The Cumulative Curriculum

Multi-media and the Making of a New Educational System

A Project Description

by Robbie McClintock

Substantial excerpts from a proposal submitted to IBM Research, Spring 1991

Length: 19,400 words

Executive Summary

The Institute for Learning Technologies at Teachers College and the New Laboratory for Teaching and Learning at the Dalton School propose to collaborate with IBM Watson Research Labs in a basic development program designed to achieve systemic innovation in education through the use of advanced information technologies. This group will aim to design, implement, and perfect a technology-intensive educational system. A range of key questions will guide this comprehensive effort to develop a new system for education based on the intensive use of information technology.

We intend to work to develop practical answers to such questions. We believe that a curriculum based, not on print media, but on interactive, multimedia electronic systems, will have distinctive cultural and educational properties. Most significantly, learning and study with these resources will be cumulative, not sequential. The new system will conduce to cooperative learning in contrast to the reliance on competitive learning in the current system. The new technologies will balance verbalization with more effective use of visualization in the acquis~ion of information and ideas. If effected, such a cumulative multi-media curriculum will transform practice into a new educational system.

Here are operational goals that we want to achieve through the Cumulative Multi-media Curriculum Project.

This project will be in, and of, and for, New York City. The ethos of the Ctty— its problems, resources, and institutions—will provide the settings and stimulus for the work of the project. Participants in the project at the outset will include:

We seek funding of approximately one million dollars annually, over five years, to support our effort to lnttiate a Cumulative MuHi-media Curriculum as the basis for developing a new, more effective educational system.

Educators need to pursue a demanding vision. They need to develop powerful tools, energized by a large market. Education can be changed, profoundly for the better. But it cannot be changed with little efforts, backed by small resources, promising marginal resuHs. Energized with new tools, education can become a material force bettering the human condttion. Through this project, we aim to joint with IBM and others in a concerted effort to extend the limits of education and culture.

1 A Perspective on the Task

Let's look ahead. In the twenty-second century, how might an historian of education sum up the major changes in pedagogical practice over the sweep of time? Imagine that we could commission a leading authority to report on the view backwards from that future time. This might be her executive summary.


Through most of history, education was a loose system of apprenticeship and indentured seNice in households, the main location of productive activity. Those who wanted their children to become learned employed tutors to help them out. A few schools existed within specialized institutions, such as cathedral priories and monasteries, but these were not like the schools that eventually proliferated, for students were not divided into classes or grouped according to age.

Around 1500 a.d., a major pedagogical transition began as printing with moveable type made an unprecedented era of educational development possible. But the transition was not a quick and simple change: to bring it off, innovators had to develop a complex of different, yet interrelated, educational strategies that together eventually made mass schooling for all a practical reality. Key steps in this process involved:

These developments were difficult because they had to come in parallel, as they were tightly interrelated. The transition required the integration of complex factors into a functional system: the design of educational space and time; a chosen pattern of educational motivation; pedagogical materials suitable for use in such places with such motivations; teachers adept at using such tools and strategies; and arguments demonstrating that the substantial costs of it all were worthwhile -- all were simultaneously essential to the historic transition to mass schooling.

Sixteenth-century educational reformers worked out solutions to these five, interrelated matters. For five hundred years, educators perfected, expanded, and developed the basic components of the educational system introduced early in the era of print, in due course creating modem systems of universal, compulsory schooling. The degree of elaboration and penetration of the system into society changed, and the specifics justifying the effort evolved to stay synchronized with cultural transformations. The main features remained stable, however. The design of the classroom and the organization of the school day, the motivational strategies employed, the scope and sequence of textbooks, the definition of good teaching practice, and the rationales for public support remained very stable. The reason for the underlying stability was rather simple: throughout it all, the character and limitations of printed textbooks remained substantially fixed, the keystone of the system.

We who inhabit the electronic ethos of the twenty-second century must remember that early in the twenty-first century, the function of printed materials changed rapidly, becoming restricted to their current role of verifying and guaranteeing standard data sets. Before then, physically printed materials had a much more central intellectual function. For five hundred years, books were the unmatched resources for making the ideas and knowledge of the culture available to students, and so long as this role was unquestioned, educators paid little attention to how the characteristics of books shaped the whole instructional enterprise.

But during the last half of the twentieth century, diverse innovations in communication and computation occurred, displacing books from their privileged cultural position and creating our current, electronic means of access to cultural achievements.

From our vantage point, we can clearly see that the microcomputer and all its attendant peripherals, that quickly matured into powerful multimedia systems, created a significant historical dilemma for educators at the end of the twentieth century. How were they to make use of these new resources in education? To what degree did the existing educational system comprise permanent, necessary arrangements? Were classrooms for twenty-five children, of similar age and talent, overseen by a single teacher, learning set subjects that had been divided into lessons, competing for grades and recognition, the way that education should be organized indefinitely into the future? To what degree were these arrangements historically relative accidents, sensible in one communication context but perhaps vestigial survivals with distorted function in a new context? In planning computer-based educational efforts, what were educators to take as givens that would remain stable, before and after the introduction of powerful information technologies?

At first, this question was not clear to educators. Early users of computers in education simply assumed that most features of the given system would remain stable, only getting better through judicious use of the new technology -- with a good deal of divergence, we might add, over what "better" might mean. We will not here rehearse all the details of these divergences. There was an initial wave of enthusiasm, and a strong undertow of skepticism, and lots of ingenious, but encapsulated, efforts to incorporate computers into the educational system. Through such efforts to introduce computers into late-twentieth-century schooling, educators became increasingly aware that the then-existing practice was a complex technical system highly adapted over centuries to making use of books as the prime medium of cultural exchange. Encapsulated innovations repeatedly engendered inflated expectations and produced disappointment and disdain.

Unfortunately, the old system had spawned a huge establishment of educational research, which functioned to optimize techniques and programs within the given system. Almost all its methods for measuring results were system-specific, assuming that existing divisions of subject matter were appropriate domains for testing, standard grade-levels were fit bases for norming results, and verbalized information was the prime indicator of learning. The bias of such research helped to protect the existing arrangements from systemic changes.

To organize education to exploit the possibilities of electronic media of cultural exchange, potentially far more powerful and flexible than the printed media, educators had to rethink the system as a whole, to take before to after the introduction of computers. Further to assess a new system, relative to the old, they had to develop a whole new type of educational research, one that did not presume in its standards of testing and measurement, that structural accidents of the old system were educational necessities of timeless applicability. The full, fundamental re-examination of educational options, and the methods for assessing them, began in the 1990's. It initiated the second historic transition in educational practice.

Looking back from the twenty-second century, the results of this reexamination are clear. Educators began to explore new solutions to all aspects of the existing system of schooling. They stopped applying computers to the educational strategies that had been developed in the early era of print. Instead, they started to search for educational strategies that seemed sensible in an era of digital information technologies.

These developments took shape in the decades preceding and following the year 2000. Educators gave up trying to introduce new technologies into the established system and they thought out an alternative system, which ineluctably displaced the old one. Since then, of course, the new system has evolved steadily, more and more thoroughly displacing the vestiges of the print-based educational system. The results have been liberating and profoundly progressive. Democracy, which had been, for the most part, a predominantly political development through the twentieth century, has gained a substantial cultural import. The persistent tendency of print-based education to reproduce and accentuate differences of power, privilege, and wealth has been decisively reversed. The great twentieth-century aspiration, verbalized by John Dewey through Democracy and Education, has become substantively fulfilled, although in an environment of pedagogical practice quite different from any he could then imagine.


