Models of Innovation: The History of an Idea

Inside Technology

Edited by Wiebe E. Bijker, W. Bernard Carlson, and Trevor Pinch

A list of the series appears at the back of the book.

Models of Innovation

The History of an Idea

Benoît Godin

The MIT Press

Cambridge, Massachusetts

London, England

© 2017 Massachusetts Institute of Technology

All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher.

This book was set in Stone Sans and Stone Serif by Toppan Best-set Premedia Limited. Printed and bound in the United States of America.

Library of Congress Cataloging-in-Publication Data

Names: Godin, Benoit, author.

Title: Models of Innovation : The History of an Idea / Benoît Godin.

Description: Cambridge, MA : MIT Press, [2017] | Series: Inside technology | Includes bibliographical references and index.

Identifiers: LCCN 2016031366 | ISBN 9780262035897 (hardcover : alk. paper)

eISBN 9780262338790

Subjects: LCSH: Technological innovations. | Models and modelmaking--History--20th century.

Classification: LCC T173.8 .G596 2017 | DDC 601--dc23 LC record available at https://lccn.loc.gov/2016031366

ePub Version 1.0

d_r0

The mind may, as it appears to me, divide science into three parts. The first comprises the most theoretical principles, and those more abstract notions whose application is either unknown or very remote. The second is composed of those general truths which still belong to pure theory, but lead, nevertheless, by a straight and short road to practical results. Methods of application and means of execution make up the third. Each of these different portions of science may be separately cultivated, although reason and experience show that none of them can prosper long, if it be absolutely cut off from the two others.

—Alexis de Tocqueville, Democracy in America (1840)

Development work is founded upon pure research done in the scientific department, which undertakes the necessary practical research on new products or processes as long as they are on the laboratory scale, and then transfers the work to special development departments which form an intermediate stage between the laboratory and the manufacturing department.

—Kenneth Mees, The Organization of Scientific Industrial Research (1920)

The principle of discovery first and utilization after is the oldest thing in man’s history.

—Willis Whitney, Science and Industry in the Coming Century (1934)

The theorist posits the basic concepts, the experimentalist tests reality, and the inventor converts the results to practical achievement.

—William Rupert Maclaurin, The Process of Technological Innovation (1950)

Table of Contents

Series page

Title page

Copyright page

Epigraph

Acknowledgments

Introduction

I Stage Models

1 The Invention-Diffusion Framework: Anthropologists and the Study of Cultural Change

2 The Stage Framework: Sociologists and the Study of Social Change

II Linear Models

3 The Research Cycle

4 The Linear Model of Innovation: A Theoretical Formulation

5 The Historical Construction of an Analytical Framework

6 The Demand-Pull Model

III System Models

7 The Research Triangle

8 A Managerial View

9 A National Perspective

Epilogue: Why Models of Innovation Are Models, or What Work Is Being Done in Calling Them Models

Conclusion

Appendixes

References

Index

Inside Technology

List of Tables

Table 1.1 The diffusion controversy: Two theses

Table 4.1 Classification of industries according to level of technological progressiveness

Table 5.1 Taxonomies of research

Table 6.1 Some early conferences on coupling and transfer

Table 6.2 Rothwell’s generations of models of innovation

Table 6.3 Three meanings of demand

Table 9.1 OECD major publications on science and technology policy, 1960–1992

Table 9.2 Transfers of funds among the four sectors as sources of R&D funds and as R&D performers, 1953 (in millions of dollars)

Table 9.3 OECD publications on national innovation systems

Table 10.1 Early models (called as such)

Table 10.2 Types of models

Table 10.3 Reviews of models

Table 10.4 Typologies of models

Table 10.5 Terms used interchangeably with model

List of Illustrations

Figure 2.1 Eugene Wilkening’s process of innovation. (From Wilkening 1953.)

Figure 5.1 Kenneth Mees’s diagram of advance of technology. (From Mees 1920.)

Figure 5.2 Clifford Furnas’s flow diagram from research to sales. (From Furnas 1948.)

Figure 6.1 Sumner Myers and Donald Marquis’s figure of the innovation process (1969). (From Gruber and Marquis 1969.)

Figure 6.2 Roy Rothwell’s diagram (1985). (From Rothwell and Zegveld 1985.)

Figure 6.3 Jacob Schmookler’s conceptual framework. (From Schmookler 1962.)

