Presented at: Twenty-Fourth Annual Telecommunications Policy Research Conference, Solomons, Maryland, USA, October 5-7, 1996.

Forthcoming: In S. Wildman, B. Cherry, & A. Hammond (Eds.), Universal service in context: A multidisciplinary perspective. Mahwah, NJ: Lawrence Erlbaum.

In this paper,we argue that universal service for advanced telecommunications will not emerge out of a grand design or new definition, but instead out of an intensely contested incremental process. New services will be added layer-by-layer to the universal service package, which will not be entirely consistent across the nation. We conceptualize that this process involves three modes of network expansion—territorial, demographic and layered.

Territorial expansion is the extension or replication of systems across geographical space over a period of time. Demographic expansion takes place within populations and expands the definition of "in-groups" participating in a system to include progressively larger sections of the population over a period of time. Layered expansion is the asynchronous addition of a hierarchy of complementary innovations to existing systems by which they are continuously modified and made more complex.

Three metaphors—the telegraph, universal suffrage, and universal education—embody the logic of network expansion in the territorial, demographic and layered modes. As the telephone network evolved, it passed through these three modes of expansion. Naturally, the metaphoric terms in which we thought about the network also changed to reflect the dominant processes in the different stages of its expansion. The telegraph metaphor, which was adequate to understand territorial expansion in the early phase of the telephone network, was replaced by the universal service principle, derived in part from the universal suffrage metaphor, when demographic expansion became the dominant mode of network expansion. But the universal suffrage metaphor,which has a definite closure—one person, one vote—is unable to tackle the open endedness of a multiple services environment.

As layered expansion becomes the dominant process in the evolution of the telephone network, we suggest that the universal education metaphor is more appropriate. The question of universal education never had a closure, as it dealt with a continually evolving phenomenon. The definition of universal education gradually expanded from elementary school to middle school to high school and even touched on higher education. How were universal education policies developed for a "service" that had no well defined boundaries? We draw on this experience with universal education to formulate broad principles for our current efforts to develop universal service policies in advanced telecommunications.

Metaphors are the building blocks of our psychological reality. From the tenuous foothold of our limited knowledge, we shape the confusing and encircling morass of everyday actuality into the neat metaphoric bricks of understanding and build them up into the edifices of a world-view. We use metaphors to bridge the chasm between the frontiers of knowledge and the unknowns of new phenomena.

Nowhere is this more apparent than in our efforts to make sense of the dawning information age. As we take tentative steps towards formulating policies for new information and communication technologies, we are confronted with critical decisions. With the unprecedented rate of technological and social change, and with little data to fall back on, our efforts are little more than the gropings of individuals in an unfamiliar dark room. All we have are metaphors, which mold our thoughts, guide our instincts, and ultimately shape our very actions.

The importance of metaphors in shaping our policies for advanced telecommunications makes it necessary for us to reexamine the appropriateness of the metaphors we use. According to Fiumara (1996), "metaphoricity is a basic mode of functioning whereby we project patterns from one domain of experience in order to structure another domain of a different kind" (p. 98). Metaphoric thought involves the discovery of recurring patterns of relationships among elements in two different regions of experience, denoted as the principal subject and the subsidiary subject (Brown, 1976; Gentner & Jeziorski, 1993; Nuessel, 1988). "A metaphor selects, emphasizes, suppresses, and organizes features of [its] principal subject by implying statements about it that normally apply to the subsidiary subject" (Nuessel, 1988, p. 14). Thus, the metaphor allows us to make sense of a new and unexplored phenomena by selecting and organizing its observed features into patterns familiar to us from our experience with "similar" phenomena (Black, 1962). The specific linguistic term employed in metaphor is only short-hand for the recurring pattern of relationships observed between the two subject areas. Therefore, metaphors are best evaluated not on the basis of the explicit terminology used in the metaphor, but at the deeper level of the implicit "experiential gestalts" invoked by the metaphor (Johnson, 1981).

In the case of universal service, our thinking has not kept pace with technological change. We continue to invoke an "experiential gestalt" drawn from a bygone regulated monopoly era while technological change has altered the rules of the game. Our efforts to develop appropriate universal service policies have attained limited success because we have sought to make modifications within the existing paradigm. We need a new starting point for our thinking. One of the best ways to do this is to adopt a new metaphor which provides a more relevant framework for our thoughts.

In this paper we explore other domains of experience— telegraph, universal suffrage, and universal education— to identify metaphors which would provide the appropriate "experiential gestalt" for the challenge we currently face with universal service. We first conceptualize the three modes of system expansion—territorial, demographic and layered—which push the system towards ubiquity and increasing complexity. We then use examples from telegraph, universal suffrage, and universal education to illustrate these three modes. Thereafter we locate the specific case of universal service in advanced telecommunications within the larger framework and draw on the relevant "experiential gestalts" to suggest new policy directions.

This section discusses three modes of system expansion—, demographic, and layered. These modes are discussed in conceptual terms without reference to specific examples which are provided in the next section.

Territorial expansion. It is the extension or replication of systems across geographical space over a period of time. Extension implies connectivity, which in turn implicates proximity or contiguity as important factors in territorial expansion. Extension is generally outward from a center to the periphery. However, territorial expansion can also take place by replication in geographically discontinuous areas. (Fig. 1) Proximity is thus not defined exclusively in geographical terms. Multiple centers may exist in territorial expansion. Under such circumstances, territorial expansion also passes through a stage of system integration. Replication may be partial or total, depending on the degree to which the new system is imitative of those already in place.

Space in territorial expansion is directional because a place is defined not by its absolute location but by its relative position in the systemized space, which in turn influences the source of innovations it adopts. For example, the former West and East Germanies, while geographically contiguous, "looked" west and east respectively and belonged to different policy spaces. Also, space is "hierarchized" as center-periphery or as node-trunk-hub-spoke-terminal in networked systems. These positions are not fixed; a terminal position in a networked system may be transformed into an important hub if it becomes the point of interconnection to a neighboring networked system.

