Economics and Politics in the Robotic Age The Future of Human Society 1527546071, 9781527546073

Citation preview

Economics and Politics in the Robotic Age

Economics and Politics in the Robotic Age: The Future of Human Society By

Qing-Ping Ma

Economics and Politics in the Robotic Age: The Future of Human Society By Qing-Ping Ma This book first published 2023 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2023 by Qing-Ping Ma All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-5275-4607-1 ISBN (13): 978-1-5275-4607-3

To my parents, Ma Yao and Li Pei-Lan, with gratitude.

TABLE OF CONTENTS

List of Figures........................................................................................... xv List of Tables ........................................................................................... xvi Preface .................................................................................................... xvii Part I. A Brief History of Production Revolutions Chapter 1 .................................................................................................... 2 The Manual Age: From the Emergence of Hominins to the British Agricultural Revolution 1. Basic Economic Entity and Periodization of Economic History 2. Hunter-Gatherers in the Paleolithic Age 2.1. The Lower Paleolithic Age 2.2. The Middle Paleolithic Age 2.3. The Upper Paleolithic Age 3. The Mesolithic Age 3.1. Mesolithic 1: Domestication of Dogs and Collection of Wild Cereal 3.2. Mesolithic 2: The Early Signs of Agriculture 3.3. Mesolithic Pottery Made by Hunter-Gatherers 4. The Neolithic Revolution: The Emergence of Agriculture 4.1. Farming and Domestication of Animals 4.2. Neolithic Tools, Pottery, and Art 4.3. Accommodations, Urbanization, and Governance4.4. 5. The Bronze Age and the Division of Labor 5.1. The Transitional Chalcolithic or Copper Age 5.2. Economy and Technological Progress 5.3. Writing, Literature, Philosophy, Science, and Art 5.4. Governance and Empire Building 6. The Iron Age and the Beginning of Ancient History 6.1. The Spread of Iron-Making Technology 6.2. Phoenician Alphabet and Written Languages 6.3. The Classical Antiquity

Table of Contents

viii

7.

From the Postclassical Era to the British Agricultural Revolution 7.1. The Postclassical Era 7.2. The Early Modern Era 7.3. The British Agricultural Revolution 8. Summary References

Chapter 2 .................................................................................................. 64 The Machine Age: From the Industrial Revolution to the 1970s Stagflation 1. The Causes of the Industrial Revolution 1.1. The Commercial Revolution 1.2. The British Textile Industry and Trade Before the Industrial Revolution 1.3. Why did the Industrial Revolution happen in Britain and Europe? 2. The Industrial Revolution 2.1. Machines and Power Sources 2.2. Innovations for Cheaper or New Materials 2.3. Innovations in Transportation 2.4. Social Changes and Reforms 3. The Second Industrial Revolution 3.1. Energy and Power Sources 3.2. Innovations in materials 3.3. Communications and Transportation 3.4. Industrialization of Agriculture 3.5. Innovations in Management 4. The Digital Revolution and the Third or Fourth Industrial Revolution 5. Summary References Chapter 3 ................................................................................................ 123 The Dawn of the Robotic Age: From Centrifugal Governors to the Rise of Artificial Intelligence and Robots 1. Automation of Physical Processes 1.1. Automation in Practice 1.2. Control Theory and Types of Control Systems 1.3. Computers in Automation of Physical Processes 2. Automation of Mental Processes 2.1. Computers 2.2. Communications and Information Technology 2.3. AI at Basic Levels

Economics and Politics in the Robotic Age: The Future of Human Society

ix

3.

Automation of Complete Job Systems 3.1. Approaches for Automation of Complete Non-Physical Job Systems 3.2. Automation of Complete Physical Job Systems 3.3. Toward Robots That Perform Complete Job Systems 4. Automation, robots, and AI in the Modern Economy 4.1. Industrial Robots and Automation 4.2. Service Robots and Automated Ordering and Checking-Out Systems 4.3. Agricultural Robots and Automation 4.4. Educational Robots and AI Systems 4.5. Healthcare Robots and Personal Helpers 4.6. Robots for Entertainment and Toys 4.7. Culture, Sports, and Entertainment 4.8. Research Robots and Automation 4.9. Military Robots and AI Systems 5. What is the Future of AI and Robots? 5.1. Creativity 5.2. Social Intelligence 5.3. General Intelligence 6. Summary References Part II. Consumption and Production in the Robotic Age Chapter 4 ................................................................................................ 204 Consumption, Satisfaction, and Life’s Purpose: A New Framework of Consumption Theory 1. Objects, Relations, and Activities That Cause Satisfaction 2. Ways to Satisfy Human Desires 3. The Process and Determinants of Consumption 3.1. The Process of Consumption 3.2. Determinants of Consumption 3.3. Consumer Rationality and Maximum Sensible Consumption 4. Modeling Constraints of Consumption 4.1. Monetary Costs of Consumption 4.2. Time Costs of Consumption 4.3. Space Constraints of Consumption 4.4. Physiological constraints of consumption

Table of Contents

x

5.

A New Hierarchical Model of Human Needs 5.1. Physiological Needs 5.2. Psychological Needs 5.3. Sociological Needs 5.4. Spiritual, Self-Actualization, and Self-Transcendence Needs 5.5. Methodological Needs 6. Utility Function and the Utility Maximization Problem 7. Implications of the Present Framework 7.1. The Individual’s Capacity to Consume 7.2. The Slowdown in Productivity Growth Since the 1970s 7.3. The Role of Information Technology and Asset Price 7.4. The Potential Future Economic Growth and Employment 7.5. Social Welfare and Politics 8. Summary References

Chapter 5 ................................................................................................ 252 Production in the Robotic Age 1. Production Factors in the Robotic Age 1.1. Capital 1.2. Labor 1.3. Human Capital: A Critique 1.4. Production Technology and Productivity 2. What is to be Produced: A Critique of the Mainstream View 2.1. Finite Physical Goods for Non-Vanity Needs 2.2. Finite Services for Non-Vanity Needs 2.3. Finite Intangible and Information Goods for Non-Vanity Needs 2.4. AI Systems and Robots 2.5. Goods and Services for Vanity Needs 3. Who Will Produce? 3.1. Horizontal and Vertical Division of Labor 3.2. Replacing First-Line Human Workers with Robots 3.3. Supervisors and Robots 3.4. Senior Managers, Homeostasis of the Internal Environment, and Robots 3.5. Measurement of Labor Productivity and Management Productivity 3.6. The Evolving Relationship between Humans and Robots

Economics and Politics in the Robotic Age: The Future of Human Society

xi

4.

How Will Goods Be Produced? 4.1. Agriculture, Forestry, Husbandry, and Fishery 4.2. Mining 4.3. Manufacturing 4.4. Energy and Water 4.5. Building 4.6. Commerce 4.7. Transport, Post, and Logistics 4.8. Accommodation and Catering 4.9. Information and Communication 4.10. Finance 4.11. Real Estate 4.12. Renting and Commercial Services 4.13. Scientific Research and Technological Services 4.14. Residential Services and Repair Services 4.15. Public Utilities and Environment Management 4.16. Education 4.17. Health and Social Work 4.18. Culture, Sports, and Entertainment 4.19. Public Administration and Social Securities 5. The Singularity 6. Summary References

Chapter 6 ................................................................................................ 307 Human Resources, Natural Resources, and Pollution 1. Human Resources in the Robotic Age 1.1. Demand for Human Talents and Technological Progress 1.2. The Curse of Success: How More Scholars Can Hinder Scientific Progress 1.3. Human Talents in the Robotic Age 1.4. Human Beings as Members of the Community 1.5. Humans as Legislators 2. Natural Resources in the Robotic Age 2.1. Natural Resources for Producing Foods 2.2. Water 2.3. Energy 2.4. Minerals and Building Materials 3. Pollution Issues in the Robotic Age 3.1. Sources of Pollution 3.2. Waste Processing 3.3. Pollution Control

Table of Contents

xii

4.