Perhaps our informant from the future depicts a vision too rosy -- she says nothing about great problems like the greenhouse effect. Nevertheless, her account draws our attention to the need to look at the whole educational system in considering how to introduce information technologies into it. If computers in education are to have substantial effect improving education, they will amount to a systemic innovation, one that changes not only the medium of cuttural exchange, but the entire context for working with that medium. We propose a comprehensive, sustained research and development effort, one that aims to understand and facilitate the full complex of changes needed to create a new educational system.

Our effort should concern our informant's five interrelated domains of innovation: environment, motivation, knowledge, teaching, and educational policy. These topics, in the large, set the research and development agenda that we propose to IBM. In what follows, we first interpret more fully what is at stake in each of these five areas. We then describe our operational goals and explain how we plan to pursue them, estimating costs and equipment needs. We close the proposal by indicating why we think these beginnings can lead to a major transformation of the educational system. After these components of the main narrative, we append supporting documents concerning our projected activities, the people and groups who will lead the effort, the pedagogical theories that will inform the work at the start, and an indication of how this project relates to other efforts to use information technologies in education.

2 Making a New Educational System

We aim through our research and development agenda to initiate a new system of education, one different from the system of print-based schooling that has dominated educational effort for the past five centuries. To make such a departure, the five components essential in the construction of the given system need to be redesigned with full awareness of the potentialities of information technologies in mind.

These questions that will generate an educational system need to be explored anew. In this proposal, we will expand on each in turn and then consider how to draw them together in an integral educational system, one that may differ significantly from the mass, compulsory schooling so familiar throughout the world.

Educational Design of Learning Environments

Education has a setting. In a print-based culture, the setting is the agegraded classroom where lessons unfold according to the time constraints of the standard instructional period. We seek first, to rethink the organization of educational space and time in light of the possibilities of advanced information technologies. This involves something much more basic than simply inquiring into the optimum size and configuration for a classroom in which students will use computing equipment.

New technologies can aHer the available ways for structuring educational space and time -- they can become resources in the educational design of learning environments. Currently, electronic mail is perhaps the most familiar example of this point, for it significantly aHers the temporal frame within which consuHations between students and teachers can take place and it may also subtly change the spatial requirements of exchange, diminishing the need for simultaneity in both time and space, as in face-to-face exchange, or even for simultaneity in time alone, as with telephone exchanges. An intensive, manycycled give-and-take can occur w~hout the parties needing to be synchronous either in time or space. Thus, the physical constraints impeding one-to-one consuHation between a teacher and a student can be greatly lowered.

Very soon, however, information technologies may become even more powerful architectural resources, allowing a single space to serve muHiple functions and different kinds of groupings by switching attentional foci electronically. Thus imagine a large rectangular space, twice as long as it is wide, su~able without crowding for about a hundred persons. When a very large screen at one end was active and all attended to it, the space would function as a large room for the whole group rather like a small auditorium. Imagine that the long walls were divided into four quarters, with a display screen on one wall in the first and third quarter and on the other in the second and fourth quarter. When those four screens were active, each with different material, the whole space would be divided into four sections, each w~h. say, twenty-four people. Retractable dividers, designed to provide simple visual and acoustic buffers, could easily strengthen that division. Imagine further that throughout the large room twenty-four well-designed display consoles were placed so that the students, in groups of four, could orient to the consoles when those were active, enabling the space to work well for small cooperative groups. Finally, equip each student with a notebook computer linked to a server by a radio coupling. When the screens of those individual machines were active, rather than those of the consoles or the display walls, each person would be in a somewhat private space within a large room, with each person going off according to his or her curiosity. A schematic, involving a slightly simpler layout, will be found on page 70.

We offer this example, not to assert that it indicates the new design that should prevail, but to illustrate ways in which information technologies may make the definition of new forms of educational space and schedule possible. The existing forms of the age-graded classroom, functioning according to the standard period, do not adapt well to the use of powerful communication media. For instance, many feature films could be moving and illuminating educational tools, but they do not fit well into existing schools: neither is a classroom for twenty-five an optimum viewing space, nor is the 50-minute standard period a good time division within which to use such resources. Atternatives are possible that are currently unexplored.

Multi-media information technologies with powerful networking, tracking, and scheduling capacHies can make the very flexible use of space and time possible. We plan to work closely with established and emergent schools that have the opportunHy and capacHy to experiment with such flexibiiHies. Our inquiry will proceed according to the basic principle that the new technologies provide resources with which space and time can be organized differently than it can be without those technologies. We are dealing with innovations that invalidate the common sense that held under prior conditions; our task will be to develop a new common sense, suHable for the new conditions. WHh the old common sense, educational environments were standardized and predictable; with the new, they will be flexible, diverse -- a challenge to the imagination.

Motivational Sources of Education

Think of a fifth-grade classroom. Imagine the class dealing with virtually any subject. The teacher has just provided an explanation of a key point summarized in the text. She asks a question -- some pupils raise their hands and wave eagerly, confidant that they know the answer. Some sit in a studious effort to avoid attracting the teacher's attention, knowing that they do not know and not wanting that fact to be registered in the public knowledge of the teacher or the class. Others seem neither eager nor reluctant, they fidget, raise and lower a hand in ambivalence, thinking they know the answer but not being sure, wanting to earn the teacher's commendation, but fearing that, if wrong, they risk rejection or rebuke. These are the signs of instructional competHion at work. From the early grades through the highest levels, the existing system motivates children by engaging them in a competitive effort to shine in recitation and examination, in which each tries to show that he or she has mastered better than others the information sanctioned to be fit for his or her level and to be correct in the view of academic authorHy. As a result of this reliance on competition, the educational system functions as a powerful sorting mechanism, and when it becomes clear to many that however they may try, they have lost the competition, they drop out.

It is remarkable how thoroughly existing educational systems, around the world, have been adapted to harness competitive motivations. It is very hard to find arrangements in schools that have been designed to encourage children to act from other motivational sources. Undoubtedly the reasons for this reliance are complex, and certainly one among them is the important fact that competHion is a very powerful, effective motivator. But there are other powerful motivators, among them cooperation and H is remarkable how few educational arrangements have been designed to motivate children to learn through cooperation. The reason for this imbalance between competHion and cooperation may have had much to do wHh the logistics of working wHh printed information.

Think of a ninth-grade teacher, preparing a unit on feudalism, lamenting—

I can't have them do group projects. There just arent enough worthwhile materials reasonably available to them. New York City has all sorts of resources, but it doesnt really help -- those who would need to go to the Cloisters wouldn1 be able to get there without all sorts of complications. The school library is good but inadequate and they can1 just simply use the high-school annex to the New York Public Ubrary -- we either stay in the school or arrange, all together, to take a trip. How do I get some to the Met, others to the Morgan, and a couple into the stacks at Butler Library? How can projects be done at a high academic level in a routine way?

If it is hard to do group projects at a high academic level in a routine way in New York City, it is far harder, most other places. Sadly, serious information management problems discourage inquiry and cooperative learning, problems that must be solved if these a~ernatives to competitive learning are to become practical, everyday a~ernatives in mass education. Compet~ive motivation arises when a group of students start from an appropriately equivalent basis, usually as measured by age, and each is then asked to master a limited, standardized body of material, with goods -- praise, grades, promotion, and acceptance by the college of choice -- being distributed in proportion to how well, in comparison to others, each performs. From the point of view of information management, this practice is very efficient; it is essential in establishing the comparison that all work with the same body of subject matter. This creates a large market for inexpensive, well-chosen, clearly-presented selections, which textbook publishers compete to provide.