Demographic expansion. It takes place within the population at each location via a process involving "in-groups" and "out-groups." [1] The process expands the definition of "in-groups" participating in a system to include progressively larger sections of the population over a period of time (Fig. 2). Early participants or adopters form the core group, to which new groups are progressively added, either by expanding the scope of the definition, or by including new categories in the definition.

Demographic expansion can be either intentional or incidental. In intentional demographic expansion, membership in the "in-group" is based upon formal criteria, and the definition is expanded by an authority empowered to do so. In incidental demographic expansion, there are no formal requirements for membership, and the "in-group" is expanded gradually through participation. Incidental demographic expansion is the closest to the conventional description of diffusion in the literature.

Layered expansion. It is the asynchronous addition of a hierarchy of complementary innovations to existing systems by which they are continually modified and made more complex. Innovations are "asynchronous" because their time of initiation is different, and they have different penetration rates. No two adoption units are confronted with adoption decisions for the different innovations in the same order. Innovations are "hierarchized" because they can be arranged on a scale of increasing complexity and the adoption of some innovations has to precede that of others (Fig. 3). Innovations are complementary because they belong to the same policy area, influence one another, and their progressive adoption can be understood in terms of movement towards a general policy goal or social objective. In that sense, innovations in layered expansion are modular. Here, the adoption decision is more complex than in territorial or demographic expansion as adopters have to make decisions at a set of points, not just about each innovation, but about the combination of innovations to be included in the "policy bundle."

The hierarchy of innovations in layered expansion is divided between the essential and the optional. Innovations in the lower rungs of the hierarchy are designated as essential, while other innovations in the higher levels are optional. The process of layered expansion involves the progressive upward movement of the level of innovations considered essential. Once a lower level is standardized and integrated into the system, there is competition between the alternatives in the next higher level for inclusion. The system expands by choosing one competing alternative and eliminating or restricting the others.

It is important to note that the three modes of system expansion do not necessarily take place in any one particular order. Under some circumstances, it is possible for a system to pass through more than one of the expansion processes described above, either simultaneously, or in varying sequential combinations.

In this section, we provide three examples that illustrate the modes of system expansion discussed above. They are the growth of the telegraph network, universal suffrage, and universal education.

Telegraph. The Morse telegraph network started commercial operations in January 1846 on a line from New York to Philadelphia, and gradually expanded its coverage. It evolved into a confederation of local telegraph companies which were interconnected with one another (Brock, 1981). The typical arrangement was that the local franchisee gave Morse 50% of the stock for the right to use the patented technology. The network essentially grew in a "hierarchized" manner as it moved down the urban hierarchy from big cities to smaller and smaller towns till it reached the outer fringes which were too sparsely populated to make the investment profitable. Gradually, a fairly extensive network developed in the populated areas of the eastern seaboard. Then a quirk of fate unleashed forces which propelled a dramatic expansion of the network in a brief five year period from 1847 to 1852 (Thompson, 1947/1972).

It all began with a rather innocuous franchise agreement between Morse interests and Henry O'Reilly for the construction and operation of a telegraph network covering the territory west of Philadelphia lying between the Ohio river, the Great Lakes, and the Mississippi. Since the Morse interests did not expect O'Reilly's line to generate much revenue, they agreed to levy a reduced patent fee—only 25% of stock. O'Reilly's line turned out to be a major success. The Morse interests regretted the fact that they had relinquished control over a vast territory for half the standard patent fee and found a loop-hole via a technicality to nullify the contract. O'Reilly, once a partner, became a fierce competitor and the ensuing rivalry set into motion the "mad era of methodless enthusiasm" (Thompson, 1947/1972, p. 97). The end result was a reckless race to occupy virgin territory. Network investment decisions were not made on the basis of potential profitability but the burning desire to be the first to establish a telegraph network in new territories. The network exploded and soon it was covering the continental expanse of the U.S.

Telegraph's growth was essentially one of territorial expansion. It did not involve either demographic or layered expansion as the telegraph never became "individualized" like suffrage, nor household-based like the telephone. Here, the primary mode of network expansion was extension rather than replication.

Universal suffrage. The territorial expansion of universal suffrage was particularly evident in the U.S. because of its federal structure, with the states taking the important decisions in enfranchising their residents. As the population expanded westward, the states which were created in newly settled territories often replicated the democratic institutions and practices of the more established eastern states. This replication was typically partial as these new states modified the institutional forms and practices they borrowed to fit their peculiar environmental factors, thus serving as crucibles for the emergence of new variations which in turn influenced the eastern states. There was considerable transfer of ideas along both directions as the U.S. moved towards universal suffrage.

The expansion of universal suffrage also had a demographic dimension. [2] Universal suffrage was neither granted to Americans by benevolent founding fathers, nor was it bestowed in any single act of great magnanimity. At first, the voting rights were restricted to the propertied elite, the "in-group." Adult franchise was extended to the rest of the population, the "out-group," in a very piecemeal manner as a result of a protracted struggle between competing political interests within the "in-group."

A democracy in many ways is a game of numbers in which the side with the largest number of votes emerges as the winner. In a situation of limited democracy where the suffrage is restricted, the numbers game can be influenced by expanding the population base of eligible voters. This possibility of manipulation has frequently led to fraudulent practices. When voting privileges were restricted to freeholders, "the practice of creating freeholds at the time of a crucial election was fairly widespread in the American colonies. When an office seeker and his friends thought an additional number of votes was necessary to win an election, it was not unusual for them to create small freeholds for the express purpose of manufacturing votes" (Williamson, 1960, p. 50). The other strategy was to ease the restrictions on suffrage so that a more partisanly disposed population could be included in the voter pool. In Pennsylvania, the Quaker party used the naturalization of aliens as the means of enlarging its voter base (Williamson, 1960). The Midwestern states extended suffrage to foreigners for economic reasons to "attract immigrants from other states to her unplowed fields" (Porter, 1918/1969, p. 18). The unrolling of universal suffrage was a series of compromises and political compacts within the "in-group" which had very little to do with the espoused ideals of freedom and the equality of "man." These negotiated interim compromises offered temporary advantages to different players and hence provided motivations for extending suffrage. Each step forward opened new doors and thereby had a snowballing effect. This political maneuvering among the sub-groups within the ever increasing "in-group" eventually led to suffrage being extended to each and every adult citizen of the U.S. The entire experience has very aptly been described as the "strange phenomenon of suffrage being carried forward on a tide of fallacies and specious doctrine" (Porter, 1918/1969, p. 65).