Figure 6.4 Chris Freeman’s representation of Schumpeter and Schmookler’s model. (From Freeman, Clark, and Soete 1982.)

Figure 6.5 Stephen Kline’s chain-linked model. (From Kline and Rosenberg 1986.)

Figure 8.1 Jack Morton’s innovation process according to phases of specialization. (From Morton 1971.)

Figure 8.2 Jack Morton’s model of the Bell System. (From Morton 1971.)

Figure 8.3 Jack Morton’s Bell System innovation process. (From Morton 1971.)

Figure 8.4 Ellis Mottur’s model of the processes of technological innovation. (From Mottur 1968.)

Figure 9.1 From OECD (1972a) and OECD (1978).

Figure 10.1 Everett Rogers’s paradigm of the innovation-decision process. (From Rogers and Shoemaker 1971.)

Figure 10.2 Kenneth Knight’s model of organizational search. (From Knight 1967.)

Figure 10.3 James Allen’s wheel, hub, and axle model. (From Allen 1967.)

Figure 10.4 Richard Nelson and Sydney Winter’s evolutionary model of firm behavior, according to Rod Coombs et al. (From Coombs, Saviotti, and Walsh 1987.)

Figure 10.5 Albert Rubenstein’s schematic diagram of the R&D process in firms. (From Rubinstein 1962.)

Figure 10.6 James Utterback and William Abernathy’s innovation and stages of development. (From Utterback and Abernathy 1975.)

Figure 10.7 Milton Coughenour’s model of the variables involved in farm practices. (From Coughenour 1960.)

Figure 10.8 Robert Mason and Albert Halter’s diagram of the innovation diffusion model. (From Mason and Halter 1968.)

Figure 10.9 Albert Rubenstein and John Ettlie’s model of the R&D/innovation process. (From Rubinstein and Ettlie 1979.)

Acknowledgments

The following chapters were published in the following journals. They are reproduced here, in a revised form, with permission of the publishers. Chapter 1: Benoît Godin (2014), Invention, Diffusion and Linear Models of Innovation, Journal of Innovation Economics & Management 15 (3): 11–37. Chapter 3: Benoît Godin (2011), The Linear Model of Innovation: Maurice Holland and the Research Cycle, Social Science Information 50 (3–4): 569–581. Chapter 4: Benoît Godin (2008), In the Shadow of Schumpeter: William Rupert Maclaurin and the Study of Technological Innovation, Minerva 46 (3): 343–360. Chapter 5: Benoît Godin (2006), The Linear Model of Innovation: The Historical Construction of an Analytical Framework, Science, Technology, and Human Values 31 (6): 639–667. Chapter 6: Benoît Godin and Joseph P. Lane (2013), Pushes and Pulls: The Hi(Story) of the Demand Pull Model of Innovation, Science, Technology and Human Values 38 (5): 621–654. Chapter 9: Benoît Godin (2009), National Innovation System: The System Approach in Historical Perspective, Science, Technology, and Human Values 34 (4): 476–501. Chapter 10: Benoît Godin (2015), Models of Innovation: Why Models of Innovation Are Models, or What Work Is Being Done in Calling Them Models? Social Studies of Science 45 (4): 570–596.

Introduction

Models abound in science, technology, and society (STS) studies, broadly defined, including studies of technological innovation (policy, management, economics) or science, technology, and innovation (STI) studies. They are continuously being invented and succeed one after the other—one author developing many versions of the same one over time. At the same time, models are regularly criticized. Such is the case with the most influential model in STS-STI, namely, the linear model of innovation. The model postulates that technological innovation—understood as the application of science and commercialization of inventions—begins with basic research, applied research, and then development. Commercialization follows.

Before the late 1960s and early 1970s, the word model rarely appeared in the literature on innovation. Theorists studied innovation in terms of a process composed of sequences and stages. Such a view was not called model, but rather a framework, paradigm, or conceptualization. Yet in a matter of a few years, linear sequence became linear model, and the alternative perspective, coupling process (research coupled to demand, as factors that explain innovation), became coupling model. Is model just a semantic convention? Or does it include more than framework or paradigm suggests?