Cooperative learning does not make sense in s~uations where each student starts with the same content with the goal of mastering more of it than anyone else. Cooperation aims at having participants do different things and then coordinating their accomplishments in a common achievement that exceeds what each would manage alone. In educational s~uations this puts far greater strain on the information resources available to the cooperating participants. Ideally, for robust cooperative learning, students should face an expansive horizon of questions, armed with extensive resources to pursue their inquiries in many directions to considerable depth. If the questions and resources available are lim~ed, their cooperative effort will not make much sense and different members of the group will find themselves working at cross-purposes with each other, repeating each others' efforts, and vying ~h one another to do the most with the few resources on which all converge.

For centuries, educational reformers have contended that cooperative learning would be a good thing, and occasional examples of learning by working together to solve real problems keep the ideal alive. It has been very hard, however, to provide the intellectual resources to sustain good cooperative learning in most educational settings. The practice has worked best with the very young, where relatively limited materials will sustain the effort, or at the most el~e levels of education where bountiful laboratories and libraries sustain the extensive specialization of inquiry that cooperative learning generates. For the age between these extremes, cooperative learning has been very difficu~ to implement. What materials will be needed to have twenty fifteen-year-olds do a two-week unit on feudalism according to the principles of competitive motivation? Each will need a copy of a well-written text and regular attendance to a teacher who can provide supplemental explanations and manage recHations and a test. What materials will be needed to have those students spend two weeks cooperatively exploring the history of feudalism, drawing together at the end a presentation of their results? The range of possibly pertinent materials is nearly limHiess and the possible roles a teacher might take in the effort is almost boundless. Consequently, the information logistics of cooperative learning strain the print-based system.

Electronic information management technologies will significantly alter the logistical constraints on cooperative learning. One of the simplest examples of such change involves the problem of movement. Traditionally, inquiry meant that children had to leave the classroom to go to the library or other locations of specialized resources. This usually was not efficient, introducing confusion about who was where and wasting time in excess movement. WHh inquiry in a well-networked electronic environment, the children can access specialized resources, almost instantaneously, with very little waste of time or effort. Such changes in logistics can have profound effects on the experience of working together. TradHionally a simple decision -- 'I'll get this and you get that" -- would draw a cooperating pair apart, often to quite different locations, perhaps with one getting stymied on the way. In an electronic environment of information management, the two can allocate their effort while remaining in close proximHy, physically and intellectually, often checking on the implications of what each is finding for the other.

We propose to study how to implement multi-media information resources in an educational environment in order to enhance the available range of educational motivation.

These, and many similar questions, need serious examination in order to broaden the motivational energies effectively harnessed in a technology-intensive educational system.

In an educational context, "eultureM re:sulb from a selection and evaluation of the sum of human acquirements, narrowing the infinite range of possibilities to a finite field, one that nevertheless exceeds the power of acquisition of any individual by a wide margin. The effort to organize culture and knowledge through this project will take place in a context set in part by disagreements between proponenb of "culturalliteracy,N a fairly narrow, canonical selection, and "multicultural" approaches, a broader, indusive selection. We do not intend to define what we mean by Mculture" relative to this polarity as if the positions within it stand above the material limitations of particular implementations of education. Instead, we think the terms of the debate between cultural literacy and mutticuhural education will be reshaped substantially by the development on a new system of education that use:s information technologies with full effect.

Organizing Culture and Knowledge

We speak of the print-based school because printed materials have been the main medium for making culture and knowledge accessible to students. So long as there has been no alternative to this reliance on printed materials, educators have paid close attention to the pedagogical features of one text compared to another, but there has been little attention to the pedagogical character of printed text, per se.

If we inquire into this latter matter, we see that the logistics of working with printed texts have much to do with the sequential character of the existing curriculum. Developmental psychologies delineated the sequences of major stages in the child's growth. But educators should not exaggerate the degree to which curricular sequences, differentiated by year and by subject, have arisen by rationally adapting curriculum and instruction to children's developmental needs. That world history should be a tenth-grade subject and American history a eleventh-grade one, or that biology should precede, or follow, physics or geology has little to do with the developmental characteristics of children. It is largely a conventional solution, one among many, arising from the need to divide the curriculum up into discrete subjects that can be presented in some sequence. The need for sequence is inherent in the constraints of print, not those of psychology. And whether it should be this sequence or that sequence is comparatively an inconsequential question.

What does it mean to move from fifth to sixth grade? A child who does so usually changes teachers and rooms, sometimes even a building, but these are not the essential changes -- the child could move from fifth to sixth grade while staying with the same teacher in the same room. What changes from one grade to the next is the curriculum, and most importantly the set of textbooks the pupils use. Sixth-grade texts differ from fifth-grade texts and so on and as the child progresses through school he or she does not cumulatively carry the texts from prior grades around. Educators have long likened the curriculum or sequence of grades to a ladder or staircase because the sequence of texts are like the sequence of rungs upon the ladder: one climbs from rung to rung, leaving the last one behind. Students in any particular grade find it hard to regain access to the materials studied in prior grades, without somehow going backwards, and they find it even harder to anticipate access to materials slotted for grades higher up. Unable to move easily, back and forth, pupils experience the curriculum as a set of sequential studies. The costs are high. If a pupil did not get one part of the sequence, the omission can be portentous, not because the sequence is the only way things could be reasonably mastered, but because, once missed, the opportunity to make it up may be very hard to regain.

Pupils will have a very different relation to a computer-based curriculum, assuming that the whole body of culture and knowledge relevant in education has been integrated into a comprehensive system, any element of which they can access at any time from any place in the school. With continuous and ubiquitous availability, the sequence of grades would loose much of its meaning and study would be experienced as a cumulative effort.

We propose to investigate how the subject-matter appropriate for a complete and excellent education can best be organized, making no assumptions about the year-by-year sequencing of its presentation. A smart, computer-based curriculum should be able to sustain an infinite number of paths through it, and it should be able to provide each student with clear reports about what he or she has so far covered, regardless of the path and sequence he or she has taken. With such resources, we will ask questions such as these:

Our main task will be to prototype a cumulative multi-media curriculum, one with a comprehensive selection of materials in it all of which will be accessible to all students at all times. We will implement this curriculum in working educational settings and pursue answers to questions like those above through intensive participant observation in these sites. In a nutshell, this set of activities defines our research agenda.

In addition to potentially making the content of the curriculum accessible to children in a much more cumulative way, we should note another, profoundly important structural effect on subject-matter that the new technologies will have. For five centuries, written materials have been the main channels of access to culturally significant knowledge. This dominance of wntten communication arose because printed texts developed a level of accessibility radically different from other modes of embodying cultural expression. Access to printed materials could be general, efficient, and enduring. Access to other forms of cultural embodiment was comparatively restricted, troublesome, and transient.

Engravings, woodcuts, and other forms of printed images are a partial exception to this assertion, except that accessing them requires one to manipulate the written language, not pictortal images. Thus, to retrieve pictures of 01artres Cathedral, one must use written catalogues and indexes.

To grasp this point, consider the theater, the drama, and its place in education. Multi-media are not new. Their significance pedagogically has grown of late. One often encounters the text of Shakespeare's Hamlet and other great plays as works taught within the curriculum. Productions of one or another play may be significant extracurricular activities in schools, and students may be encouraged to see a professional staging of them, should such performances be accessible in their locale. Nevertheless, the performances, whether student produced or professionally produced, have not been the central educational use of the drama during the era of pnnt because access to the performance has been highly idiosyncratic and temporary, whereas access to the text of the play would be general and enduring. In the era of print, wntten materials have dominated educational effort from the most elementary to the most advanced levels because these have been the materials to which access has been general, efficient, and endunng. A radical departure is afoot because now electronic information technologies can provide general, efficient, and endunng access to a much broader range of culturally significant materials: recorded performances of the play can be as easily retrieved as its text. The educational consequences of this development will be vast.