The movement towards universal suffrage stopped at demographic expansion because it had a definite closure— person, one vote. Universal suffrage by its very definition could not accommodate qualitative differences in the voting privileges it granted citizens. Therefore, layered expansion was not an issue in the case of universal suffrage.

Universal education. The system of universal education as we know it today is a bundle of different policies including compulsory attendance laws, sources for school finance, provision of schools and colleges in territorial jurisdictions, mandatory standards for teachers and syllabi, and enforcement mechanisms for all of the above. The evolution of this complex system involved all three modes of expansion—territorial, demographic, and layered—which often occurred simultaneously in an intertwined manner. This non-sequential system development is evident in the following discussion which traces the growth of universal education since early 17th century.

The first immigrant groups settled the continent's eastern seaboard in relatively isolated and independent colonies with distinct national and religious identities. The educational initiatives of the colonies were also more or less independent of each other. After independence, federalism ensured that each state within the union enjoyed considerable freedom to choose the educational system most suited to itself. However, their independent identities did not prevent colonies (and later states) from borrowing ideas from each other. Successful innovations were copied from other territorial jurisdictions, especially those regarded as exemplars. Exemplars were usually the states considered more progressive, as well as the ones with which the adopting state had the most in common. Thus, exemplars were often states from the same region as the adopting state. For instance, there were major differences between the educational initiatives of New England and the southern colonies. In New England, concern with the religious instruction of the young and the republican ideal of the equality of all citizens prompted community elders to create the early statutes making education compulsory for all children. The colonies of Massachusetts and New Haven were the exemplars for colonial New England (Kotin & Aikman, 1980). Their laws were copied by all the New England colonies except Rhode Island between 1642 and 1671, making it a mandatory requirement for all children to be taught by their parents, or by the masters in the case of apprentices.

Educational requirements were gradually expanded from a prescribed minimum to incorporate new and higher level requirements. For example, the first Massachusetts law of 1642 requiring all children to be taught reading was amended in 1648 to say that pupils should be able to read "perfectly" (Kotin & Aikman, 1980). In 1660, the New Haven colony added a writing requirement.[3] As each colony adopted educational statutes, it had the option to devise new requirements, or to borrow from more than one source. The Plymouth colonyÕs education law of 1671 illustrates this. It borrowed the educational requirements described above from Massachusetts, and enforcement provisions from the New Haven law. [4] Significantly, it did not have any requirement that children should be taught a trade, which both the Massachusetts and New Haven laws had, demonstrating that adoption was not whole-sale but quite selective.

At the same time, the southern colonies had Virginia for their exemplar. Notably, no southern colony adopted a law requiring all children to be given education, as was the case in New England. According to Urban and Wagoner (1996), the Virginia model used "every man according to his own ability" (p. 23) as its motto. Since well-to-do families would naturally provide their wards with the best education they could afford, the state needed to concern itself only with the education of those incapable of acquiring an education on their own, namely "orphans, poor children, illegitimate children, and mulattos born of white mothers" (Kotin & Aikman, 1980, p. 19). In 1705, Virginia introduced a law stating that all apprentices should be taught a trade and to read and write. These Virginia laws were copied in the other southern colonies.

Two competing models were thus available for other colonies/states. Ultimately, it was the Virginia model of apprenticeship education that prevailed. Even in New England, universal compulsory education ceased to be a concern after the late-1600s, to resurface only in the 1800s. [5] In the intervening period, education statutes in America referred not to universal education, but only to the responsibilities of masters toward their apprentices.

The 18th and 19th centuries were marked by a large number of educational initiatives which flourished for some time, and then died out or evolved into new forms. As Good points out, "one may discern the faint outlines of several systems" (1956, p. 37) in the educational diversity of this period. These "systems" of education offered educational alternatives to students with different educational aspirations and economic means. For the common masses there were dame schools, parochial schools, old field schools, and district schools which prepared children to become tradesmen or clerks. [6] In the cities, there were venture schools and private schools through which pupils could acquire a good education preparing them for the intermediate professions. In these schools, practical skills like accounting and book-keeping were imparted with relatively less emphasis on scholarship. Finally, there were the preparatory schools and Latin grammar schools which prepared future scholars, doctors, clergymen, and lawyers for college. Most of the children in these schools belonged to privileged families.

It should be noted here that the emergence of educational institutions did not follow any sort of graded structure. For example, the first colleges were instituted almost contemporaneously with the first elementary schools. [7] What we witness as systems evolve is the integration of existing institutions into a graded structure, so that they form an educational track or hierarchy. Existing forms are sometimes modified and combined to fill an intermediate position in the evolving hierarchy. An example of this is the combination of the college preparatory grammar school and the terminal venture school into the academy, a secondary level institution. These hybrid institutions combined the desirable features of both predecessor institutions. For example, the academy offered a variety of programs including both college preparatory and terminal, and was private like the venture school but had the grammar school's rigorous academic standards (Hillesheim & Merrill, 1971).

Another trend that marked the 18th and the first half of the 19th century was the increasing role of the state in education. This was not without opposition from advocates of parental rights or the rights of local communities. But by the early decades of the 20th century, the rights of the state to set reasonable educational standards had acquired considerable public and judicial sanction. [8] The standard-setting role of the state paved the way for the evolution of state-wide and nation- wide systems of education. In the absence of this standard-setting role, nascent educational systems would probably have undergone continuous mutation without convergence.

In the early 19th century, a constellation of factors resurrected the old Puritan ideal of education for all in an era marked by industrialization, rural-to-urban migration, and massive immigration. Many believed that the consequent social dislocation will lead to class conflict and discord. Educational reformers like Horace Mann advocated that the school system should be used as a means of assimilating diverse ethnic and linguistic groups into a common American identity. As the country rapidly industrialized, there was increasing demand for better educated, more skilled workers. There was also a need to inculcate in prospective factory workers qualities like punctuality, team-work, and obedience to authority. The school system was an effective means of achieving these purposes.