Attention Time as a Resource 4.1. Firms Posting Advertisements 4.2. Firms Displaying Goods 4.3. Live Commerce, Internet Celebrity Commerce, and SelfMedia 5. Job as a Resource in the Robotic Age 5.1. The Disappearing Human Jobs 5.2. Work as a Scarce Resource 6. Summary References

Part III. Wealth Distribution, Politics, and Preparation for the Future Chapter 7 ................................................................................................ 348 Income and Wealth Distribution in the Robotic Age 1. Motivation, Productivity, and Equality 2. Technology, Employment, and Wealth Distribution 2.1. General Employment Enhancing Technology 2.2. Employment Reducing Technology 2.3. Return of Capital with Decreasing Labor Demand 2.4. Productivity Growth and Social Polarization 3. Globalization and Wealth Distribution 3.1. The Law of Comparative Advantage and its Limitations 3.2. Rising Protectionism and its Causes 3.3. Globalization in the Robotic Age 4. Intellectual Properties and Wealth Distribution 4.1. Intellectual and Intangible Properties as Revenue-Earning Assets 4.2. Intellectual and Intangible Properties and Wealth Distribution 4.3. Intellectual Properties, Information Technology, and Productivity 5. Immigration and Wealth Distribution 5.1. Immigration and Unemployment 5.2. Immigration and Social Security 5.3. Failed States, Refugees, and the International Community 6. Guaranteed Income and Social Security System 6.1. Mass Unemployment and Minimum Income Support 6.2. Politics, Employment, and Social Security 6.3. Entrepreneurs and Capital Owners in the Robotic Age 7. Summary References

Economics and Politics in the Robotic Age: The Future of Human Society

xiii

Chapter 8 ................................................................................................ 395 Social Organization 1. Individuals, Families, and Households in the Robotic Age 1.1. Individuals in the Robotic Age 1.2. Family Relations 1.3. The Impact of Science and Technology on Family 1.4. Human-Robot Relationship 2. AI Systems and Robots as Administrators in Local Communities 2.1. AI Facilitated Local Direct-Democracy 2.2. Robots and AI Systems as Community Administrators 2.3. AI or Machine Morality 2.4. Ethics in Using Robots and in Human-Robot Interaction 3. Local Governments Run by Robots and AI Systems 3.1. The transition from Indirect Democracy to Direct Democracy 3.2. Robot-Run Local Government with Human Supervision 3.3. The Role of Local Governments in the Robotic Age 4. Social and Professional Organizations 5. Nation-State 6. International Community 6.1. Economics and International Relations 6.2. Communication Technology and Problem-Solving 6.3. Unbalanced Development in the International Community 7. Strong AI and Super-Intelligent AI Society 7.1. Strong AI and Mind 7.2. Will strong AI be a Threat? 7.3. Strong AI Society 8. Summary References Chapter 9 ................................................................................................ 427 Preparation for the Future 1. Individuals: What Should We Learn, and What Could We Do? 1.1. Accepting That Robots and AI Systems Will Take Most Human Jobs 1.2. Working to Welcome the Robotic Age 1.3. Ordinary People Participate in Political Decisions 2. Schools and Universities 2.1. The Origin of Universities 2.2. Preparation by Universities for the Robotic Age 2.3. Cultivating Responsible Citizens and Training in Basic Skills

Table of Contents

xiv

3.

4.

5.

6.

Firms: Self-Transformation into the Robotic Age 3.1. Embracing AI and Robots in Production, Service, and Management 3.2. Reconfiguring for the Changing Landscapes of Production 3.3. Owner-Robots Collaboration Professional Organizations 4.1. AI Professional Advisors and Human Professionals 4.2. Professional Sports and Entertainment 4.3. Robotic Researchers and Human Researchers Governments 5.1. Industrial Policy 5.2. Implementing Guaranteed Basic Income and Cultivating Responsible Citizens 5.3. Automation of Government Summary

Appendix A. List of Abbreviations ........................................................ 460 Index ....................................................................................................... 463

LIST OF FIGURES

1-1 An economy as the externalization of a monkey’s “production” mechanism 1-2 A typical Oldowan simple chopping tool 1-3 A typical Acheulean handaxe 1-4 A typical Mousterian point 1-5 A typical El Khiam point microlith 1-6 A potter bowl with a pig image from the Hemudu site 2-1 A model of the spinning jenny 2-2 A model of the Arkwright water frame 2-3 A model of the self-acting mule 2-4 A diagram of the Newcomen steam engine 2-5 A diagram of a 1784 steam engine designed by Boulton and Watt 2-6 A diagram of a reverberatory furnace 4-1 Maslow’s hierarchy of human needs is portrayed as a pyramid 4-2 A consumer’s time budget constraint 4-3 Productivity growth in developed countries during different periods 4-4 An illustration of why productivity growth has slowed down since the 1970s 5-1 Production factors in the Manual and Machine Ages 5-2 Autonomous factories and the support center (human) 5-3 Corporate management line charts at different stages of the Robotic Age 5-4 Supply of 3-D printer consumables 5-5 Application of ICT, AI, and robots at the early and mature stages of the Robotic Age 7-1 International traders only import goods cheaper than domestic ones regardless of comparative advantage 9-1 The management line and feedback routes in the Robotic Age

LIST OF TABLES

1-1 Periodization of economic history 4-1 Types of goods in a broad sense 4-2 Types of private common goods 4-3 Types of novel products 5-1 Types of capital in different ages 5-2 Classification of individual production activity 5-3 Types of information goods 6-1 Ability required in different periods 6-2 Different intelligence levels of robots 6-3 Natural resource solutions in the Robotic Age 6-4 Classification of pollution

PREFACE

The rapid development of artificial intelligence (AI) and robotics is one of the most critical phenomena in the twenty-first century. For the pessimists, this will cause mass technological unemployment, and even humanity’s future survival might be at stake if AI and robots are allowed to become more intelligent than humans. For the optimists, these new technologies will create more jobs than those they will destroy; as human history has repeatedly demonstrated, there is no need for scaremongering. Economists, AI researchers, business scholars, and popular science writers of both pessimist and optimist colors have published numerous books on the impacts of future AI and robots. Despite their opposing views on the impact of AI and robots on human employment, their policy recommendations are shockingly or understandably similar, that is, to improve education and retraining. The pessimists are particularly worried about the prospect that human workers have no jobs to do. Hence, their more detailed prescriptions usually contain how to let workers have jobs to avoid the three evils of boredom, vice, and poverty. The plan to write a systemic analysis of human society’s future direction arose from my experience of using robots in drug discovery and my view of human consumption from a physiological perspective. While measuring neuropeptides with radioimmunoassay during my Ph.D. in Neurophysiology/Neuroscience study, although radioactivity counters were automatic, manually handling test tubes and reactions allowed me to work with only one or two neuropeptides in dozens of test tubes. I could screen hundreds or even thousands of compounds a day with robots doing experiments. Looking back at the production history of humanity, we can find that it is a history of continually reducing human efforts or input in each output unit. Therefore, AI and robots taking over all human jobs should be the fulfillment of a dream long held by humanity. It should be celebrated rather than feared. Viewed as such, the history of production began with the Manual Age, in which human muscles were the main power source, while tools were invented to enhance the efficiency of human muscle power. According to the materials used to make the tools, the early days of humanity were divided into the Stone Age, Bronze Age, and Iron Age. Labor productivity

xviii

Preface

increased more and more with better tools so that farmers could feed themselves and artisans, teachers, thinkers, administrators, and rulers. The further improvement of manual tools and a better understanding of natural forces brought humanity to the Machine Age with the Industrial Revolution. During the Machine Age, human physical strength is no longer essential in production as non-biological power sources drive machines. However, human workers still control production processes by contributing the intelligence needed by the production process. The further reduction of human input naturally is to replace human intelligence required for the production process with machine intelligence, which is the task of robots and AI. Intelligent machines will take control of most production processes in the coming Robotic Age. If robots and AI take all human jobs in production, the human input in each output unit will reduce to zero. Orthodox economics thinks that humans desire infinite consumption with the axiom of non-satiation. As any medically qualified professional can testify, unlimited consumption is impossible because human physiology does not allow us to do so. As a neuroscientist, I understand that people consume materials to satisfy their senses and minds, which are constrained by their physiological capacity. Therefore, the quantity consumed sensibly by an individual is finite, which implies that everyone’s non-vanity demands can be satisfied when productivity reaches a certain level. Inequality in outcomes has been the most significant driver of efficiency and productivity. However, civilization can promote equality at the expense of efficiency when AI and robots have dramatically increased productivity. There is no need to worry about what workers will do in the Robotic Age. Everyone could live like intellectuals on government stipends in ancient China or nobilities in feudal Europe with a much higher standard of living. The prime objective of this book is to examine from a broad perspective how the economy will operate in the robotic age. As a fundamental economic change will inevitably affect politics, this book touches on political operations in the Robotic Age. The word robot used in this book means robots with artificial intelligence. It can also indicate AI systems (software) that do not have a physical existence, although I tend to refer to robots and AI systems in this book. While intelligent robots perform physical jobs, the AI systems that replace professionals for non-physical tasks can be viewed as virtual robots. The Robotic Age is an age of physical and virtual robots performing productive roles. Humans are still masters in the Robotic Age, just like they are masters in the Machine Age.