"Verbalization" here refers not only to the spoken word, but even more essentially to the written word and even conceptualizations communicated through the symbolic notations of mathematics and the like. In its most comprehensive form, the basic proposition of verbalization is that higher· order thinking consists in manipulating symbolic notations.

Networked, multi-media electronic systems provide general, efficient, and enduring access to cultural works of nearly every form conceivable. In the era of print, written works had a supenor cultural usefulness than other resources. People could distribute, store, cite, retrieve, and use printed resources far more effectively than they could work with other forms of cultural expression. Essentially, one has long been able to refer other people to printed materials without knowing the particular physical location of the particular instance of the material that people will consult, for one cites ednions -- Plato, The Republic, Book IX, 592b -- the numerous instances of which are scattered at many places. Paintings, plays, sculptures, and buildings, in contrast, cannot be referenced in this generalized way -- they exist in unique locations and access to them can require taxing trips, even a pilgrimage. Owing to this supenor accessibilny, printed materials, usually wntten materials, have more and more mediated the production and communication of knowledge in modem culture. Let us sum up this development: in the era of print, verbalization increasingly dominated education.

More rigorously, video tape has so far initiated only half the process. The superior accessibility of printed materials has rested on two related, but different, techniques •• title pages and pagination. T.Ue pages allow people in many places at many times to easily gain access to the same work. This can be immensely helpful, but if the work in question is large and complex, it still leaves people with a diffiruh problem of Ngetting to the point." Pagination within the edition solves this problem, al~ing a reader not only to get to the right work, but also to the right place in the work, with eaae. The commercial availability of videotapes allows a community of viewers, dispersed in time and $p8C8, easy access to the same film, exacdy as a community of readers, dispersed over centuries, have had with important books. The second technique, analogous to pagination, has not generally been reached, however, evan though technically it is a trivial task to put searchable timecode on the tape. Videodisc and the full conver$ion of analog media to digital form are rapidly completing the second stage of retrievability.

Slowly through the twentieth century, and building rapidly at its end, other modes of exchanging information, ideas, and knowledge between people are gaining cultural power relative to printed text. For centuries, texts have been available "at any place at any time' -- that has been their power. With the rise of the broadcast media, first speech through radio and then the moving image through television gained part of the power of print, becoming available "at any place," provided one tuned in at the right time. The recording industry gave music full accessibility, independent of particular place and time. Video tape is giving the same to the moving image, and very soon, with fully interactive multimedia systems, the superior accessibility of text compared to other forms of expression will completely disappear.

As we here use "verbalization~ to describe far reaching assumptions about the relation between words and symbolic notations lo higher-order thinking, so we here use "visualization" expansively to situate reflective thinking in pre-linguistic forms of perception and awareness, which may then be expressed through words and symbolic notations, or through images, sounds and all manner of asaociations and actions. In this sense, •visualization" is not a mere opposition to verbalization, not a simple attemative to it, but a Hegelian Aufhebung of it, the upheaval of it Into something else in which the original form remains nevertheless included and preserved. "Visualization" in this extended sense indudes "verbalization" as one among a number of different forms of reflective thinking, which all should be integrated into a comprehensive and many-sided culture and education.

When people speak about interactive, mu~i-media systems, they are speaking about a process by which the full gamut of human expression will integrate into one complex system, with all components, regardless of form, being generally, efficiently, and enduringly accessible. This integration, enhancing the accessibility of all forms of expression, we will call visua/ization. A powerful trend toward the visualization of education is taking hold. Its historic effect will be to broaden effective participation in the culture greatly.

AND Communications, for instance, has produced an interesting demonstration tape for IBM Educational Systems in which James E. Dezell, Jr. shows how shJdents might work with a muhimedia unit on Tennyson's 'Ulysses.~ The demonstration provides a good example of what we here call visualization and Mr. Dezell appeals at the end to his audience asking whether, in the eighteen minutes the demonstration took, they haven't learned an unusual amount about Tennyson's poem and all that associates with it. The audience acclaims an affirmative, but one wouki be hard pressed to devise a rigorous measure of what they learned, and even more, at what individual pupils working independently with the system might learn. What sort of student evaluation would be appropriate at the condusion of such a unit? Surely not a simple test, multiple choice or otherwise, of the n..denfs recall of information in the unit that the tester deemed essential. That would defeat the central idea of the system, which is that there should be more that is available and worth laarning than any single user will master and that many different samples of that totalrty represent worthwhile engagements with the material. Evaluation of learning with such systems will have to somehow assess the scope and sample of a s'b.Jdenfs knowledge relative to the universe of what might be learned. This will not be a simple thing to do.

We propose to investigate how to facilijate such visualization of education. This requires that we find how to create muHi-media presentations and to stock the curriculum with them and to present the whole effectively to students. Such inquiry reaches very deeply into the basic assumptions about teaching and learning and educational common sense. Gone is the finite body of subject matter that should be teachable and that students can be held responsible for learning. The idea that good learning consists in apprehending what has been taught no longer will hold. Wijh multi-faceted curricular resources, which can sustain many valid paths of inquiry within them without any inquirer exhausting all their contents and permutations, one cannot specify precisely what has been taught. Hence, our effort also requires that we develop important new forms of evaluation. Verbalization pervades existing forms of evaluation, which usually test the recall of verbal information and assess students' capacities to express their ideas in writing. With extensive visualization in education, other ways of testing the resuHs will become essential.

Preliminary thoughts about principles that can suitably guide the construction of such systems are in Appendix 8, Section 2, "Toward a Pedagogy for the Cumulative Curriculum," page 167.

In sum, then, with the effort to develop a computer-based system of education, we not only need to explore new ways of organizing culture and knowledge, we need to recognize that we change fundamental assumptions about education by doing so. With the print-based system education has consisted primarily in imparting authoritative selections of material to students who are responsible for learning them. With the electronic system the scale of the authorijative selections of material jumps significantly and the student can no longer be held responsible for simply learning them in full. Instead the student becomes responsible for intelligently exploring them and taking from them a unique but sound and useful sampling. Formal learning thus becomes much closer to experiential learning. The student needs to become a skilled explorer, not a docile learner; the teacher becomes, not the master, but the native guide, like Virgil to Dante, interpreting, elucidating, cautioning, exhorting. We propose to try to uncover and understand such changes by implementing a prototype of the new system and attending closely to how it changes patterns of educational activity wRhin it.

Improving the Conditions of Educational Work

Teaching in the print-based system has required skilled professionals. The earliest Protestant theorists of schooling pointed to the importance of well-trained teachers, if the system were to be effective. And the need has been constant since then. Nevertheless, the conditions of educational work within the printbased system have had significant deficiencies. Teaching a set curriculum with set texts tends to be highly repetitive, year to year, and teachers often find their work routinized. They cannot do much beyond the text and after a few times through, the text becomes a familiar locale that ceases to challenge their imaginations.

Allied to routinization is deskilling, which is a kind of routinization that happens, not as a by-product, but as the purposeful resuH of policy. When work requires higher levels of skill than the average worker may possess, managers have often tried to simplify the job, believing that to be a more efficient solution than improving the skills of the worker. Complex tasks once performed somewhat unpredictably by high-paid skilled artisans were analyzed into component steps that anyone, following instructions, could passably perform. Unskilled workers replaced the artisans with the process tightly managed according to the principles of Frederick Winslow Taylor, and the output became predictable and the production costs minimal. Curriculum developers have sometimes used these techniques to seek a "teacher-proof" curriculum, hoping thereby to better guarantee resuHs and to get by with lower pay for less-skilled teachers. In many industrial areas, such processes have reduced numerous artisans to mere machine-tenders, mindlessly repeating dumb tasks as products wend toward completion along the line.