Also important was the sociological transformation of the American work-place. In the colonial period and for many decades after that, able-bodied workers were difficult to find, and all available hands had to be employed in field or workshop. Within this context, child labor was condoned or even encouraged in the name of the Puritan work ethic. Continuing immigration increased the labor supply and thereby reduced the need for child labor. The labor interests realized that the continuing employment of children in the work-place was depressing wages for all. It was therefore necessary to remove a part of the child work-force from the labor pool. Chief Justice Brenner of the Supreme Court was alluding to this when he referred to the "economic function (of the educational system in keeping children of certain ages off the labor market and in school." [9]

Under these influences, the common school—free, tax-supported, non-denominational, and universal—began to generate more and more popular enthusiasm. Universal education gained acceptance as both a desirable ideal and a realizable goal. To effect the transition to a universal and compulsory system of public education, reforms were required on a number of fronts: compulsory attendance laws to compel the presence of all school-age children in educational establishments, child labor laws to prohibit them from the work-place, and school finance. Initiatives in all three areas were taken by different state jurisdictions and the innovations diffused rapidly over the union once their utility was established. Rhode Island was the first to initiate a Child Labor law in 1840 (Alexander & Jordan, 1973). By 1853, six states had child labor laws, and by the end of the century this had risen to twenty-eight (Kotin & Aikman, 1980). Compulsory attendance laws were first introduced in Massachusetts in 1852, requiring all children aged 8 to 12 to be sent to school for a minimum of twelve weeks per year, with at least six weeks running consecutively (Kotin & Aikman 1980). The District of Columbia followed suit in 1864 and Vermont in 1867. By 1900, 38 states had compulsory education statutes, with the southern states enacting their own laws by 1918.

Innovations in school finance were much more controversial because they directly affected the tax liabilities of all citizens. Earlier, schools had been financed by private endowments, land-grants, parental contributions or permanent school funds, created mostly from the sale of land, the income from which was earmarked for schools (Edwards & Richey, 1963). Once the school systems expanded, an institutional arrangement that could manage funds, oversee their disbursement, and regulate their utilization became necessary. In 1812, the state of New York established an office of the superintendent of common schools, adopted a district system, and permitted a district tax for school finance. This system was rapidly emulated by the other states. By 1860, twenty-eight of the thirty-four states had chief state school officers, and the district system had become firmly entrenched (Edwards & Richey, 1963). The creation of an educational bureaucracy was in itself an impelling force towards the institutionalization of universal eduction (Katz, 1976). Thus, during the first few decades of the 20th century:

compulsory schooling was transformed from a relatively simple statute requiring a fixed period of school attendance into a complex network of interrelated rules. This network of rules involved not only requiring school attendance but also the hiring of truant officers, delegating jurisdictive power, and dealing with a host of child labor regulations (Katz, 1976, p. 21).

Soon after the war, President Truman established the Commission on Higher Education to study the civilian implications of this war experience. The Commission recommended that tuition-free education should be made available to everybody for the first two years of study either in traditional colleges or community colleges (President's Commission on Higher Education, 1947). It felt that "only in this way can we be certain of developing for the common good all the potential leadership our society produces, no matter in what social or economic stratum it appears" (President's Commission on Higher Education, 1947, p. 38). The postwar success of the G.I. Bill further confirmed the belief that economic barriers were preventing a large number of otherwise capable persons from fully developing their potential (Bowles, 1966; Munday & Rever, 1971; Willingham, 1970). Further, studies suggested that the G.I. Bill eventually cost the tax payers nothing because the increased taxes paid by the beneficiaries more than compensated for the original investment (Educational Policies Commission, 1964).

The movement towards universal higher education was given a further boost by a critical view of higher education that started striking roots in the 1960s. It saw the institutions of higher education as reproducing and perpetuating inequality (Arboleda, 1981; Birenbaum, 1971; Lavin, Alba & Silberstein, 1981; Rossman, Astin, Astin, & el-Khawas, 1975). They were viewed as sorting and screening mechanisms which distributed power and privilege by creating and credentialing an elite (Schrag, 1971; U.S. Office of Education, 1971; Willingham, 1970). According to this view, the purpose of higher education should not be to pick winners but to maximize the educational potential of each and every student at whatever level he or she enters the higher education system (Karabel, 1972, Rossman et al., 1975).

The war experience, reports of the President's Commission on Higher Education and other bodies, the advent of the critical view of higher education, and other elements of the sociocultural milieu of the 1960s, including the Civil Rights movement, converged to create pressure for universal higher education. The biggest impediment for any movement forward was a near total lack of conceptual understanding of what exactly universal higher education entailed. It was yet another ideal which was conceptually powerful but difficult to operationalize. The past experience with secondary education offered few precedents because higher education could not be prescribed on an uniform basis to everyone. Should higher education be made available to everyone or should it be restricted to those who have the required talent? Should it have an uniform curriculum or should it offer great diversity? These questions were not a problem earlier because in the past progress was linear and incremental as each additional grade was an additional layer on a curriculum which was already in place. Universal higher education, on the other hand, represented a major discontinuity. The questions it raised were profound and the answers elusive.

Yet, the lack of definitive answers to these fundamental questions did not stop the evolutionary process. The process continued forward as exemplars emerged and defined what was possible in practical terms. The course of universal higher education was greatly influenced by two very different models represented by the California Higher Education System and the City University of New York. California adopted a stratified strategy of "differential access" with three levels of colleges—University of California system, California State University system, and a community college system. The upper eighth of high school graduates were placed in the University of California campuses, the upper third in the California State University campuses, and the community colleges were open to everybody with a high school diploma. All students were assured admission to the system but not to a specific college. The system, however, allowed for upward mobility as a student could work his/her way up the hierarchy (Jaffe ∧ Adams, 1972). The City University of New York adopted a very different "open admissions" model which assured admission to all the high school graduates from New York city who ranked in the top half of their graduating class. The resulting increase in student body was accommodated by hiring new faculty and increasing class sizes. Furthermore, a concerted effort was made to offer remedial programs to take care of the needs of the weaker students (Lavin et al., 1981). The other states developed their own arrangements which were hybrids of elements drawn from these two models. The fascinating thing about this whole experience is that all these major policy initiatives were taken within a context of great ambiguity.