Economics and Politics in the Robotic Age: The Future of Human Society

xix

To understand how the economy will operate in the Robotic Age, it is helpful first to trace the production evolution in human history and try to find clues from the development in the Manual Age and the Machine Age as well as the progress in the current germination stage of the Robotic Age. Examining historical and present evidence forms the first part of this book. Chapter 1 examines the Manual Age, Chapter 2 the Machine Age, and Chapter 3 the development of automation, information and communications technology (ICT), AI, and robots. I call the present time the dawn of the Robotic Age. Part 2 investigates the economy in the Robotic Age, which will focus on consumption and production. The widespread application of robots will fundamentally change the production process and consequently affect consumption as well as utilization of resources. Chapter 4 provides a framework for consumption theory, which includes the physiological, time, and space constraints for consumer choice in addition to the monetary budget constraint. Chapter 5 examines how the economy operates in the Robotic Age. Chapter 6 looks into how resources, especially human resources, are utilized. Part 3 examines wealth distribution and political operations in the Robotic Age and discusses how individuals, firms, governments, and the international community should prepare for the coming Robotic Age. Chapter 7 focuses on income and wealth distribution. Chapter 8 investigates social organization in the Robotic Age, looking at how individuals, local communities, governments, organizations, countries, and international communities behave differently from in the Machine Age. Chapter 9 discusses how individuals, universities, firms, professional organizations, and governments should prepare for the Robotic Age. I hope this more detailed and broader-based approach can shed more light on the future of human society, which is currently at the dawn of the Robotic Age. I also hope readers will find this book helpful in understanding the impact of AI and robotics on the economy and society. I want to thank my family, friends, and colleagues for their support and help. Qing-Ping Ma

PART I A BRIEF HISTORY OF PRODUCTION REVOLUTIONS

CHAPTER 1 THE MANUAL AGE: FROM THE EMERGENCE OF HOMININS TO THE BRITISH AGRICULTURAL REVOLUTION

To understand the future of human society, a useful starting point is to look back at how humanity has arrived at where we are now. The primate ancestor of modern humans (Homo sapiens) can be traced to the primitive catarrhines, the common ancestors of old-world monkeys and apes. From catarrhines arose the hominoids (apes, members of the superfamily Hominoidea) over 20 to 23 million years ago (Kumar and Hedges 1998; Hedges and Kumar 2003; Pilbeam and Young 2004). Around 16.8 million years ago, the lesser apes, such as the gibbons, separated from the great apes (orangutans, gorillas, chimpanzees, and humans, members of the family Hominidae) (Carbone et al. 2014). The hominines (gorillas, chimpanzees, and humans, members of the subfamily Homininae) parted with orangutans around 12 to 14 million years ago. The gorillas diverged from the hominins (the chimpanzees and humans, members of the tribe Hominini) around eight million years ago, and the chimpanzees split off from the hominians (humans) and Australopithecus around 6.3 million years ago (Hill and Ward 1988; Pilbeam and Young 2004; Cote 2004; Fuss et al. 2017; Böhme et al. 2017). The hominians, the modern humans, and their close relatives are members of the subtribe Hominina. The Australopithecus evolved in eastern Africa around four million years ago. The genus Homo was thought to be derived from Australopithecus about three million years ago, and its earliest known member, Homo habilis, appeared 2.8 million years ago (DiMaggio et al. 2015; Villmoare et al. 2015). Human society or, more appropriately, human communities emerged with the genus Homo. In this chapter, we will examine socio-economic development in the long period from the emergence of humanity to the Industrial Revolution, which we call the Manual Age because production depended mainly on people’s manual power.

The Manual Age

3

1. Basic Economic Entity and Periodization of Economic History An individual ancestry primate such as a catarrhine has already demonstrated some essential elements of an entity in a modern economy. A monkey is both a producer and a consumer. Its four limbs provide tools for production and mobility; its nervous system provides the process control, communication, and motivation; its muscles are the power source for the production process; its digestive and circulatory systems transport the materials and energy to where they are needed; its skin and fur provide some essential protection against adverse factors in the environment (Fig.1). The consumption (eating) happens almost simultaneously with the production (picking up fruits and other foods). They also play to have fun, want companions and might enjoy congenial environments that comfort their senses. These activities and conditions bring them satisfaction or utility similar to humans. The later economic progress of human society is just externalizations, expansions, divisions, scale-ups, specializations, separations, and alienations of these processes, as well as additions and scale-ups of raw materials and products.

Fig.1 An economy as the externalization of a monkey’s “production” mechanism.

Externalization is the partial or whole replacement of an internal or proximal process by an external or distant process. For example, compared with a monkey’s hand-to-mouth production-consumption mode, the goods

4

Chapter 1

we consume are generally produced far away without our participation. Expansion is an increase in the range or variety. The division is to split a process into subsegments that can be performed by different people or at other times. Scale-up is to conduct the same process on a large scale. Specialization narrows what a person has to do in a complex process down to a specific task. Separation transfers some rights from their owners to somebody else. Alienation is the disconnection of products and processes from their creators, which enables them to acquire a life of their own. We will find numerous incidents of these activities when we examine the history of humanity. The first sign of progress was using and making tools as extensions of human hands and limbs. When primitive humans began to make stone tools, they entered the Stone Age. The development of civilization is usually divided into three stages according to the materials used to make tools: The Stone Age (2.6 million to 6,500–4,000 years before present, BP), the Bronze Age (3300 BCE to 1200 BCE in the Near East), and the Iron Age (starting at around 1200 BCE), with some transitional periods between two adjacent ages. Bronze and iron enabled humanity to make better tools and massively increased labor productivity compared with the Stone Age. The Stone Age can be further divided into the Paleolithic Age, the Mesolithic Age, and the Neolithic Age. The Paleolithic Age was preagricultural. The beginning of agriculture coincided with that of the Neolithic Age. Agriculture provided the foundation for more complex social structures than hunter-gatherer communities. The Mesolithic Age was the transition period between the Paleolithic and Neolithic Ages (Bellwood 2004; Ammerman and Cavalli-Sforza 1971). The technological progress in the Stone Age was extremely slow, and the same technology was often invented at multiple sites because of the lack of effective communication and transportation means. We will look into these periods in the following sections. Although iron and steel continue to be the primary material for making tools up to modern times, the Iron Age was thought to end with the beginning of recorded history because the three-age division is used to describe the prehistory and history of humanity. Since the production process was largely dependent on human physical strength until the First Industrial Revolution in the eighteenth century, the long period from the emergence of humankind to the Industrial Revolution was a Manual Age. The Industrial Revolution brought humanity to the current Machine Age, where production no longer depends on human manual power.