According to the literature from philosophy, the natural and biological sciences, and economics, a model is a representation of reality in a simplified form.¹ Such is also the standard view in STS-STI: a model is a greatly simplified abstractions of the situation (Nelson and Winter 1982, 402). However, this is far from the whole truth. To others, a model is an example or emblematic paradigm. For example, in 1969, Bela Gold, professor of industrial economics at Case Western Reserve University, summarized the literature on technological innovation and called the assumptions shared in the field a guiding or synoptic model that reflects essentially similar, though unstated, conceptions of the basic system of relationships involved in [the field]. It may be useful to outline a structure of such implicit hypotheses so as to crystallize views which appear to be widely accepted and to highlight the contrasting hypotheses to be offered later (Gold 1969, 392). Taken together, these building blocks² yield a model which combines the appeals of simplicity, rationality and seeming relevance both to widespread interpretations of recent business experience and to common conceptions of the decision-making processes of management (393–394).³

This book offers a third view. Not only is a model a thing; it is also a concept. The term model started to be used in the studies of innovation around the 1960s, as in other social sciences. Certainly a model exists with or without the term. Yet the term serves a specific purpose: it gives social existence to a theoretical construct and contributes to making of it a visible theoretical construct in a field.

There exist two kinds of models in the literature: analytical and mathematical. The first type is sometimes accompanied with measurement but is more conceptual in nature. The latter type is grounded in arithmetical formulas and simulations. Mathematical models have a clear function: measuring and predicting. In contrast, conceptual models are, by definition, purely conceptual. This book is concerned with conceptual models, hereafter called models. Basically a model is a conceptualization or theorization put into a schema, graph, or diagram. Such models are usually schematized with boxes or concentric circles and arrows, as is the case for the linear model:

Calling such a conceptualization a model serves practical or pragmatic purposes, in addition to organizing knowledge. It highlights societal and policy uses and also serves rhetorical purposes. Because of its simplicity, a model allows the conceptualization to travel easily in the academic and public fields and between the two fields.

The book studies the emergence and the diffusion—and death—of the most influential models of innovation from the early twentieth century to the late 1980s. It has two main objectives. First, the book offers a history of models. Why did the particular models come into being? For what purpose? What use was made of them? How did models compete among themselves? The book looks at the models developed in the light of the contextual factors involved: civilizing and modernizing social practices and societies (part 1); funding research, because it is instrumental to socioeconomic progress (part 2); and supporting firms’ innovative activities (part 3). To these factors correspond different models (stage model, linear model, system model) and actors (respectively, anthropologists and sociologists; industrialists and economists; managers and policymakers) whose specific contribution is studied here through the schools and individuals responsible for the models.

The second objective of the book is to put to the test the narratives that the theorists of innovation have developed on models and the history of models. Scholars blindly like to trace back the origin of their ideas, when they do, to other scholars. In the case of models of innovation, scholars are silent on certain pioneers; they remain unaware of the intellectual source of their models; they invent mythic fathers. Here, I document a greater diversity of thinkers who contributed to models of innovation than emblematic authors, some forgotten, others ignored. I trace the genealogy of models back to anthropology and to industrialists, practitioners, and policy-makers in the first half of the twentieth century and their subsequent formalization in STS-STI.

I have always been fascinated by the ease with which scholars use the word model; by the multiplicity of models in the literature, every scholar developing his or her own model, hoping to become a leader in the field, thanks to his or her model; and by the quickness with which models become popular and turn into a fashion—and their names or labels into buzzwords—but also how fast models die. At the same time, I am always disappointed by the lack of reflexivity on models of innovation (models have a history that is too often forgotten); the mythic accounts on the origins of models; the lack of criticism or lack of serious consideration of criticisms of models, or, at the opposite end, extreme criticism of models in the name of one’s own model; and the equivocal interpretations of what a model is or refers to. This book offers a more complex history of models of innovation than any other in existence; consideration of a large number of contributors to the idea of models of innovation, larger than that suggested in current narratives; and a study of the discourses of scholars on models—the concept of model embodies a rhetoric that gives legitimacy to a scholar’s theoretical construct. This rhetoric is at the same time disciplinary, community based (scholars), and transdiscursive (between scholars and society).

Making Sense of Models

Models are central to the theorists of innovation and to policymakers. They organize knowledge and guide action. But models change continuously. Why? Is the simplification of reality that defines what a model is too simplified—or simplistic? Is it that a model is a resource to scholars in search of symbolic capital, changing according to fads, perspectives, and eras? Could it be that the assumptions of models—and modelers—are historically constructed and context dependent, making models ephemeral objects? This book suggests that nothing in models of innovation makes sense but in the light of history. We take for granted that models are scientific objects that explain as theories do. We may have forgotten that a model is a concept and, like other concepts, has history (meanings and uses change); that a model is a fluctuating, unstable, and contested construct; that its virtue lies beyond science alone. A model has to be looked at in relation to other models, not the world. A model is, first of all, a conceptualization, and competes with other conceptualizations to explain society.