The development of universal education followed a more complex evolutionary path than that of either the telegraph or universal suffrage. It required a number of separate but complementary initiatives to be put together over a period of time. Initiatives arose in different jurisdictions at different points in time, and diffused over the federal system at different rates. Educational requirements began with elementary skills and were gradually upgraded to include higher levels. There was competition between alternative forms as new layers were added to the educational system. This ÒlayeredÓ process makes the development of the universal education system a good illustrative example of layered expansion.

The influence of metaphors is very evident when they are explicitly used to frame an issue. Their influence is more subtle and often more powerful when they are so deeply embedded in a discourse that even the participants are not aware of it. For example, Lakoff and Johnson (1980) point out how common discourse on "argument," laden with words like "attack," "defence," and "demolish," is structured by the conceptual metaphor "argument is war." [10] We comment, analyze, and actually engage in an "argument" within the conceptual structure of this metaphor without even being aware of it.

In this section, we perform a similar analysis of the implicit conceptual metaphors which informed the development of the telephone network in the different stages of its growth and thereafter suggest a new metaphor for the future. We do not claim that these metaphors were explicitly used in the development of universal service policies. However, we suggest that like the "argument is war" metaphor in the above example these metaphors formed the conceptual structure within which different initiatives were undertaken.

As the telephone network evolved, it went through territorial, demographic, and layered modes of expansion. Naturally, the metaphoric terms in which we thought about the network also changed to reflect the dominant processes in the different stages of its expansion. The telegraph metaphor, which was adequate to understand territorial expansion in the early phase of the telephone network, was replaced by a derivative of the universal suffrage metaphor, the universal service principle, when demographic expansion became the dominant mode of network expansion. As layered expansion becomes the dominant process in the evolution of the telephone network, the metaphors we employ will have to be changed again.

Figure 4 illustrates this migration of metaphors and shows that the universal education metaphor encompasses the layered expansion processes currently active in network expansion.

The first stage of telephone's development was guided by the telegraph metaphor. This was quite natural because the telephone was an unexpected technological mutation which emerged out of efforts to improve telegraph. Even the patent filed by Bell for his new invention was titled "Improvements in Telegraphy." It did not even mention the word "telephone" (Brooks, 1976). The influence of the telegraph metaphor pervaded all aspects of the telephone business. "Even the language of the telephone revealed its ancestry; telephone calls were for many years labeled as messages and measured in message units" (Fischer, 1992, p. 81). In fact many of the key inventions like telephone exchanges which facilitated the development of early telephony grew out of ideas first developed for telegraph (Fagen, 1975; Garnet, 1985; Mueller 1989).

While the telegraph network expanded mainly by extension, the telephone network grew by both extension and replication. The growth of the Bell System represented the former and the independents the latter. In the beginning, the Bell companies focused almost entirely on the urban markets. As in the case of the Morse telegraph network, the Bell System grew in a "hierarchized" manner down the urban hierarchy. It refused to serve the rural areas in spite of vociferous demands from the rural population. As soon as the Bell patents expired, independent telephone companies mushroomed all across the rural landscape. They were often crude homegrown telephone networks put together by farmers which "replicated" the Bell networks. The surprising success of the independents forced the Bell companies to enter the rural areas. The introduction of competition radically altered the pace of network development (Mueller, 1993). There was almost a nine-fold increase in the per capita density of telephones in the period 1893-1902 (Fischer, 1987). In this phase of the networkÕs development, the foundations were laid for universal service'a geographically ubiquitous telephone network which interconnected all communities.

The next step in the development of universal service was the extension of telephone service to all citizens. The mode of network development at this stage was very different from the previous one as the extension of the network to "everyone" followed a very different logic from the earlier extension to "everywhere." [11] It raised questions which the telegraph metaphor could no longer address. Should society subsidize the local loop for those who could not pay the full cost? The metaphor that helped address this question was that of universal suffrage. [12]

Initially, the telephone service was limited to the rich and the business community, the "in-group." It was almost a luxury. Slowly, as prices fell, the subscribership continued to expand beyond the small group of wealthy elites to those on the lower rungs of the economic ladder as more and more people could afford to pay for the telephone service. The "in-group" was positively predisposed towards universal service because the expansion of the network increased the economic value of their telephone service as they could communicate with a larger universe of subscribers.[13] The only hitch was that this expansion necessitated a subsidy from the "in-group" to pay for the local loops for the "out-group." The amount of subsidy determined the price of service for the "out-group" and thereby defined the size of the "in-group." Within this context, the universal suffrage metaphor played an important role by providing justification for public policy intervention as it suggested that access to telephone, a critically important technology for participation in the life of the community, was almost a right of every citizen. An elaborate system of subsidies set up by the regulators eventually expanded the telephone service to "everybody."

The extension of telephone service to almost the entire population settled the question of universal telephone service until the recent convergence of telephone and computer technologies transformed the telephone network from a vehicle for the transmission of voice telephony signals to a platform for the delivery of a whole host of information services such as 911 emergency service, home banking, distance learning, remote medical diagnostics, surveillance, energy management, special services for the hearing impaired, automatic language translation, voice mail, computer conferencing, access to data bases, and others. This proliferation of telephone-based services has changed the issue of universal telephone service from "one household, one telephone" to "one household, how many services?" There is general agreement that the definition of universal service needs to be extended beyond basic voice communication (Gillan 1986; Hadden, 1991a, 1991b; Hudson, 1994; Information Infrastructure Task Force, 1993; NTIA, 1988, 1991; O'Connor, 1991; Office of Technology Assessment [OTA], 1990; Parker, Hudson, Dillman, & Roscoe, 1989; Williams, 1991). However, nobody has been able to develop universally accepted criteria for including some services in the universal service package and excluding others. Within this changed context, the universal suffrage metaphor which had a definite closure—one person, one vote—is unable to tackle the open-endedness of a multiple services environment. Quite obviously, there is a need for a new metaphor.