The Manual Age

5

After the replacement of human manual power by machines, the next epoch will be the Robotic Age, in which intelligent robots replace both human manual and mental abilities. The digital revolution or information revolution that began during the period between 1950 and 1970 is a transitional period between the Machine Age and the coming Robotic Age. In the Robotic Age, human mental power is substituted by artificial intelligence (AI) in the production process, so labor will no longer be a critical factor for economic growth. We propose that economic history be periodized according to how human contributions to the production process are supplemented and substituted by tools and machines (Table 1). We divide economic development into the Manual, Machine, and Robotic Ages. Table 1 Periodization of Economic History Period Paleolithic Age Mesolithic Age Neolithic Age Chalcolithic Age Bronze Age Manual Age

Subperiod Lower Paleolithic Middle Paleolithic Upper Paleolithic

Copper Age Early Bronze Age Middle Bronze Age Late Bronze Age

Iron Age Archaic Period Classical Antiquity Post-classical Era Modern Era

Machine Age

Contemporary History

Classical Greece to Roman Empire Late Antiquity Middle Ages Early Modern Period Late Modern Period

Time 2.6 -0.3 MY BP 300-50 KY BP 50,000-12,000 BP 10,200-3,300 BCE 4,500-3,200 BCE 3,300-2,100 BCE 2,100-1,550 BCE 1,550-1,200 BCE 1,200-550 BCE* 8th–6th centuries BCE 5th century BCE– 5th century CE 250–750 CE 500–1500 CE 1500–1800 CE 1800–1945 CE 1945–Present

Robotic Age * In the Neareast. BP: before present. KY: thousand years. MY: million years.

Chapter 1

6

2. The Hunter-Gatherers in the Paleolithic Age The typical Paleolithic society was a hunter-gatherer economy (Stavrianos 1991). The oldest stone tools found before 2015 were those from the excavation sites in Gona, Ethiopia, which could be firmly dated to 2.6 million years ago. The Paleolithic Age is considered to have started 2.6 million years ago. It could be further divided into Lower (2.6 million– 300,000 BP), Middle (300,000–50,000 BP), and Upper (50,000–10,000 BP) Paleolithic ages. Recently, more primitive stone tools dated over 3.3 million years ago have been excavated in the Lomekwi 3 site in Kenya (Harmand et al. 2015), and they are called Pre-Mode 1 tools. These stone tools are thought to have been produced by Australopithecus.

2.1. The Lower Paleolithic Age In the Lower Paleolithic Age, the most basic human need was the physiological need for food and water to survive. The primitive Mode I and II stone tools were produced and used in this period. Early humans were omnivorous, so the early stone tools, such as choppers, scrapers, and pounders, represented a new category of products that met human needs for acquiring and processing foods more efficiently and extending the function of their hands. They used natural caves as a shelter against adverse environmental conditions. 2.1.1. Oldowan Stone Tools The Mode I tools were characterized by their simple construction (Stavrianos 1991; Klein 2009). They came from the earliest Paleolithic stone tool-making technology, the Oldowan industry, named after its type site found in Olduvai Gorge, Tanzania. The Oldowan industry was a percussion technology that predominantly used core forms. It used river pebbles or similar rocks to produce cores and struck them with a spherical hammerstone to cause conchoidal fractures to remove flakes from one surface. The removal of flakes created an edge and often a sharp tip at the distal end of the core, and the proximal end remained blunt. The hominin grasped the blunt proximal surface, bringing the distal surface down hard on an object they wished to detach or shatter. The flakes could be used as light-duty tools. Oldowan tools included choppers, scrapers, pounders, burins (with points for engraving), and awls (with points for boring). Fig. 12 illustrates an Oldowan chopper. These stone tools could help users prepare food for better consumption or make tools from other materials.

The Manual Age

7

Fig.1-2 A typical Oldowan simple chopping tool from the Duero Valley, Valladolid, Spain (by José-Manuel Benito Álvarez, CC-BY SA2.5, https://commons.wikimedia.org/w/index.php?curid=548566). The three images are views from three different directions.

The Oldowan industry flourished with early species of Homo, such as Homo habilis and Homo ergaster, in southern and eastern Africa between 2.6 and 1.7 million years ago. It could have been invented by the species Australopithecus garhi or Homo habilis (De Heinzelin et al. 1999; Toth and Schick 2009). Homo habilis was the hominin that used most Oldowan tools in Africa; about 1.9–1.8 million years ago, Homo erectus and Homo ergaster inherited them (Semaw 2000). Homo ergaster (meaning “working man”), living in eastern and southern Africa between 1.9 million and 1.4 million years ago, is also called African Homo erectus (Gabunia et al. 2000; Antón 2003; Rightmire, Lordkipanidze, and Vekua 2006). It is often thought to be the same species as Homo erectus (meaning “upright man”) living in Asia between 1.9 million years ago and 143,000 years ago, such as “Java Man” and “Peking Man” (Antón 2003; Indriati et al. 2011; Swisher III, Curtis, and Lewin 2001). The Oldowan tools are found across much of Africa, the Middle East, South and East Asia, and Europe. The lithic technology was thought to have spread out of Africa and into Eurasia by traveling bands of Homo erectus. The consumption of hunter-gatherers depended totally on the natural endowment of their region, so they needed to migrate from time to time, and their legs were their transport “tools.” They reached Riwat in Punjab in northern Pakistan by 1.9 million years ago (Dennell, Rendell, and Hailwood 1988), Caucasus by 1.8 million years ago (Gabunia et al. 2000; Garcia et al. 2010),

8

Chapter 1

Java by 1.8 million years ago (Swisher et al. 1994) and Northern China by 1.66 million years ago (Zhu et al. 2004). Oldowan tools were used at Lézignan-la-Cèbe in southern France 1.57 million years ago (Crochet et al. 2009) and at Monte Poggiolo in Italy 0.85 million years ago (Muttoni et al. 2011). The slow pace for the technology to spread reflected the lack of fast transport and communication tools. 2.1.2. Acheulean Stone Tools Homo erectus started the more complex Acheulean (Mode II) stone tool industry around 1.8 or 1.65 million years ago in the West Turkana region of Kenya (Roche et al. 2003; Lepre et al. 2011). Acheulean tools, characterized by their bifacial, oval, and pear-shaped “hand axes,” are found across Africa, Asia, and Europe. The Acheulean tools of the Madrasian culture in South India included flake tools, microliths (small stone tools), and other chopping tools, besides the characteristic bifacial handaxes and cleavers (Armand 1985; Avari 2016). The oldest tools date to 1.5 million years ago (Pappu et al. 2011). Fig.1-3 illustrates a typical Acheulean handaxe.

Fig.1-3 A typical Acheulean handaxe from the Duero Valley, Valladolid, Spain (by José-Manuel Benito–own work, Public Domain, https://commons.wikimedia.org/w/index.php?curid=548596). The four images are views from four different directions. The small flakes on the edge were from reworking.

Stone tools excavated at the Gongwangling site in Lantian, in the Shaanxi province of China, appear to be Acheulean (Wang, Lu, and Xing 2014). These include cores, flakes, choppers, hand axes, spheroids, and scrapers.

The Manual Age

9

The remnants of Lantian Man, Homo erectus lantianensis, found there were dated to 1.63 million years ago (Zhu et al. 2015). Acheulean tools were used in the Bose Basin in Guangxi, China by 800,000 years ago (Xu et al. 2012; Hou et al. 2000). Acheulean flint tools dating around 500,000 years ago were found at the Boxgrove site in West Sussex, UK (Pitts and Roberts 1997). About 250,000 years ago, Acheulean stone tools almost completely replaced Oldowan tools. 2.1.3. Fire The use of fire was an essential milestone in the history of human technology. A controlled fire could be used for cooking and keeping away large predators. It is generally accepted that the control of fire by Homo erectus had begun some 400,000 years ago (James et al. 1989), but a fire might have been used by the Lower Paleolithic hominin Homo erectus 1.5 million years ago (McClellan III and Dorn 2015). Burned animal bones found at the Xihoudu site in Yunan and dated to 1.8 million years ago are one of the earliest pieces of evidence of using fire by humans. In Shaanxi, blackened mammal bones and ash from campfires were found at the sites of Yuanmou Man and dated to 1.7 million years ago (James et al. 1989; Qian et al. 1991).

2.2. The Middle Paleolithic Age The Middle Paleolithic Age saw the emergence of anatomically modern humans, Homo sapiens. The fossils of Homo sapiens excavated in the Omo Kibish area, Ethiopia, were dated to 200,000 years ago (McDougall, Brown, and Fleagle 2005). The fossils unearthed at the Jebel Irhoud site in Morocco that were thought to be Homo sapiens were dated 300,000 years ago (Hublin et al. 2017). New tools, such as the Mousterian tools, replaced the Acheulean tools. 2.2.1. Stone Tools The Mode III or Mousterian (160,000–30,000 BP) industry, named after Le Moustier in the Dordogne region of France, adopted the Levallois or other prepared-core techniques to produce handaxes, racloirs, points, and other smaller and sharper knife-like tools (Haviland et al. 2012). The new approach allowed the invention of stone-tipped spears by hafting sharp, pointy stone flakes onto wooden shafts. Fig.1-4 shows a typical Mousterian point. These new tools represented a category of new goods, the attacking weapons for hunting. In Europe, the Mousterian industry was developed and

10

Chapter 1

used primarily by the Neanderthals (Semaw et al. 2003). In contrast, in North Africa and the Near East, it was used by anatomically modern humans.