An industrial system that achieves production efficiencies by steadily lowering the skill requirements in many forms of work over several generations can find itse~ in trouble should the skill requirements of work suddenly increase. Advanced technologies in the workplace have caused precisely this shift in recent decades. In factory and office, deskilling jobs had made much work diseducative. And educational preparation for work in such jobs put a premium on rote learning and routinized teaching in the "factory school" where students were primarily acculturated through drill and practice to follow instructions wRh uncomprehending accuracy.

Twenty years or so ago, teachet'·proofing the curriculum was a fashionable topic in the profesaional education literab.Jre. Now teacher-burnout has displaced it as a topic of considerable concern.

Increasingly, high technology reverses the polarity on the skill needs of labor in the industrial and service sectors. Machine-tending jobs, performing a single task according to a prescribed manner in a complex division of labor, are growing scarcer. Process-managing work, controlling a complex system by mon~oring information about the condition of its parts, has become more prevalent. In them, a mindless mistake can prove most costly. This shift in polarity carries all the way through the educational enterprise. Learning to learn and critical thinking are fast becoming important educational resuits, not only for the most successful, but for all who go through the system. In such a situation, the demand arises for more highly skilled, fully engaged teachers. Hence it is becoming socially important, not to simplify instruction so that any teacher, no matter how unskilled, can make it work provided he follows instructions, but to structure it so that the teacher will continually develop his skills, growing more and more adept with more and more experience.

We propose to investigate ways in which a computer-based educational system can be designed to engender continuous growth and development in teachers, to provide them with an intellectual environment that does not become repetitive or static, to equip them with resources with which they can be effective from the start of their careers, while continuing to develop their skills as their experience grows. Above, we surveyed possible changes in how educators motivate students and how they organize cuiture and learning to provide broader, more engaging access to it. These changes can have profound effects on the work of teaching, changing what teachers do and can be held accountable for, expanding the opportunities for personal growth. As that happens, the profession of teaching may become more attractive to the talented and ambitious, salaries may increase, and the prestige of the teaching profession may rise with the public. Such are the hypotheses that we seek to explore in this sector of our effort.

Strengthening Public Participation In Education

In order to construct a technology-based educational system, the level and structure of expenditures on education must change. Currently, K-12, expenditures on instructional materials per pupil amount to a small fraction of the total per pupil expenditures. Most go instead for salaries of teachers and staff. Here are the estimated average per pupil expenditures and national totals for public elementary and secondary schools for 1987-88.

Per PupilPer CentTotal
Elementarv & secondary schools, total$3,703100.0$149,420,000,000
Administration$1664.5$6,701,000,000
Salaries of Instructional Staff, total$2,15158.1$86,798,000,000
Other expenditures for instruction$3419.2$13,779,000,000
Operation of Plant$2797.5$11,238,000,000
Maintenance of Plant$1042.8$4,215,000,000
Fixed Charges$3639.8$14,637,000,000
Other School Services$2998.1$12 051,000,000
Capital Outlay$395$15,932,000,000
Interest on School Debt$122$4,925,000,000
Total enrollment40,350,000

The line, 'Other expenditures for instruction," seems at first to represent a nearly sufficient sum, close to $350 per pupil per year, to have substantial technology outlays for each pupil. The following table provides a breakdown of it, showing far less to be potentially available for investment in technological resources.

Per PupilPer CentTotal
Other expenditures for instruction$3419.2$13,779,000,000
Clerical assistance to instructors$842.3$3,372,000,000
Teacher aids & para professionals$280.8$1,112,000,000
Free textbooks$260.7$1,056,000,000
School library books$170.5$677,000,000
Teaching supplies$491.3$1 ,994,000,000
Other instructional expenses$1383.7$5,569,000,000

These figures are approximations, roughly sound, pending receipt of the official figures for the most recent reporting period. They were generated by applying the 1971-72 proportions for each category to the total expenditure for public elementary and secondary schools for 1 987-88.

The first two subcategories represent further salary expenses and the last line covers things like travel, graduation ceremonies, and other special aspects of the academic program. If all textbooks, library books, and teaching supplies were devoted to technology in a computer-based system with expenditure patterns otherwise similar to those currently in force, $92 per pupil per year (2.5% of expenditures) would be available for hardware and software per pupil. True, this amount would aggregate nationally into a lot of money, but it would be far from sufficient to provide the technical investment needed per pupil for a fully computer-based educational system.9

With a heavy infusion of technology, other educational costs will not decline. Were we to assume that the technology makes it possible to have fewer teachers, a questionable assumption, the level of average salary for teachers would likely rise proportionately, keeping labor-related costs even. To implement a technology-based educational system, total per-pupil expenditures will need to increase significantly with a substantial increase in the expenditures for instructional materials and equipment becoming possible through that increase. Let us guess: a robust computer-based system would require spending about $500 per pupil per year (about 12% of a total $4200) on hardware and software, with other expenditures remaining about as they are. This works out to an increase in total public school expendnures of about 13.5%, an increase sufficiently large to require a compelling public justnication before it would be incurred. At that level of expendnure, however, we are postulating an annual American market for educational hardware and software for public elementary and secondary education, alone, of over 20 billion dollars, so n may well be worth exerting the serious effort!

Having to develop powerful justifications for substantial increases in educational expendnures is not a novel challenge. Universal, compulsory school systems evolved as nations found reason to devote increasing percentages of their GNP's to such costly instructional efforts. Over the past five hundred years, the provision of education has never become more efficient, providing significantly more output for substantially less input. Rather it has become more important, more valued, with the public deciding that increased educational resutts are worth increased costs. In precisely this W81of, policy justifications for a computer-based educational system will need to convince the public that the increased costs bring benefns that justify the added expendnures. Developing those justifications will be an integral element of efforts to invent the new system. Wnhout them, the only things that will seem realistic from a marketing point-ofview are the creation of specialized adaptations to the existing system that fit within the existing per pupil expendnure patterns. Those might make a few entrepreneurs weatthy, but they will not create a systemic change in education.

To institute a new system of education, educators will need to marshall large arguments of broad public purpose. These are of three general types: arguments of relative national advantage; arguments concerning the achievement of basic civic goals; and arguments that something is a good an end to be sought for its own sake, not merely as means to something else. We propose to investigate the degree to which the development of a new, computerbased system of education can provide grounds for each of these types of arguments.

Relative national advantage. In the late 1950's the pursuit of relative national advantage through education resulted in the National Defense Education Act. Now the quest for military preparedness has given way to the problems of remaining competitive in a global economy. A fully developed computer-based educational system can have three types of significant effect on the over-all competitiveness of the American productive enterprise.

We hypothesize that all three of these benefits would develop with a shift to an intensive use of information technologies in education and we will seek to develop and test such hypotheses.

Achieving basic civic goals. Although the agenda of freedom has taken great strides in most recent history, that of equality has been stuck in stasis. Significant portions of the population are addicted, nearly unemployable. Uncounted families cling to survival -- homeless street people hawking Street News, begging, and scamming, and groveling through the refuse of the relatively rich for cans and bottles redeemable for a nickle each. Many, prideful on having made it, blame social failure on the failings of those who suffer the failure, and espouse social policies designed primarily to help those already adept at helping themselves. In our willingness to bail out bankers who mismanaged the savings of the middle class while we cut back on programs to serve those most in need, we poorly represent the centuries of humanitarian progressivism that has animated our traditions. The agenda of equality merits substantial civic effort.