The universal education metaphor seems more appropriate because it also has an open-endedness to it which is akin to modern telecommunications. The question of universal education never had a closure as it dealt with a continually evolving phenomenon. The definition of universal education gradually expanded from elementary school to middle school to high school and even touched on higher education. How were universal education policies developed for a "service" that had no well defined boundaries? This experience with universal education perhaps has much to offer for our current efforts to develop universal service policies for advanced telecommunications services.

In this concluding section, we draw on the "experiential gestalt" of our previous efforts in universal education to sketch out implications for the development of universal service in advanced telecommunications. Based on our analysis in this paper, we have distilled the following three broad principles which have a bearing on our current efforts:

1. An a priori definition of universal service is not a necessary condition for the development of universal service in advanced telecommunications. Our fixation on defining universal service has led us to focus our energies almost entirely on developing a new definition. The assumption here is that once a new definition is developed, the rest of the decision-making process will be relatively simple. This logic has such a strong grip on our imagination that repeated failures have not deterred us from our efforts to develop a new definition for universal service. The history of universal education beckons us to rethink our assumptions.

The development of universal education continued even in the absence of an adequate definition for a basic concept like the "public school." While it may seem trivial in retrospect, a considerable amount of energy was expended in the search for such a definition in which legislatures, religious groups, and even the courts were involved. The debate was especially contentious because designation as a public school made an institution eligible for public finance, but also subjected it to state supervision over instructional content (Good, 1956). However, progress in universal education did not await the resolution of these conflicts. It progressed in spite of, and possibly because of, this definitional ambiguity.

Initiatives in universal higher education were undertaken in spite of no clear answers for many fundamental questions: What exactly is higher education? Should its definition be restrictive and limited to four year colleges, or could it be expansive and include community colleges, technical institutes, and all post-high school vocational programs? What are the benefits and costs of higher education? There was at best an expectation that universal higher education would provide benefits in the form of enhanced productivity and an enriched democratic process. It was, however, difficult to quantify these benefits (Bowles, 1966, Hansen & Witmer, 1972). Even an understanding of the costs involved was hazy at best as estimates of the percentage of population who would benefit from higher education ranged from 25% to 49%, creating uncertainty about the required quanta of investments (Miller, 1971, p. 97). Universal higher education grew in an environment within which higher education was "unable to define with precision its purposes, to measure with clarity its processes, or to quantify with certainty its outputs" (Olivas, 1979, p. 1). It was within this overall context of great ambiguity that the final and perhaps the most important question—who should have access to higher education?—was tackled.

The parallels with universal telecommunications service are striking. We are now faced with a similar set of questions: What should be the definition of universal service? What will be its benefits and costs? Who should have access to advanced telecommunications services? The experience with universal education suggests that our current efforts to define universal service is a misdirected effort. It will perhaps never be possible to develop a national consensus on a new definition. Furthermore, the lack of definitive answers to these questions is not necessarily a roadblock for the development of universal service.

2. Exemplars in the form of implemented programs have a much greater influence on the nature and scope of universal service than formal definitions. The concept of universal service is inherently an ambiguous idea. It springs from notions of equality and participatory democracy. The idea resonates with high ideals but has a vagueness which makes its translation into implementable programs problematic. However, precisely because of this dilemma, tentative real-world solutions have a much greater impact on the further development of the idea than theoretical formulations. Even if formally imprecise, these "solutions" have a metaphoric impact which makes the practical heroic and thereby bridges the gap between the mundane and the ideal. These "solutions" become the exemplars against which all subsequent efforts are measured. Like Massachusetts and Virginia for early education, and the California Higher Education System and the City University of New York for higher education, exemplars become the surrogate for the ideal which is difficult to conceptualize. The current debate about universal service is still in the pre-exemplar stage. The policies that encourage decentralization and allow the different states to develop their own solutions will facilitate the emergence of exemplars. They will draw attention away from the search for definitions, to an examination of the pros and cons of different exemplars, and the development of hybrids suitable for each state.

3. The universal service policy is in reality a "bundle of policies" whose elements arise from different sources in an asynchronous manner and are later harmonized into an integrated policy framework with local variations in each jurisdiction. We tend to view universal service as a monolithic concept that can be realized by a masterful grand plan. The development of universal education reveals that in reality the process is far more diffused and disorganized. The policy framework which sustains universal education is highly modular: it has in fact been described as a "complex network of interrelated rules" (Katz, 1976). These modules develop separately and are later organized into integrated systems. There is considerable variance in the make-up and composition of each module, and they in turn can be arranged in different combinations to create a wide variety of systems.

This diversity is reflected in the great variation in the education systems of different states all of whom offer universal education but in their own peculiar ways. This experience with universal education suggests that a similar modularized approach would be more appropriate for universal service in advanced telecommunications. Instead of working on an overarching framework, we should create a decentralized policy environment which would facilitate the development of a wide variety of policy modules— criteria, service packages, finance, and others— could later be organized into integrated systems.

1. The terms "in-groups" and "out-groups" have a long history in identity politics, cultural studies, and other related areas. Not all the connotations of these terms are invoked by their usage here. We have used these "loaded" terms partly for want of better terminology, and partly to highlight the differences in the interests and perspectives of those who have access to a service or resource and those who do not. Back to text.

2. This discussion on the demographic expansion of universal suffrage has been extracted with minor modifications from Sawhney (1994, pp. 386-387). Back to text.

3. As Urban and Wagoner (1996) point out, the addition of a compulsory writing requirement was radical for the times. Compared to reading, for which primers and other books were cheap and readily available, writing material was expensive and hard to obtain. It was regarded as a specialized skill enabling its practitioners to enter lucrative professions in law, business, and public affairs. Writing was also taught by largely male teachers, while reading was the domain of female instructors. Back to text.