Fig.1-4 A typical Mousterian point from the Darai Rockshelter in the Sirwan Valley of Haweaman, Zagros, Iran (by ICAR, Iran–own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=107930591). The four images are views from four different directions. The small flakes on the edge were from reworking.

In China, Mousterian stone tools have been excavated from the fourth and fifth layers of the Sanlongdong cave site in Chifeng of Inner Mongolia and dated to 50,000 years ago (Shan et al. 2017). Paleoanthropologists also found Mousterian stone tools in the Dahe Paleolithic cave site in Fuyuan of Yunnan, which included choppers, scrapers, carvers, and prepared cores with charcoal, burned animal bones, and burned clay. Its lower layer was dated by the uranium isotope method 40,000 years ago (Ji et al. 2006). 2.2.2. Cooking, Rafts, and Non-Lithic Tools The use of fire became common in the Middle Paleolithic Age (McClellan III and Dorn 2015; Toth and Schick 2007). The early humans began to cook their food at the latest in the early Middle Paleolithic (250,000 BP). Cooking may have improved their health because cooked foods are more digestible, tastier, and less likely to contain pathogenic bacteria (Wrangham and Conklin-Brittain 2003; Marlowe 2005). Heating may have degraded toxins contained in plants. This early cooking was the origin of the modern catering and food industry. Cooking can be viewed as an extension and externalization of the human digestive system.

The Manual Age

11

So far, the first and most efficient means is water transport. Humans used rafts to travel over large bodies of water to colonize other land (Bednarik 2014; Simmons 2016). The ancestors of Indigenous Australians are believed to have arrived in Australia 40,000 to 60,000 years ago, during a period of glaciation when New Guinea and Tasmania were joined to the continent (Roberts et al. 1994; Hiscock 2007). As the journey still required sea travel, they were among the world's earliest mariners (Butlin 1983; Roberts et al. 1994). Raft-making represented the emergence of the transport category of human products, which extended and externalized the function of their legs, backs, and shoulders to carry heavy loads and cross deep waters. Some researchers think the Lower Paleolithic hominid Homo erectus had possibly invented rafts (800,000 or 840,000 BP). Some of them used rafts to reach the island of Flores and evolved into the small Homo floresiensis. The Middle Paleolithic Homo erectus used bones and antlers to make tools such as awls and spear points. Harpoons made of bone were invented during the late Middle Paleolithic (90,000 BP) (Yellen et al. 1995). Before the invention of harpoons, humans might have caught fish with their bare hands, but harpoons made fish a more abundant food supply in human diets. Neanderthals hunted large game animals by ambushing them and attacking them with mêlée weapons such as thrusting spears and might have used projectile weapons (Richards and Trinkaus 2009; Boëda et al. 1999; Lazuén 2012).

2.3. The Upper Paleolithic Age The transition from the Middle to the Upper Paleolithic Age in the Levant (Syria, Lebanon, Israel, and Palestine) was the Emiran culture (50,000– 40,000 BP). It arose from the local Mousterian and kept many LevalloiseMousterian elements with the locally typical Emireh point (Copeland 2000). It used numerous stone blade tools, including curved knives. Western Europe's corresponding Châtelperronian (41,000–39,000 BP) industry produced denticulate stone tools and distinctive flint knives with a single cutting edge and a blunt, curved back. Paleoanthropologists found ivory adornments on the Châtelperronian sites. Art appeared at the end of the Middle Paleolithic Age and the beginning of this period. 2.3.1. Stone and Other Tools The Aurignacian (38,000–29,000 BP) tool industry followed the Châtelperronian and Ahmarian (46,000–42,000 BP) in Europe and southwest Asia. It produced flint tools and microliths, including fine blades

12

Chapter 1

and bladelets struck from prepared cores (Mellars 2006). In China, stone tools at the Shuidonggou site in Lingwu of Ningxia are comparable to Mousterian and Aurignacian ones and dated 30,000–20,000 years ago. In the Levant, the Antelian (32,000–20,000 BP) culture had evolved from the Emirian culture and incorporated some typical elements of Aurignacian. The Gravettian (33,000–21,000 BP) stone-tool industry in Europe was characterized by small pointed blades for hunting larger animals. Imprints of textiles pressed into clay were found at the Dolní VČstonice site (Svoboda et al. 2009); the weavers of the Upper Paleolithic produced plaited basketry, nets, and sophisticated twined and plain woven cloth (Adovasio, Soffer, and Klima 1996). The Gravettians also used nets for hunting smaller animals. Primitive cloth represented a new product category, which was the origin of the textile and garments industry. Bows and arrows may have been invented 60,000–70,000 years ago (Backwell, d'Errico, and Wadley 2008; Lombard and Phillipson 2010; Backwell et al. 2018). Egypt's Khormusan industry (42,000–18,000 BP) used animal bones, hematite, and stone to develop advanced tools, including small arrowheads (Vermeersch et al. 1982). The earliest definite finds of arrows and bows were associated with the Ahrensburg culture (eleventh to tenth millennia BCE) in north-central Europe (Riede 2009). Bows and arrows enriched the arsenals of attacking weapons. Perforated rods were found at Aurignacian sites and considered spear throwers or shaft wrenches (Shaw and Jameson 2008). The Solutrean (22,000–17,000 BP) tool-making industry and the Magdalenian culture (17,000–12,000 BP) used bone, antlers, ivory, and flintstone. Solutreans had relatively finely worked, bifacial points made with lithic reduction percussion and pressure flaking. They used antler batons, hardwood batons, and soft stone hammers for knapping and generating delicate slivers of flint to make light projectiles and elaborate barbed and tanged arrowheads. Tools and ornaments made from bone and stone, as well as remains of Homo sapiens found in the Upper Cave at Zhoukoudian, were dated to 10,000–20,000 years ago (Boaz et al. 2004). 2.3.2. Art Humans began to produce art such as cave paintings, rock art, and jewelry and engage in rituals. Art represents another category of human products that satisfy psychological needs for beauty and harmony. The oldest known cave painting, a red hand stencil in the Maltravieso cave at Cáceres in Spain, was dated more than 64,000 years ago and thought to have been made by a

The Manual Age

13

Neanderthal (Hoffmann et al. 2018). Previously, the oldest artworks were those from the caves in Sulawesi, Indonesia, dated 40,000 years ago (Aubert et al. 2014). Some Stone Age rock paintings found in the Bhimbetka rock shelters in the Raisen District in the Indian state of Madhya Pradesh are approximately 30,000 years old (Hoffmann et al. 2018). The Aurignacian culture created some of the earliest known cave art, such as the animal engravings at Trois Freres and the paintings at the Chauvet Cave in southern France. The works in the caves of Altamira, El Catillo, and Tito Bustillo of Spain have been dated to the Aurignacian culture and are approximately 35,600 years old (Pike et al. 2012). The Aurignacian people also made pendants, bracelets, ivory beads, and figurines. Hundreds of Venus figurines have been found on Gravettian sites. The Magdalenian tools, such as perforated batons and intricately engraved projectile points, served functional and aesthetic purposes. They also made artistic and personal adornments, including figurines, sea shells, and perforated carnivore teeth, presumably necklaces (Audouze 1987).

3. The Mesolithic Age The Mesolithic Age, sometimes called the Epipaleolithic Age, was the transitional period between the Paleolithic and the Neolithic ages. The Kabaran or Kebarian culture (21,000 or 18,000–12,500 BCE) in the Levant and Sinai areas started to collect wild cereals, which was the first step toward the Neolithic Revolution.