Educationally the problems with the agenda of equality appear in the prevalence of drugs, dropping-out, and the difficulty of making schools work in areas of chronic urban and rural poverty. These are big problems and they will require complex solutions. Efforts to deal with these problems by further improving the current system of schooling will not work. Unfortunately, the current system is part of the cycle of causality, not a means of breaking it. Interpreters of the education system tend to be people who have done well within it. They experienced schooling as a happy system for self-development and self-advancement. Interpreters often therefore have difficulty seeing how the experience of the less successful was fundamentally different: for many others, the same system functions as a powerful social sorting mechanism, frustrating their self-development and reinforcing their disempowered status. A computerbased system of education may help break the cycles holding the truly disempowered in thrall in three ways.

First, the way the current system handles subject-matter is invidious. Size limns on textbooks require that a very limited selection of ideas and information be packaged together. The materials chosen become, ipso facto, sanctioned by inclusion in the standard texts and tests. The resu~ will harmonize with the experience of some children and be at odds with that of others; some will find more with which they can identify emotionally, and others less. Interest groups realize that what the selection includes and excludes has import, good and bad, for their interests. However, where the scope and compass of texts is seriously limned, the polnics of textbook development and curriculum design has become a contest to exclude whatever particulars may offend an articulate sensibilny. Increasingly, such efforts to exclude all possible cu~ural bias tend simply to render the curriculum pedagogically impotent for all. In contrast, computerbased curricula can be comprehensive and inclusive. The politics of a computerbased system has the possibilny of opening the narrow confines of the standard curriculum to genuine mu~i-cu~ural possibilities. Wnh the new system, the politics of curricular development will cease to be exclusionary, becoming instead a many-sided effort to ensure that what may empower this or that interest finds its place within the spacious system. We propose to work with diverse racial and ethnic groups to develop multi-cu~ural curricula through which high levels of disciplinary mastery can be achieved along numerous paths of interest and inspiration.

Second, reliance on printed sources in the current educational system provides a narrow access-path to the power of knowledge. Those who experience the existing system as disabling do not do well with book learning. To be sure, in theory the system offers them vocational tracks, which put greater stress on learning to make productive use of hand and body. But these tracks have a stigma associated with them because everyone knows that in a printbased culture the only real access to knowledge is through verbal facility, and no matter how manually skilled one becomes without high levels of verbal knowledge one will be held mentally second-rate. Insofar as a computer-based system can complement the verbalization of print media with the visualization of electronic media, mu~iple access paths for acquiring and manifesting mental excellence will open. This will not do away with distinctions between people with respect to intelligence and intellect, but it can broaden the existing structure of intellectual opportunity. We want to implement a thoroughly technology-based curriculum so that n will make full use of the power of visualization in our cu~ure.

Third, the way the existing system motivates educational effort through pervasive competnion creates a sorting mechanism that deprives disempowered groups of their more able members. Those who succeed in the competnive assent often assimilate to the dominant elnes. Imagine an educational system that did a better job at fully developing the potentialnies of each person while less effectively grouping and sorting the members of age-cohorts according to their performance on a narrow set of mandarin acquirements. Such a system would be a very different response to the Jeffersonian idea that talents distribute randomly through a population. Rather than co-opting those talents to the service of power and privilege, it would preserve those talents in their random distribution, leavening the whole through a multiplicity of communal excellences. A computer-based educational system that can put a premium on cooperative learning and offer a mu~i-cu~ural curriculum with many paths to mastery within it has the possibility of functioning in this more genuinely Jeffersonian manner. We seek to implement a technology-based system of education in order to explore the possibility of such socio-cu~ural effects.

Education as a good In Itself. Soon the ethos of "More" must give way to an ethic of "Enough." As that happens, problems of public purpose will remain, but they will undergo revaluations. In the era of print, the basic justification for the support of education has described schooling to be a useful means to the achievement of publicly sanctioned ends. For the past few centuries, those publicly sanctioned ends have often been variations on "More" -- more power, more wea~h. more influence, more adherents, more law and order, more of what-have-you. What good did the print-based system serve as it mobilized competitive energies to distribute broadly a level of literate skills through the population and sorted the young effectively according to the quality of their performance within the system? It served best as a means in the pursuij of "more." Would it serve well in support of an ethic of enough? One might suspect that a system that relied on cooperative learning, that could attract participation in educational self-development, not as a means but as an end itse~. providing an enhanced quality of life, bonds with others, shared experiences of personal meaning, would be an education well adapted to the ethic of enough.

A competitive ethos of "more" can take hold among people when they feel they can safely compete for possession of finije, limited goods. Where the competijors become aware that the competition is fundamentally unsafe and unstable they withdraw from the unbridled continuation of it. In the late twentieth century, the age-long competitions for national power, pursued through the pursuij of more population, more armaments, more material output, has become increasingly unsafe and unstable as armaments become too destructive to use, populations too large to feed and nurture, and material output exhausts natural resources and threatens to destabalize wor1d climates and ecologies. Unlike the various forms of power, which are finite and relative, education is not a limijed good. More education for one need not mean less for another. Hence, education can be a public purpose, one pursued by each and all, without an inherently limiting competition arising, without one person being pitted against the other. Taken as a means to relative advantage, people have an interest in acquiring learning and withholding it from others. But taken as an end in itse~. people have an unbound mutual interest -- the educational attainments of others enriches the educational possibilijies that I enjoy. "A learning society" is a society adapted to a wor1d of finite resources. A computer-based educational system is not the only possible basis for a learning society, but insofar as it can supplant competitive educational motivations with cooperative ones, and insofar as ij can genuinely broaden educational opportunity by opening mu~iple channels to knowledge, it will facilitate the emergence of such a society. We propose to explore how well education can serve, in-nself, as the basic goal of civic life, serving not merely as a means to some other purpose, but as the good animating the common pursun of the good life.

3 What Is To Be Done

We aim to prototype a new system of education, one based on the intensive use of mu~i-media information technologies as the primary means of access to the cultural content of education. To accomplish this purpose we need to be mindful of its complexity. We cannot simply apply multi-media technologies to existing components of the educational system, designing technology-delivered lessons for the existing curriculum. Likewise, we cannot simply layout the five major components of the task in a sequential order -- first designing a new environment, then developing a new pattern of motivation, with that followed by a comprehensive effort to organize the cultural content of education according to the new principles, proceeding to set up training programs for teachers, and concluding with an impassioned call for public support of the new system. To work well on the tasks we need to proceed with them all, as much as possible in parallel, with an effort to note which define the critical path, the timing determining when the whole can begin to function.

Let us try to situate where things stand in a schematic way. From that baseline we can chart what is to be done. In this, we will use single terms -Environment, Motivation, Knowledge, Profession, and Policy -- to denote the complex domains discussed above. The current situation confronts us with isolated pieces and parts, each of which has been fashioned without concern for its place in the whole. Educators are gaining experience with the diverse components of a new system, but that is not the same as beginning to put a preliminary version of the new system together.

Summaries of the Current Situation

Technological innovation usually follows an S-curve, along which change accelerates, ascends, and then tapers off. Many of the specific innovations in the acceleration and ascent phase but not very indicative of what the technology will be like as it levels off into ma'b.Jrity. The standalone miaocomput:&r with a character-based screen and software on a floppy disk is as much an artefact of transition technology as are punchcards, starter cranks, or house-top TV antennas. We will use the term Rmature technology,• in contrast to transition technology, to describe the technical characteristics of a system after the pace of innovation has slowed.