4. The enforcement provisions in the New Haven law may be regarded as the first recognition of the rights of the state in the education of the young. The law allowed colonial, rather than local authorities to supervise enforcement of the law, and the negligence of parents or masters to provide a suitable education to their wards was punishable by fines (Kotin & Aikman, 1980). Back to text.

5. See Kotin and Aikman (1980) and Hillesheim and Merrill (1971), for some of the reasons why the New England model was gradually eclipsed. First, continued immigration of an ethnically and religiously diverse population diluted the homogeneity of early settlements and made consensus on educational content difficult. Secondly, with the increasing secularization of education, the church which had taken a leadership role in colonial education gradually abdicated control to local authorities. Thirdly, as populations moved westward, settlements became too dispersed to support a school system and made compulsory education statutes impossible to implement. Back to text.

6. As the name indicates, dame schools were run by women in their homes where very young children were taught "their letters." Old field schools, parochial schools, and district schools were all elementary schools under different forms of management. Old field schools, run on the basis of a contract between teacher and parents, got their name from being located on waste or exhausted land belonging to the town. Parochial schools and district schools were run by the church and an elected board respectively (Good, 1956, see pp. 37-38). Back to text.

7. Nine colleges were founded in America during the colonial period, the first, Harvard, as early as 1636 (Hillesheim & Merrill, 1971). Back to text.

8. As the Supreme Court in the Pierce v. Society of Sisters case ruled, "No question is raised concerning the power of the state reasonably to regulate all schools, to inspect, supervise and examine them, their teachers and pupils; to require that all children of proper age attend some school, that teachers shall be of good moral character and patriotic disposition, that certain studies plainly essential to good citizenship must be taught, and that nothing to be taught which is manifestly inimical to the public welfare" (quoted in Katz, 1976, p. 26). It should however be noted that the same judgment negated the rights of states to compel the attendance of all children in public schools, provided their parents imparted some form of education to them. The Pierce v. Society of Sisters case examined the constitutionality of a 1922 Oregon statute that compelled public school education for all children 8 to 12 years old. Back to text.

9. Wisconsin v. Yoder, 406 U.S. 205 (1972). Back to text.

10. The metaphor "argument is war" can be seen operating in our everyday language as seen in the following statements:

Your claims are indefensible.
He attacked every weak point in my argument.
His criticism were right on target.
I demolished his argument.
I've never won an argument with him.
You disagree? Okay, shoot!
If you use that strategy, he'll wipe you out.
He shot down all of my arguments

(Lakoff & Johnson, 1980, p. 4). Back to text.

11. The extension of the telephone network to "everyone" required considerable investment in the local loop which connected individual subscribers to the telephone network. This created many problems because the local loop is a dedicated and not a shared facility and hence there is a distinct cost for connecting each individual subscriber to the network. This issue did not arise in the case of telegraph because the network was not extended to each individual subscriber. Back to text.

12. Though we did not come across any historical evidence explicitly linking universal suffrage and universal telephone service, it is quite apparent that the principle of universalism was at the root of both. For example, Assemblywoman Gwen Moore during her universal service campaign in California made this connection when she said "If the freedom to communicate is a fundamental right then access to the means of communication must also be a fundamental right. Without access, one cannot be a part of the telecommunicating community" (Jacobson, 1989, p. 59). Thus we see that the universalist ideal created the conceptual structures within which all the "universal" initiatives were taken. Back to text.

13. According to Littlechild (1979), network externalities have been recognized as a factor in telephony right from the 1890s. The generation of a consumer surplus due to larger network size is fairly obvious, though there is presently little agreement on the pricing implications of this positive externality. See Taylor (1994) and Mitchell and Vogelsang (1991) for rigorous economic analyses of network externality. Back to text.

Alexander, K., & Jordan, K. F. (1973). Legal aspects of educational choice: Compulsory attendance and student assignment. Topeka, KS: National Organization of Legal Problems in Education.

Arboleda, J. (1981). Open admissions to higher education and the life chances of mlower class students: A case study from Columbia. Unpublished Ph.D. dissertation, Indiana University.

Birenbaum, W. M. (1971). Something for everybody is not enough. In W. T. Furniss (Ed.), Higher education for everybody? (pp. 65-82). Washington, DC: American Council on Education.

Black, M. (1962). Models and metaphors: Studies in language and philosophy. Ithaca, NY: Cornell University Press.

Bowles, F. (1966). Observations and comments. In E. J. McGrath (Ed.), Universal higher education (pp. 235-245). New York: McGraw-Hill.

Brock, G. (1981). The telecommunications industry: The dynamics of market structure. Cambridge, MA: Harvard University Press.

Brown, R. H. (1976). Social theory as metaphor: On the logic of discovery for the sciences of conduct. Theory and Society, 3, 169-197.

Brooks, J. (1976). Telephone: The first hundred years. New York: Harper & Row.

Educational Policies Commission. (1964). Universal opportunity for education beyond the high school. Washington, DC: National Education Association of the United States.

Edwards, N., & Richey, H., G. (1963). The school in the American social order. Boston: Houghton Mifflin.

Fagen, M. D. (Ed.). (1975). History of engineering and science in the Bell system: Vol. 1. The early years, 1876-1925. Warren, NJ: Bell Telephone Laboratories.

Fischer, C. (1987). The revolution in rural telephony: 1900-1920. Journal of Social History, 21, 5-26.

Fischer, C. (1992). America calling: A social history of the telephone to 1940. Berkeley, CA: University of California Press.

Fiumara, G. C. (1995). The metaphoric process: Connections between language and life. Routledge: New York.

Garnet, R. (1985). The telephone enterprise. Baltimore: Johns Hopkins University Press.

Gentner, D. & M. Jeziorski. (1993). The shift from metaphor to analogy in western science. In A. Ortony (Ed.), Metaphor and thought (pp. 447-480). New York: Cambridge University Press.

Gillan, J. (1986). Universal telephone service and competition: The rural scene. Public Utilities Fortnightly, 117, 21-26.

Good, H. G. (1956). A history of American education. New York: Macmillan.