3.1. Mesolithic 1: Domestication of Dogs and Collections of Wild Cereal Genetic findings suggest that dogs were domesticated in Siberia 23,000 years ago (Perri et al. 2021). It might have occurred much earlier in other regions in the Upper Paleolithic Age (Germonpré et al. 2009; Druzhkova et al. 2013; Sablin and Khlopachev 2002). Domesticated dogs were a great extension of many human sensory and motion functions involved in hunting, such as observing, searching, chasing, fighting, and catching prey. Kabarans and some other Mesolithic cultures had domesticated dogs. Kebarans were a highly mobile nomadic population of hunters and gatherers in the Levant. They used bows and arrows and made small tools with bladelets struck off single-platform cores, burins, and end scrapers (Dayan 1994). The Kebarans began building oval-shaped brushwood huts that

14

Chapter 1

averaged between 2.7 and 4.9 meters long. These huts represented the beginning of the human housing industry that ended the human reliance on natural caves for accommodation. Housing and clothing were primarily externalizations of protective body functions against adverse external conditions. In Hut 1 on the Ohalo II site, seeds of 13 species of fruit and cereal, including brome grains, wild barley, and millet grass grains, were found around a grinding stone, suggesting extensive preparation before consumption (Weiss et al. 2008). Proto-weeds grew near human camps and probably in small-scale, cultivated plots (Snir et al. 2015). The site has been dated to around 19,400 BP (Mithen 2011), some 11,000 years before the accepted onset of agriculture. Along the Nile Valley and Nubia in Egypt, the Halfan culture (20000– 12,000 BCE) was characterized by three primary tools: Halfa flakes, backed microflakes, and backed microblades (Smith 1976). The Halfans depended on hunting, fishing, and collecting for survival. The hunter-gatherer Zarzian culture (18,000–8000 BCE) of the late Paleolithic and Mesolithic Ages in Southwest Asia was also associated with the domesticated dog and the introduction of the bow and arrow. Plenty of microliths were found on Zarzian sites (Wahida 1981).

3.2. Mesolithic 2: The Early Signs of Agriculture The Natufian culture (12,500 to 9500 BCE) succeeding the Kebaran might have started deliberate cultivation of rye at Tell Abu Hureyra, a sign of agriculture (BarဨYosef 1998), but Natufians were generally still huntergatherers. They hunted gazelles, deer, aurochs, wild boars, onagers, and caprids (ibex). Waterfowl and freshwater fish formed part of the diet in the Jordan River valley. They might have collected wild cereals, legumes, almonds, acorns, and pistachios. Natufians had domesticated dogs. The Natufians built semi-subterranean habitations, often with a dry-stone foundation and the superstructure probably made of brushwood. The Natufians had a microlithic industry based on short blades and bladelets. Researchers found geometric microliths (including lunates, trapezes, and triangles), backed blades, sickle blades, shaft straighteners made of ground stone, and heavy ground stone bowl mortars at Natufian sites. There was a prosperous bone industry, including harpoons and fishhooks. Natufians also had rich artistic products: pendants and other ornaments made of stone or bone and animal and human figurines made of limestone. The Ain Sakhri Lovers, a carved stone object found in the Ain Sakhri cave in the Judean

The Manual Age

15

desert, is the oldest known depiction of a couple having sex (Boyd and Cook 1993). The Qadan culture (13,000–9000 BC) people in Upper Egypt not only harvested and consumed wild grasses and grains but also watered and cared for the local plants (Darvill 2003). They also developed sickles and grinding stones to aid in the collecting and processing of these plant foods. The Harifian (8800–8000 BCE) culture in the Negev Desert of southern Israel was characterized by semi-subterranean houses, often more elaborate than those found at Natufian sites. For the first time, arrowheads were found among the stone tool kit. Harifian industry was characterized by microlithic points (Salem 1976; BarဨYosef 1998).

3.3. Mesolithic Pottery Made by Hunter-Gatherers The earliest pottery yet known anywhere in the world was those found at the Xianren Cave, located in Wannian County, Jiangxi, China. The prehistoric pottery shards were dated between 20,000 and 19,000 BP, and the site bore evidence of early rice cultivation (Wu et al. 2012). Another candidate for the oldest pottery was in the Yuchanyan Cave in Daoxian County, Hunan, China. The cave's pottery shards and other artifacts were dated by analysis of charcoal and bone collagen to a date range of 17,500 to 18,300 BP (Boaretto et al. 2009). Pottery was another important category of products, providing containers and cooking pots that broadened the types of food so humans could enjoy soups, porridges, stews, and other boiled foods. Before this invention, humans could only bake and roast their foods. The Jǀmon period (14,000–300 or 1000 BCE) was a hunter-gatherer culture in Japanese prehistory. Its pottery is generally accepted to be among the oldest in East Asia and the world (Kuzmin 2006). The name “cord-marked” derived from the pottery style characteristic of the first phases of Jǀmon culture, which decorated pottery by impressing cords into the surface of wet clay. The Ertebølle culture (5300–3950 BCE) in southern Scandinavia and the Dnieper–Donets culture (5000–4200 BCE) in the area north of the Black Sea/Sea of Azov, among others, were also making pottery at the end of the Mesolithic period (Jordan and Zvelebil 2009).

16

Chapter 1

4. The Neolithic Revolution: The Emergence of Agriculture The Neolithic culture first emerged in the Levant around 10,200–8800 BCE. The Khiamians (10,200–8800 BCE), which succeeded the Natufian culture, had experimented with farming (Cauvin 2000). The main difference between the Paleolithic and Neolithic Ages was not simply an improvement in making stone tools. Instead, it was a major change in the dominant production technology (Ma 2019, 2020), from hunting-gathering to agriculture, i.e., farming animals and plants. The appearance of agriculture has been the most important technological progress for humanity. It provided a more sustainable food supply than hunting and gathering, enabling more complicated social structures. On the west bank of the Euphrates in Raqqa of northern Syria, barley, rye, and Polygonum were harvested and cultivated. The following and overlapping Pre-Pottery Neolithic A (PPNA) culture (9500–8000 BCE) was the first stage in early Levantine and Anatolian Neolithic culture (Cauvin 2002; Aurenche et al. 2001).

4.1. Farming and Domestication of Animals The PPNA cultivated local grains such as barley and wild oats and built granaries for storage, but their cultivation is considered “pre-domestication.” Iraq ed-Dubb (the Cave of the Bear) northwest of Ajlun in the Jordan Valley shows the earliest evidence for domestic cereals with a date range from 9600 to 9475 BCE (Kuijt and Goodale 2006; Colledge et al. 2004). The PrePottery Neolithic B (PPNB) culture (8800–6500 BCE) began to depend more heavily upon domesticated animals to supplement their earlier mixed agrarian and hunter-gatherer diet. The eight so-called founder crops, emmer and einkorn wheat, then hulled barley, peas, lentils, bitter vetch, chickpeas, and flax, occurred more or less simultaneously on PPNB sites in the Levant. Tell Aswad (9300–7500 BCE) in Syria was one of the oldest PPNB sites with domesticated emmer wheat dated to 8800 BCE (Salamini et al. 2002). Goats were evident in the early stages of its occupation, and there were pigs, sheep, and cattle from the middle PPNB period (Horwitz et al. 1999). In China, millet and rice were among the first domesticated crops, followed by beans, mung, soy, and azuki. The Nanzhuangtou culture (9500–9000 BCE) near Lake Baiyangdian in Xushui County of Hebei had domesticated dogs (Yuan 2010) and cultivated millet (Yang et al. 2012). The Peiligang (7000 to 5000 BCE), Houli (6500–5500 BCE), and Cishan (6500–5000