Project planners, faced with such a complexity of tasks, need to decide where the critical path to their implementation lies. Work can proceed in parallel on several aspects of the problem, and in due course a fruitful interaction between the different aspects can emerge. But one of them will be pivotal, critical, defining how quickly the whole project can move ahead. With respect to the new educational system, the critical sector is that of Knowledge, the reorganization of knowledge for presentation through multi-media systems in ways that differ significantly from the presentation of knowledge through printed materials. Without that, new environments will be visionary, not usable; strategies of motivation will be hypothetical, not actual; reforms of the teaching profession will be utopian, not practicable; and policy rationales will be unrealistic, not down-to-earth. In order to develop the new system, we need to draw together a full multi-media knowledge-base that can serve as the cultural content of a complete education. Without that we are not coming to grips with the challenge

Operational Goals of the Project

We do not suggest here that the way any particular child experiences the multimedia curriculum will be devoid of sequence. The curriculum must sustain many different sequential encounters, and hence it cannot be phased in, step one, step two, step three, according to a cannonical sequence. Here we are discussing the problems of curriculum development, not the characteristics of the pedagogies that should guide its study.

In order to think out the process for creating this full muHi-media knowledgebase, we again need to put established curriculum-development paradigms in abeyance. Essentially we must substitute a cumulative resource for a sequential one. With the sequential curriculum, developers would work grade-by-grade, following, with one or another distinctive variation, the typical and mandated syllabuses for subject and age-group. We are discussing a new system precisely because a comprehensive electronic curriculum will not be sequential in the same way that the traditional curriculum is and consequently we cannot use the same strategy of design and development. In order to develop a traditional curriculum step-by-step, one can do it a grade at a time. To develop the new curriculum through a series of steps, something different must be tried, namely filling out a sparse matrix.

Think of a painter composing a large canvas. She does not treat the whole as a sequence of component panels, say three rows of four, starting at the lower left and completing the first panel, then moving on to the one to its right and so on until all twelve are executed, one by one. She would, instead, sketch the whole composition, making sure that the essential components of the composition balance and interact effectively. Then she might roughly fill in much of the painting and bring some parts to an early, tentative finish while leaving other parts close to the first sketch. Working here, then there, sometimes making significant alterations in the original conception, she completes the whole work. Such a procedure works, with adjustments, in diverse areas of creation, where people create functional wholes. We call this process one of filling out a sparse matrix, and it is the form of implementation especially appropriate where the elements have not been previously organized into a set sequence.

A movie theater generally fills also in a random way compared to the sequential boarding of the airplane, with people seeking to situate themselves in the theater with least obstructed views according to one or another preferred perspective. Some want to be up close, others way back; some by an aisle, others in the middle. Even with only a few people in the theater, most of the seating regions will have someone in it, and as the theater fills, the more popular areas will pack dense first, but all the areas will fill out according to the play of preference. The multimedia electronic curriculum should to be implemented in this way, with something in all its parts, but more in some areas, the whole filling in, something here and something there, until all that should be included has been included

To get a better sense of how the lack of sequence affects the process of completion, consider a different, but similar, example. Compare how people distribute themselves in a subway car with how they board an airplane when having been assigned sequentially ordered seats. The airline attendants will call the passengers by rows, boarding, in this case, from the back to the front in a way similar to how one implements a traditional curriculum grade-by-grade. The subway car in a more random way, as a sparse matrix, with each new passenger usually seeking to situate hirnseH with the maximum open space between hirnseH and other people. The basic algorithm of maximum face-to-face distance will ba further adjusted by factors based on gender, ethnicity, age, class, and comportment, but however modulated the car fills up, not sequentially, but diffusely, with all the space baing occupied all the time at variable levels of density. The multimedia electronic cur~culum needs to ba Implemented In this way.

Immediately here the question springs forth: how will younger children be ready to study such an advanced body of knowledge? We cannot answer this questions, lacking suitable curricular tools to have acquired experience with which to form an answer. We aim through our project to create those tools in order to seek answers to this question. We can observe at his point, however, that the question gains much of its force from our inveterate habit of thinking about the course of study as a sequential one in which the child moves through the succession of grades as a preparation for exposure to the content of undergraduate study. We confuse the way we do things with the way things must be done to the point of loosing sight of the fact that we also do things in other ways. Thus, the child did not learn his mother tongue through sequential study: he was immersed in the milieu where it was spoken and he learned it by exploration and use, with that learning being cumulative for life.

How large should our subway car be, so to speak? What should be the full capacity of the cumulative curriculum? However sparsely filled out at the beginning, its scope should ba the full range of knowledge that a highly able undergraduate might be axpacted to have acquired on having completed his or her general education. We want to emphasize that this college-centered scope is appropriate for the K-12 curriculum. We have two main reasons for adopting this scope. First, the existing system already incorporates the content goals of undergraduate education by providing a means by which students can earn 'advanced placement' credit. It would not ba appropriate to have a lesser scope for the new system. Second, if the new system has a future, it will need to perform more effectively than the old one, by a significant margin. If the existing system allows the gifted few to achieve advanced placement, the new system should engender something approaching that level of performance from the average student. For these reasons, we conclude that the scope of the new curriculum should to be that of good undergraduate general education.

Given this scope, how will the project proceed? What will its specific activities be and how will it be organized? Here are operational goals that we want to achieve, with an indication of the timespan during which we will actively work on them.

In each of these areas we have made preliminary, but comprehensive, inventories of activities to be accomplished. Here we provide a summary indication of the work to be carried out in each area.

We believe these costs can be kept to a minimum. Too many people think of educational software using an analogy to the textbook. We suggest, instead, that the textbook will be displaced by a well developmed implementation of educational technology. Such an implementation will be much more like a school library than a school textbook. We think that a large collection of tapes and digitized materials can be acquired like books in the library and used withou1 further royalties throughou1 the school as requested by individual students. To make this possible, it will be important that the linking and navigating software store information abou1 materials in idependent databases and not require that the information be uniquely recorded in the material to which it points. Thus, for instance, if our curriculum makes use of a sequence from the Adams Chronicles or any other tape by including in it machine instructions and capacities to load, cue, and play the tape, we would not need to acquire rights to the tape, but would rather simply need to have bought a copy of the tape that we hold in the curriculum collection.

Project Structures

[Not included]

4 Designing a Historic Transition

To fulfill the potential of information technology in education, educators need to invent new solutions in the basic areas that generate an educational system:

But can these things be accomplished? Why should we believe that we can change the current system, with its vast inertias of scale -- contracts, regulations, procedures, customs, and conventions -- and all its vested interests -- the bureaucracies, unions, associations, clients, and pressure groups -- that thwart smaller-scale reforms? Have we longer-term strategies for converting our prototypes, if they succeed as prototypes, into systemic innovations?

To begin, let us note what would define success for our prototype. We base this project on an implicit premise: traditional schooling with the sequential curriculum has been optimized as an educational technology and now offers few opportunities for significantly improving its performance. Some evidence seems to challenge this premise. For instance, international comparisons of school achievement indicate significant differences in the level of attainments, one nation to the next. So, too, the substantial differences in achievement between schools suggests that if all traditional schools could perform as well as the best among them, a major advance in educational performance might be attained.