Hadden, S. (1991a). Regulating content as universal service (Working Paper, Policy Research Project: "Universal Service for the Twenty-First Century"). Austin, TX: The University of Texas at Austin, Lyndon B. Johnson School of Public Affairs.

Hadden, S. (1991b). Technologies of universal service. In The Institute for Information Studies (Ed.), Annual review, 1991: Universal telephone service: Ready for the 21 st Century? (pp. 53-92). Nashville, TN: The Institute for Information Studies.

Hansen, W. L. & Witmer, D. R. (1972). Economic benefits of universal higher education. In L. Wilson & O. Mills (Eds.), Universal higher education: Costs, benefits, options (pp. 19-39). Washington, DC: American Council on Education.

Hilleshiem, J. W. (1971). Theory and practice in the history of American education: A book of readings. Pacific Palisades, CA: Goodyear Publishing.

Hudson, H. (1994). Universal service in the information age. Telecommunications Policy, 18, 658-667.

Information Infrastructure Task Force. (1993). The national information infrastructure: Agenda for action. Washington, DC: National Telecommunications and Information Administration, National Information Infrastructure Office.

Jacobson, R. (1989). An open approach to information policy making: A case study of the Moore Universal Telephone Service Act. Norwood, NJ: Ablex.

Jaffe, A. J. & Adams, W. (1972). Two models of open enrollment. In L. Wilson & O. Mills (Eds.), Universal higher education: Costs, benefits, options (pp. 223-251). Washington, DC: American Council on Education.

Johnson, M. (1981). Introduction: Metaphor in the philosophical tradition. In M. Johnson (Ed.), Philosophical perspectives on metaphor (pp. 3-47). Minneapolis, MN: University of Minnesota Press.

Karabel, J. (1972). Perspectives on open admissions. In L. Wilson & O. Mills (Eds.), Universal higher education: Costs, benefits, options (pp. 265- 286.). Washington, DC: American Council on Education.

Katz, M. S. (1976). A history of compulsory education laws. Bloomington, IN: Phi Delta Kappa Educational Foundation.

Kotin, L., & Aikman, W. F. (1980). Legal foundations of compulsory school attendance. Port Washington, NY: Kennikat Press.

Lakoff, G. & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press.

Lavin, D. E., Alba, R. D., & Silberstein, R. A. (1981). Right versus privilege: The open-admissions experiment at the City University of New York. New York: Free Press.

Littlechild, S. C. (1979). Elements of telecommunications economics. New York: Peter Peregrinus.

Miller, J. L. Jr. (1971). Who needs higher education. In W. T. Furniss (Ed.), Higher education for everybody? (pp. 94-105). Washington, D.C.: American Council on Education.

Mitchell, B. M., & Vogelsang, I. (1991). Telecommunications pricing: Theory and practice. New York: Cambridge University Press.

Mueller, M. (1989). The switchboard problem: Scale, signaling, and organization in manual telephone switching, 1877-1897. Technology & Culture, 30, 534-560.

Mueller, M. (1993). Universal service in telephone history: A reconstruction. Telecommunications Policy, 17, 352-369.

Munday, L. A., & Rever, P. R. (1971). In P. R. Rever (Ed.), Open admissions and equal access (pp. 75-96). Iowa City: The American College Testing Program.

National Telecommunications and Information Administration. (1988). Telecom 2000: Charting the course for a new century (NTIA Special Publication 88-21). Washington, DC: U.S. Government Printing Office.

National Telecommunications and Information Administration. (1991). The NTIA infrastructure report: Telecommunications in the age of information (NTIA Special Publication 91-26). Washington, DC: U.S. Government Printing Office.

Nuessel, F. (1988). Metaphor and cognition: A review essay. In M. Danesi (Ed.), Metaphor, Communication, and Cognition. Toronto, Canada: Toronto Semiotic Circle. pp. 9-22.

O'Connor, B. (1991). Universal service and NREN. In The Institute for Information Studies (Ed.), Annual Review, 1991: Universal telephone service: Ready for the 21 st Century? (pp. 93-140). Nashville, TN: The Institute for Information Studies.

Office of Technology Assessment. (1990). Critical connections: Communications for the future (OTA-CIT-407). Washington, DC: U.S. Government Printing Office.

Olivas, M. (1979). The dilemma of access: Minorities in two year colleges. Washington, DC: Howard University Press.

Parker, E. B., Hudson, H. E., Dillman, D. A., & Roscoe, A. D. (1989). Rural America in the information age: Telecommunications policy for rural development. Boston: University Press of America.

Porter, K. H. (1969). A history of suffrage in the United States. New York: Greenwood. (Original work published 1918)

President's Commission on Higher Education. (1947). Higher education for American democracy, Vol. 1. Washington, DC: U.S. Government Printing Office.

Rossman, J. E, Astin, H. S., Astin, A. W., & el-Khawas, E. H. (1975). Open admissions at City University of New York: An analysis of the first year. Englewood Cliffs, NJ: Prentice-Hall.

Sawhney, H. (1994). Universal service: Prosaic motives and great ideals. Journal of Broadcasting & Electronic Media, 38, 375-395.

Schrag, P. (1971). Open admissions to what? In P. R. Rever (Ed.), Open admissions and equal access (pp. 49-53). Iowa City: The American College Testing Program.

Taylor, L. D. (1994). Telecommunications demand in theory and practice. Boston: Kluwer Academic Publishers.

Thompson, R. L. (1972) Wiring a continent : The history of the telegraph industry in the United States, 1832-1866. New York: Arno Press. (Original work published 1947).

Urban, W., & Wagoner, J. (1996). American education: A history. New York: McGraw-Hill.

U.S. Office of Education. (1971). Report on higher education (OE. 50065). Washington, DC: U.S. Government Printing Office.

Williams, F. (1991). The new telecommunications: Infrastructure for the information age. New York: Free Press.

Williamson, C. (1960). American suffrage: From property to democracy, 1760-1860. NJ: Princeton University Press.

Willingham, W. W. (1970). Free-access higher education. New York: College Entrance Examination Board.