The Manual Age

17

BCE) cultures raised dogs, pigs, chickens, goats, sheep, and cattle (Yuan 2010). Wheat, barley, and jujube are thought to have been domesticated by 9000 BCE in the Indian subcontinents. During the Mehrgarh Period I (7000–5500 BCE), people cultivated barley, einkorn, and emmer wheat and raised sheep, goats, and cattle. This period saw the first domestication of the elephant (Gupta 2004). In Greece's Late Neolithic period (5300–4500 BCE), sheep and goats were raised for their wool and used to weave garments (Barber 1991). In southeastern and central Europe, the Vinþa culture (5700–4500 BCE) kept livestock for milk, leather, and as draft animals (Šljivar 2006). The Cucuteni-Trypillian culture (4800–3000 BCE) in Eastern Europe might have domesticated the horse in its Early Period (4800–4000 BCE) (Anthony and Brown 2011). The oldest known depiction of a wagon on the Bronocice pot from Poland was credited to the Funnel-beaker culture (4300–2800 BCE). The invention of the wagon and the training of draft animals were a new addition to the transport tools to extend human legs and shoulders to travel and carry heavy loads on land. The Yangshao (5000–3000 BCE) and Longshan (3000–1900 BCE) cultures in China practiced silkworm cultivation and made fabric from silk or hemp (Chang 1986; Liu and Chen 2012). In southeastern Turkey, Syria, and northern Iraq, the Halaf culture (6100 and 5100 BCE) practiced dryland farming by exploiting natural rainfall (Liverani 2013). In northern Mesopotamia, the Samarra culture (5500–4800 BCE) developed irrigation agriculture (Blackham 1996). The first canal irrigation at Choga Mami seemed to operate at about 6000 BCE. In southern Mesopotamia, during the Ubaid phase I (5400–4700 BCE), people around Eridu grew grains in the extremely arid conditions of Southern Iraq, taking advantage of the high local water tables (Roux 1992). In the Ubaid phase II (4700–4500 BCE), developing extensive canal networks from major settlements became the first project that required a collective effort and centralized coordination of labor in Mesopotamia (Kurt 1995; Wittfogel 1957). The Yangtze River Delta's Liangzhu culture (3400–2250 BCE) engaged in drainage and irrigation, paddy rice cultivation, and aquaculture (Weisskopf et al. 2015). Maize was domesticated from the wild grass teosinte in West Mexico by 6700 BCE (Piperno et al. 2009). Squash, avocados, and bottle gourds were domesticated by 8000 BCE; beans by 7000 BCE; cassavas, peanuts, and cocoa by 6000 BCE; cotton by about 4000 BCE; and pepper by 3000 BCE (Piperno 2011). The tomato was also domesticated in America. In the

18

Chapter 1

Andean region, the potato was domesticated approximately 7,000–10,000 years ago (Engel 1970). The remnants of six finely woven textiles and cordage made from plant fibers, found in the Guitarrero Cave, Peru, dated between 10,100 and 9080 BCE, are the oldest known textiles in the Americas (Jolie et al. 2011). The guinea pig was first domesticated as early as 5000 BCE for food by tribes in the Andean region of South America. Camelids (primarily llamas and alpacas) were domesticated by 4,000 BCE (Rischkowsky and Pilling 2007).

4.2. Neolithic Tools, Pottery, and Art The transitional Shepherd culture (10,200–8800 BCE) represented an industry of small flint tools from the Hermel plains in the north Beqaa Valley of Lebanon (Fleisch 1966). The tools included short denticulated or notched blades, end scrapers, transverse racloirs on thin flakes, and borers with solid points. The oldest chert arrowheads with lateral notches, the socalled “El Khiam points” (Fig.1-5), the appearance of small female statuettes, and the burying of auroch skulls characterized the transitional Khiamian culture (Cauvin 2000). The succeeding PPNA lithic industry was based on blades struck from regular cores (Edwards et al. 2004). The axes shaped by transverse-blow and polished adzes appeared for the first time (Garfinkel and Nadel 1989; Bar-Yosef 1989).

Fig.1-5 A typical El Khiam point microlith from El Khiam, Syria (by José-Manuel Benito–own work, public domain, https://commons.wikimedia.org/w/index.php?curid=1030893).

The PPNB and the pottery-lithic Jarmo and Hassuna people used grinding equipment, containers made of stone or mud for processing and storing

The Manual Age

19

farming products, and baking ovens (Nishiaki and Le Mière 2005; Adams 1983). One of the major PPNB flint tool elements was the naviform core. Many PPNB tools and weapons, including Aswadian and Jericho point arrowheads, were made of flint (Edwards 2016). Stone sickles, cutters, bowls, and receptacles of engraved marble were found at the Jarmo sites. In China, the Nanzhuangtou people used stone-grinding slabs and rollers and made bone artifacts. The Peiligang, Cishan, Xinle (5500–4800 BCE), Hemudu (5500–3300 BCE), Beixin (5300–4100 BCE), Yangshao, and Liangzhu people used stone arrowheads, spearheads, ax heads, chisels, awls and sickles (Liu and Chen 2012). The Hemudu people also used animal bones to make tools, such as hoes, harpoons, bows, and arrowheads. The Yangshao people used grinding stones to make flour (Chang 1986). Grinding grains for easy consumption represented a quality improvement in production. Basketry and weaving seemed commonplace at the PPNB and the pottery lithic sites, evidenced by the imprint of embroidered fabric on a plaster fragment (Wendrich and Ryan 2012; Verhoeven 2000; Schick 1988). The earliest known Neolithic textile production evidence is a cloth woven from hemp at the Çatalhöyük site, dated around 7000 BCE. Weaving also appeared in Egypt’s Faiyum A culture (6000–4800 BCE) (Simpson 2018). The Peiligang, Cishan, and Beixin people used fishing nets made from hemp fibers. Beixin people used wild hemp fibers to weave fabric for clothing and various thread, twine, and rope forms. At the Liangzhu sites, remains of boats, oars, a wooden pier, and an embankment have been found (Liu 2005; Liu and Chen 2012). Sailing boats were used for riverine and maritime during Mesopotamia's Ubaid period (6000–4300 BCE) (Carter 2006). Austronesians from southern China were the first actual ocean-faring people who expanded widely in the Pacific and Indian oceans during 3000–1500 BCE (Bellwood 1991). The sail was an important invention that used wind power to drive boats/ships, indispensable for trade and geographic discoveries in later periods before the invention of steamboat/ship. The earliest agricultural tool is thought to have been the digging stick, used to prepare the soil for planting or dig out underground food such as roots and tubers (Allen 1970; Tylor 1881). Hoes were another important ancient technology. Digging sticks, hoes, and mattocks were invented in the early Neolithic Age (Milner, Hammerstedt, and French 2010; Elliott 2015). A true plow/ard that turns the soil for sowing seed needs more draft power than a human worker. The animal-drawn true ard was developed after the domestication of oxen in Mesopotamia and the Indus valley in the sixth millennium BCE (Isaac 1962). Plowing makes soil soft and friable, aerates

20

Chapter 1

the ground, and improves its ability to retain moisture, making land tilling easier and faster. Tools represent an essential human need, the methodological need for tools to make meeting other needs possible or efficient. Agricultural tools facilitate agricultural production. Pottery was not yet in use in the PPNA culture. PPNB people had the earliest proto-pottery, white ware vessels made from lime and gray ash, built around baskets before firing (Garfinkel 1999a). The pottery found at the Nanzhuangtou Ruins in China was dated 10,200 BP (Liu and Chen 2012). The manufacture of pottery in various shapes and sizes and for different uses at Yarmukian culture (6400–6000 BCE) sites gives this cultural stage its name, the Pottery Neolithic (Garfinkel 1993). Other cultures further improved pottery technology and produced very fine-glazed earthenware decorated with geometric paintings or stylized animals in one or more colors (called polychrome), white or red-inlaid clay decoration, and various geometric designs (Liverani 2013). The potter's wheel was thought to be first invented in Sumer around 4500 BCE (Kramer 2010). The Neolithic Age saw the creation of more artworks. The PPNB, Merimde, Yarmukian, Halaf, Cucuteni-Trypillian, Neolithic Knossos in Crete, Sesklo cultures, and Mehrgarh Periods I, II (5500–4800 BCE) and III (4800–3500 BCE) all had geometric shape ornaments as well as animal and human figurines made of fired clay or carved on stone (Liverani 2013; Demoule and Perlès 1993; Possehl 2002). Most human figurines were female images, interpreted as representing a goddess. The Knossos figurines included nude sitting or standing females with exaggerated breasts and buttocks. Some Sesklo statuettes of women appeared pregnant, probably reflecting the influence of the widely hypothesized prehistoric fertility cult. The Merimde people made the first Egyptian lifesize head of clay. Halaf people made stamp seals of stone, thought to mark the development of concepts of personal property. The Mehrgarh people produced the first button seals from terracotta and bone. They also developed stone drills, updraft kilns, tanning, and bead production technologies. Some researchers conjectured that the Vinþa symbols were an early form of proto-writing. Neolithic China was also full of artistic creativity. At the sites of Xinle, Hemudu, Dawenkou (4300–2600 BCE), Yangshao, Hongshan (4700–2900 BCE), Liangzhu, and Shijiahe (2500–2000 BCE), miniature clay or stone figurines, and artifacts made of jade, ivory, or turquoise are commonly found. One wooden carving excavated at a Xinle site was dated 7,200 years ago. The Hemudu people also produced lacquer wood and musical instruments such as bone whistles and wooden drums. Fig.1-6 shows a

The Manual Age

21

pottery bowl with a pig image found at the Hemudu site. The earliest examples of alligator drums appeared at Dawenkou sites. The Longqiuzhuang (5000–3000 BCE) culture in Gaoyou of Jiangsu is famous for the sherds and deer antlers bearing Neolithic signs. Neolithic signs were also found on Dawenkou pottery and might be related to the earliest Chinese writing (Maisels 2001; Liu and Chen 2012).