The models studied in this book, in chronological order, can be summed up in two categories:

Process models

Stage model

Linear model

Demand model

System models

The first level of this typology—process and system models—owes its inception to Ronald Havelock. In 1967, Havelock, a prolific author, from the University of Michigan, on knowledge transfer or dissemination of knowledge wrote, There seem to be two ways to conceptualize [knowledge] utilization: One way is as a system and the other is as a process. The same is true of studies of innovation in which process means either a sequence of activities in time or a system of institutions and their relationships. In his review of models of change, Havelock defines system models as concerned with the flow structure of the use of knowledge and using the concepts of organization, group, person, agent, position, role, channel, and link. In contrast, process models study the way or the mechanism through which knowledge is used (what is going on at each of the exchange points or linkages in the flow structure), using concepts such as relationship, linkage, transfer, exchange, translation, diffusion, and communication (Havelock and Benne 1967, 50, 60). Such a categorization of models was commonplace in the 1960s. For example, a few years before Havelock, the sociologist Robert Chin offered a similar typology. The process or development model develops a time perspective which goes far beyond that of the more here and now analysis of a system model. A system model emphasizes primarily the details of how stability is achieved … [whereas] the developmental model assumes constant change and development, and growth and decay (Chin 1961, 211–212).

Briefly stated, a process model is one concerned with time, that is, the steps or stages involved in decision making of action leading to innovation (emergence, growth, and development of an innovation). A system model deals with the actors (individuals, organizations, and institutions) responsible for the innovation and studies the way the actors interact. Time and space, as the OECD put it, are the two fundamental frameworks to understand the innovation process (OECD 1978). A process model is historical, and a system model is social; one is developmental, the other functional. Yet the distinction between process model and system model is not as clear-cut as it might appear at first sight. Both types of models are models of a social process in a large sense. In creating technology, states Bernard Carlson, it is not enough to possess scientific or craft knowledge; one must also locate this knowledge in a social organization that can act upon it. … If one wishes to understand the innovation process, one must ask not only about the knowledge base but also about the organization within which that knowledge is developed and used (Carlson 1991, 7). Process here does not stress the time dimension, at least not as a central dimension in need of explanation, but the social one, that is, the social process of actors, activities, and environment that brings an invention to the market.

Among the process models, the literature on innovation puts stress on what is known as the linear model of innovation, a model said to be the first theoretical framework in the study of innovation (c. 1950), and now widely rejected. Yet before the linear model, there were precursors or other process models. One is the invention-diffusion framework from the early twentieth century. This framework, which was not called a model at the time, comes from anthropologists in the 1920 and 1930s and served to analyze changes in cultural traits among societies. Another early process model is the stage model from sociologists. From the 1940s onward, rural sociologists studied the diffusion of innovation as a sequential process composed of stages. These two models paved the way for the linear model of innovation.

Instead of (macro) stages of diffusion (invention, adoption), the linear model looks at the generation of innovation and postulates a series of (micro) steps or activities conducted in sequence (basic research, applied research, development, commercialization).⁴ The linear model was short-lived—about a decade. Or was it? From time to time, there have been theoretical rehabilitations of the linear model of innovation: the critique has gone too far (Balconi, Brusoni, and Oresnigo 2010); the model should be read backward, moving basic research to the end with feedbacks from end to beginning (Cowan 2005); the model has real empirical—social and spatial—bases (Henry, Massey, and Wield 1995). According to innovation theorists Roy Rothwell and Walter Zegveld, Despite the increasing acceptance of the interactive model of innovation [the substitute for the linear model], it nevertheless remains clear that many governments—and indeed many industrial companies [I would add scholars too]—continue to adhere, at least implicitly, to the technology-push model (the linear model of innovation), namely, the belief that more R&D does indeed result in more innovation (Rothwell and Zegveld 1985, 50).

Be that as it may, criticism of the linear model gave rise to the demand-pull model (c. 1965), which places at the origin of the process of innovation social needs or market demand, depending on the discipline, rather than research or research and development (R&D). This model was as short-lived as the linear model. First, the field (STS-STI) saw the emergence of studies documenting that the innovation process involves a diversity of factors, of which research and demand are but two. Moreover, there is feedback between the factors responsible for innovation, not a fixed sequence. These ideas gave rise to interactive views or models of innovation: a linear model with feedback.