Both comparisons, we think, suggests illusory expectations for further perfecting the prevailing system. Neither set of comparisons discriminates persuasively between the school-based determinants of educational achievement and the non-school determinants. Thus the observed differences do not necessarily reflect differences in the quality of schooling, but rather differences in the conditions under which the schooling takes place. American social policy is seriously retrograde in comparison to social policy operative in welfare of our least advantaged citizens, urban and rural, compares poorly to conditions in other advanced nations. Such disadvantages can account for substantial parts of the differentials in school achievement revealed by international comparisons. Indeed, our low ranking on indicators of school achievement parallels similar low rankings on many other social indicators. Further, differentials across nations, localities, classes, and ethnicities may arise from all sorts of non-school factors -- the way the culture of mass entertainment and consumption catches the attention of the young, how an ethos of boredom, despair, and underachievement spreads within groups and regions. In short, we assert our premise believing that most of the observed differences in school achievement result from non-school determinants: the existing system of schooling is a mature system that cannot perform significantly better than it does as a result of adjustments within it.

Our prototype will be successful if it can serve as the enabling factor for a new system of education that will have a different, more desirable spectrum of achievement associated with it. The prototype will succeed as a prototype if it demonstrates the feasibility of an alternative system, a system based on a cumulative, not sequential, curriculum, one that enables access to culture through an integrated mix of media. Assuming for now that the prototype succeeds as a prototype, have we reason to believe that it can translate into a system?16 It is easy to imagine a successful prototype that, despite an effort to convert it into an alternative system of education, does not do so. Failed reforms are legion in history. The early twentieth-century movement for progressive education stands as an evident case in point. Progressive prototypes worked well; their conversion into a new system flopped. Why would the Cumulative Multi-media Curriculum be different?

One of the most successful innovations in twentieth-century education, standardized testing, exemplifies such materiality. The tests were a new physical tool, administered in places and times set apart for them, scored with new technologies. The material reality of standardized testing is crucial to the way it has so quickly and so widely taken hold, the way it has generated the resources to support its further development and extension, the way it commands attention, forcing schools to teach to the tests and arbitrating success and failure throughout the system.

The late Dr. Ben D. Wood, whose generosity and vision led to the creation of the Institute for Learning Technologies, was a pioneer in the testing movement and someone who keenly sensed the importance of tangibility in agents of educational change.

To change a system, one must change its material conditions. Many failed educational reforms have been reforms in idea only, changes in method and guiding principles that left in tact the material life of schools. Such failed reforms have been designed to take hold in the given environment. Indeed, people even hailed that characteristic as a sign of the practicality of potential reforms. But ideas that do not affect material life have little historic staying power. The new system of education that we envision will change the daily realities of education; it will change the patterns and structures of educational activity, the locus of study and the tools employed. The new system will transform education by transforming the artifacts of educational activity.17

Indeed, one might argue that a reform as intellectualist as the Protestant Reformation took hold in those areas where the material reality of Catholic liturgy and worship were effectively suppressed and did not in areas where it survived. The reformers redesigned places of worship, radically altered their decor, changed clothing styles and all sorts of details of material life.

Stop to wonder how a craftsman succeeds in teaching his apprentice or a mother passes her skill in cooking to her child. The pedagogical art is small but sure because the lesson is enforced by the actuality of the activity. The master points to the task and gives a tip when the time is ripe for the apprentice to grasp a new knack. The hammer has a balance and a feel that allows the novice to quickly sense how the tool should be used. Culture, not as things thought and said, but as things made and used, determines what people do and do not know how to do. To change educational ideas and principles one needs to change the tools and conditions of education, otherwise the old ideas and principles will reassert themselves through the weight of doing what is done.16 The teacher's craft is not independent of her tools, but evident, as with any craft, in the mastery with which she wields them.

Unlike many efforts at change and reform that have centered on educational "method," leaving structures and conditions in tact, we aim to design an alternative material reality, changing the place of education, its schedule, and the tools and resources in daily use. Our prototype, if successful, will show that a different set of material conditions for education are workable, adoptable, and desirable conditions. The first reason for believing that our prototype may indeed metamorphose into a systemic change is its materiality, its substantiality. A second reason arises from a willingness to attend to the realities of power and interest.

Educational reformers often do not distinguish sufficiently between the social processes through which an innovation takes hold in practice and the social consequences of the innovation that has taken hold and begun to reshape everyday life. Pick a major technical innovation -- rail travel, the telephone, the automobile, air travel, television, or what-have-you. The social consequences of all these systems have been generally liberating and egalitarian: once established, they have brought options of movement, communication, information, and entertainment within the reach of very wide segments of the population, most of whom did not enjoy such options previously. These social consequences need to be distinguished from the social processes associated with the introduction of the technology into use, for these processes are invariably elitist and anti-egalitarian. The wealthy and powerful benefit first and their interests drive such processes of innovation. We do not say that is the way things ought to be; simply that is the way they are.

Educators have been poor effectors of innovation because they have been too eager to be on the right side of every issue, to champion reforms that will redress injustice, to help primarily the disadvantaged and those in need. Head Start provides an example. It was a good program serving a good purpose, but its base of support was too narrow for the program to have effective staying power and its locus of operation was too circumscribed for its benefits to endure through the whole of a child's education. Many educational reforms designed to redistribute the benefits of the existing system in favor of the least advantaged succumb to opposition because the political base strongly favoring them is too narrow. A redistributive change usually starts in an aura of good will, or highmindedness in the face of crisis, and groups opposed to it on grounds of ideology or self-interest will defer, letting the change begin unopposed, saving their resistance until the time for it is more opportune. At the first sign of hard times and dubious results, however, support for the change narrows and its natural opponents emerge, soon to prevail.

We aim, not to redistribute the benefits of the current system of education. We aim to substitute for it a new system, based on the cumulative curriculum, one that will improve the benefits for all. In doing that, we can mobilize a broad base of support, one with staying power in the face of opposition and constraint. In developing a new system, we can use the social dynamics of contemporary power to put in place innovations, the consequences of which will be liberating and egalitarian.

To change the system, educators must engage the interest and power of dominant elites, to work through their interests to the service of all. Our strategy for innovation is to mobilize the social dynamics of self-interested and powerful groups to institute changes, the social consequences of which will broaden access to excellent education and empowering cultural resources. An innovation that serves the advantaged and the disadvantaged alike is of greater relative worth to the disadvantaged, by virtue of the law of diminishing returns. For agenda. It is time to reassert it, robust and proud.

New York, New York, with your glorious complexity of peoples and needs, with your soaring concentration of power and wealth, with your capacity to attract talent and attention, with your gruff directness and abrasive realism -- with all your contradictions, New York, New York, its up to you to implement this strategy of change. The City's elites are the most diverse of the world; and they can mobilize great power, influence, and purpose in the service of all. They want their children well educated. They want reversed and redeemed the terrible collapse of hope and life chances for the truly disadvantaged, the consequences of which continuously threaten all. They want the daily level of skill and intelligence, coursing through the intricate worlds of their activities, to rise steadily, ever more competent and humane. These are not simple amenities; these are complex developments that neither their wealth can buy nor their will command. These wants are their stake in a substantially transformed, and a powerfully transforming, system of education. If they believe that a better system can be constructed, they will seek it for their own and use their influence to spread it far and wide simply because in New York, New York, unlike more sheltered, controlled communities, the elites have a substantial stake in the common good.

The elites of New York, New York, are, at once, the Rainbow Coalition, the American tradition of patrician reform, and the communication nexus of the world. They are Black and Hispanic and WASP and Jew and Catholic, new people and old, Italian and Irish, Korean and Chinese, German and Pole, Czech and Russian, Nordic and Greek, Indian and Pakistani, and on, and on -- peoples from everywhere, pulsing with energy and aspirations. They are resourceful -full of resources -- the resources of real estate, law, finance, communication, design, industry, science, engineering, fashion, art, film, and philanthropy. The existing educational system serves them poorly. They can change it. Simple prototypes can metamorphose into a new system of education, despite the opposition of inertia and ensconced interests, because the great energies of a great city will drive the process, from its inception, to its completion.