Second, a new kind of model made its appearance: the system model. As mentioned above, what distinguishes system models from process models is that the latter are dynamic models or, as some call them, a natural history of innovation (Havelock 1969, 10.81). The identification of stages acts as a theoretical framework for studying the process of innovation over time, from the generation of an invention to its adoption or diffusion. In contrast, a system model stresses the organizations or institutions and their relationships. Certainly innovation as a system is studied as a process, but not in the sense of a sequence of stages or activities. Scholars look at the constituents or parts or subsystems and the way the constituents interact with each other in order to serve a common output or purpose: the production and adoption of (technological) innovation.

Structure of the Book

The book is organized according to the two types of models noted: process and system. Parts 1 and 2 are devoted to the history of process models, respectively stage models and linear models, and part 3 to system models. To be sure, there are other ways, many other ways, to categorize models. Yet the typology I offer, or suggest, is a basic one for historical purposes. First, it corresponds to and thus allows the critical analysis of the narratives of scholars in the field: generations of models would have evolved from process models to system models. Second, it enables me to fill in the holes in current narratives: a large number of models and inventors are absent from current narratives. I include here forgotten models or, rather, precursors, like the research cycle from industrialists. I also unearth forgotten authors, like Maurice Holland and Rupert Maclaurin, as the precursor and the inventor of the linear model, respectively, and Jack Morton for his early system model. I also bring back to life models not considered in current narratives, or rather typologies, of models from STS-STI, particularly from narratives in the management, policy, and economics literature: the invention-diffusion framework from anthropologists, the stage model of sociologists, and the system approach from the Organization for Economic Cooperation and Development. All in all, the history documented here runs counter to the narratives that fill the literature. A model rarely comes from a single individual, however important this person is as a scholar or public figure. Models have history. One of the objectives of this book is to rethink and debunk the historical narratives created by today’s theorists of innovation.

Chapter 1 documents the emergence of one of the first theoretical constructs or frameworks in the study of innovation, from the first half of the twentieth century: the invention-diffusion framework. Certainly, at that time there was no use of the word model and few uses of the term innovation. Yet the study of culture in terms of either invention or diffusion gave rise to the sequence invention → diffusion that remained influential in later models of innovation. In fact, there exist two sequential or linear models of innovation in the literature. One is the linear model of innovation as such, discussed in part 2. The other model, of which the linear model of innovation is one part or stage, is that of innovation as a process of invention followed by diffusion. This model, or rather the theory or framework on which it is based, comes from anthropology and was invented as a solution to a controversy on how invention versus diffusion explains civilization and culture change.

Similar to the approach of anthropologists, every later study of innovation looks at innovation as a process. But what is a process? Chapter 2 studies the sociologists on the diffusion of innovation as the pioneers of the idea. Beginning in the 1920s, sociologists studied innovation as a process over time. To this end, they imagined sequences and stages through which innovation diffuses, thus contributing to the modernization of social practices and societies. In a matter of a few decades, the idea of innovation as a stage process was present everywhere in literature on innovation, from sociology to management and economics. It is precisely such a dynamic process that later models of innovation modeled.

One influential framework developed for understanding technological innovation (rather than innovation, as anthropologist and sociologists do) and its relation to the economy is what is called today the linear model of innovation. The model postulates that technological innovation is a process that starts with basic research, followed by applied research and development, and ends with commercialization. The precise source of the linear model remains nebulous. Several authors who used, improved, or criticized the model in the previous fifty years rarely acknowledged or cited any original source. The model is usually taken for granted. According to others, it comes directly from Vannevar Bush’s Science: The Endless Frontier (1945).

The next two chapters look at particular authors as pioneers of the idea of a linear model of innovation. In 1928, Maurice Holland, director of the Engineering and Industrial Research Division at the US National Research Council, produced a paper on what he called the research cycle. He portrayed the development of modern industries as a series of sequential stages from basic research to commercialization of technological inventions. Chapter 3 documents the source or context of the idea of the research cycle, the arguments on which it relies, and the end to which it was put: persuading more industrialists to build research laboratories in order to accelerate the development of industries. The chapter suggests that Holland turned a frequently heard but poorly formalized argument into a theory or framework, paving the way for what came to be called the linear model of innovation.

Chapter 4 turns to an economic historian, Rupert Maclaurin from MIT, a student of Joseph Schumpeter. Schumpeter is a key figure, even a seminal one, on technological innovation. Most economists and STS-STI scholars who study technological innovation refer to Schumpeter and his pioneering role in introducing innovation into economic studies. However, despite having brought forth the concept of innovation in economic theory, Schumpeter provided few, if any, analyses of the process of innovation itself. This is important to keep in mind. Schumpeter’s theory of economic change is not discussed in this book. This chapter suggests that the origin of systematic studies of the process of technological innovation owes its existence to Maclaurin. In the 1940s and 1950s, Maclaurin developed Schumpeter’s ideas further, analyzing technological innovation as a process composed of several stages, and he proposed a theory of technological innovation that would later be called the linear model of innovation. Maclaurin’s purpose was no more the study of civilization and modernization in the abstract but explaining the way or steps that research takes or should take to have effects on socioeconomic progress. This remained the main purpose of scholars’ models of innovation for the next two decades.

Chapter 5 extends the analysis to industrialists, management and official statisticians, and economists. The linear model of innovation is the work of many individuals, conducted over several decades. It would be nonsense to attribute it to one or two individuals only (Holland and Maclaurin). A diversity of actors and scholars introduced their respective point of view or disciplinary matrix into the discourse of innovation that led to the fine-tuning of the linear framework over time. The chapter also argues that statistics is one of the main reasons explaining why the linear model of innovation is still alive despite criticism, alternatives, and having been proclaimed dead.

The models studied in this book up to and including Maclaurin’s do not carry the name model as such. The term was applied in retrospect. In the 1960 and 1970s, the term model was in vogue in every discipline (Heyck 2014), and the theorists of innovation started to apply it to their own construct and that of others, even if the others did not use the term. One catalyst in the use of the term in STS-STI was the critique of the newly named linear model of innovation. Beginning in the 1960s, people from different horizons started looking at technological innovation from a demand rather than a supply perspective. The view was that technological innovation is stimulated by market demand rather than by scientific discoveries. Few traces of the demand-pull model remain in the literature today. Chapter 6 looks at what happened to the demand-pull model from a historical perspective at three points in time: its birth, crystallization, and death. The chapter suggests that the idea of demand as a factor explaining technological innovation emerged in the 1960s, was formalized into models in the 1970 and 1980s, and was then integrated into holistic models. From then on, the demand-pull model disappeared from the literature, existing only as an object of the past, like the linear model of innovation.

In their place, system models made their appearance.⁵ In the last few decades, the meaning of process has changed. The term refers less to time than to a set of organizations and institutions involved in innovation, a system whose center is the innovative firm. Criticism of the linear model of innovation is responsible for this. As a result, a holistic view of innovation developed. Research is not the central factor explaining innovation. Many other activities (other than science) and actors (other than scientists) are necessary. It is common today to view science and technology as an innovation system composed of institutional sectors in relation to each other. Where did this approach come from? Chapter 7 studies the emergence of industrial research as a key factor in the development of a holistic approach to technological innovation. From the late nineteenth and early twentieth centuries, universities were no longer alone in conducting research; there was a research triangle, as some called it in the 1940s (today, some say a triple helix, an old metaphor),⁶ a more complex system composed of universities, industries, and governments. The chapter analyzes the early industrial discourses held in the name of a holistic approach to research, or scientific whole, following World War I. To industry, a holistic approach would put industrial research on the national research map, contributing to public recognition of the phenomenon. This would help make the case for universities’ contributing to industries’ needs, and industries benefiting from the government’s funding efforts.

The holistic discourse on research was further developed when scholars started looking at technological innovation as a system. Chapter 8 examines one of the first system models of innovation, from managers. It studies the contribution that engineer Jack Morton, a manager at Bell Laboratories, made to models of technological innovation in the 1960s and 1970s, a system model of the total process of innovation (total was a key term in the literature of the time, as whole was before). Innovation encompasses a large variety of people and activities acting together toward a common goal. A system view is essential, claimed Morton, to understand and manage the total process of innovation. Morton’s views of technological innovation as a system were shared by many at the time. They gave rise to a new kind of models of innovation that culminated in the idea of a national innovation system.

The national innovation system approach suggests that the research system’s ultimate goal is technological innovation, and that the system is part of a larger system composed of institutional sectors like government, university, and