Fig.1-6 A pottery bowl with a pig image from the Hemudu site, Yuyao, Ningbo, Zhejiang, China (by G41m8, own work, CC BY-SA4.0, https://commons.wikimedia.org/w/index.php?curid=47063664).

Jadeware was crucial in early Chinese culture. Jade pig dragons and embryo dragons, clay figurines of pregnant women, and small copper rings were found at Hongshan sites. The Liangzhu culture was characterized by finely worked large ritual jades commonly incised with the taotie (a ferocious mythical animal) motif, such as Cong (cylinders), Bi (discs), and Yue axes (ceremonial axes) (Maisels 2001). Some axes were crafted using diamond tools and polished to a mirror-like luster. This was the earliest known use of diamond tools worldwide, thousands of years earlier than elsewhere. Jade pendants with engraved representations of small birds, turtles, and fish were also common (Lu et al. 2005). It was the only prehistoric culture known to work in sapphire. The jade carving technology exhibited by the artifacts at Shijiahe sites exceeded that of the Liangzhu and Hongshan cultures renowned for their jades (Yu 2015). The PPNB people in the Levant imported obsidian from Anatolia. The Halaf and Samarra pottery vessels had been widely exported. Sea shells from the Mediterranean, polished stone vessels made of alabaster or marble, and blades made from obsidian from Turkey (700 km away) were discovered at the Yarmukian Sha'ar HaGolan site, indicating trades with distant communities

22

Chapter 1

(Garfinkel 1999b). The Badarian culture (5000–4000 BCE) near Der Tasa in Upper Egypt had imported shells, turquoise, copper, and porphyry slabs (Brunton and Caton-Thompson 1928). Long-distance trade also existed during Mehrgarh Period in the Indus valley as sea shells came from the distant sea shore and lapis lazuli from as far away as present-day Badakshan, Afghanistan (Jarrige 1975). Different endowments of natural resources are one of the causes of trade between regions even now, as is the technological difference. Therefore, cultures with advanced pottery techniques exported pottery.

4.3. Accommodations, Urbanization, and Governance The PPNA constructed round semi-subterranean houses with stone foundations, terrazzo-floors, and unbaked clay mudbrick upper walls. The most famous PPNA site was at Jericho, with the famous tower of Jericho, a communal structure. It was thought to be the world's first town (10,000 BP), containing a population of up to 2,000–3,000 people, protected by a massive stone wall and tower. The higher sustainable productivity of agriculture was the economic foundation of Neolithic towns, which also represented a scaleup, expansion, and alienation of the social interactions within a Paleolithic hunter-gatherer herd or the interactions among an individual’s internal systems. Houses at Tell Aswad were round, elliptical, or polygonal. The architectural styles of the PPNB people in the southern Levant became primarily rectilinear. Homes had a thick layer of highly polished white clay plaster floors made of lime produced from limestone (Mazar 1990). The Neolithic Klimonas (9100–8600 BCE), Çatalhöyük (7500–5700 BCE) in southern Anatolia, Knossos (7000–5000 BCE), Peiligang in China, and many other Neolithic cultures all built elliptical-, polygonal-, round-, or rectilinear-shaped houses with wood, wattle-and-daub or mudbrick. The larger settlements could comprise about 500–800 homes with perhaps up to 5,000 people. The Klimonas people built half-buried mud brick communal buildings 10 meters in diameter and surrounded by dwellings (Vigne et al. 2014). The Yarmukian Sha'ar HaGolan site had large courtyard houses, ranging between 250 and 700 m² in area, with a central courtyard surrounded by several small rooms (Garfinkel 1999b). In the Middle Neolithic (50004000 BCE) Knossos, the Great House, a 100 m2 (1,100 sq ft) area stone house, was divided into five rooms with meter-thick walls, which was likely to be for communal use and might have been the predecessor of a palace (Evans 2013).

The Manual Age

23

Increased farming productivity enabled the community to raise an elite class of administrative people responsible for maintaining the social order and organizing production and artisans who produced goods that met people’s psychological needs. Eridu was founded around 5400 BCE at the beginning of the Ubaid phase I (5400–4700 BCE). Sizable unwalled village settlements, multi-roomed rectangular mudbrick houses, and the appearance of the first temples of public architecture in Mesopotamia characterized the Ubaid culture. Eridu had long been considered the earliest city in the region, and the settlement was centered on an impressive temple complex (the House of the Aquifer) built of mudbrick (Leick 2002). According to the Sumerian King List, Eridu was the first city in the world. In Sumerian mythology, Eridu was the home of the Abzu temple of Enki, the god of deep waters, wisdom, and magic, and one of the five cities built before the Deluge occurred. Eridu had specialized craftspeople, potters, weavers, and metalworkers. Such early forms of division of labor increased productivity and supply of non-agricultural goods. Farmers with better tools made by artisans could also increase their labor productivity. A unique ritual complex associated with the Hongshan culture was found in Niuheliang. The structure constructed of stone platforms with painted walls was named Goddess Temple due to the discovery of a female clay head with jade inlaid eyes. Housed inside the Goddess Temple were clay figurines three times the size of real-life humans (Nelson 1997). The largest ancient Liangzhu city found so far was in a wetland on the plain between Daxiong Mountain and Dazhe Mountain. Its interior was 290 hectares, surrounded by clay walls with six city gates. A palace site at its center spanned 30 hectares (Renfrew and Liu 2018). The early appearance of temples in various large Neolithic cultures demonstrated the importance of the human spiritual need to account for their purposes of life and their relationship with nature or the universe’s controller/creator(s). The Shijiahe site cluster in Tianmen County of Hubei was 120 hectares (300 acres) in area, likely with between 15,000 and 50,000 inhabitants within its walls. Both its size and population were larger than the Bronze Age Erlitou site. Boats seemed to be the primary means of travel, and residents constructed channels to connect urban core areas to adjacent rivers or from towns to main rivers. In addition to walls, moats were dug around towns and urban centers. At the town site at Chengtoushan, the moat was about 40–50 m wide. The oldest fired bricks in the world were found at Chengtoushan, dating back to 4400 BCE (Yasuda 2013). Some researchers considered Shijiahe an ancient state (Zhang 2013).

24

Chapter 1

5. The Bronze Age and the Division of Labor From about 3300 BCE, humankind gradually entered the Bronze Age. Bronze is an alloy consisting primarily of copper, commonly with about 12% tin (tin bronze) or sometimes with non-metals or metalloids such as arsenic (arsenic bronze). Copper occurs primarily in copper carbonate hydroxide Cu2(CO3)(OH)2, malachite. Copper(II) oxide can be obtained by expelling carbon dioxide and water into the atmosphere in a process called “roasting” and then directly reduced to copper by a reducing environment. The archaeological site of Belovode on the Rudnik mountain in Serbia contains the world's oldest securely dated evidence of copper smelting at high temperatures, from 5000 BCE (Radivojeviü et al. 2010). A similar reduction process can produce tin from cassiterite (SnO2). The main difference between the Neolithic and Bronze Ages was not the dominant production technology; it was a revolutionary improvement in the tools used in the agricultural economy. Tools made of bronze were harder and more durable than copper. Bronze tools far exceeded stone tools in their efficiency and versatility. Using bronze tools increased the productivity of agriculture and other production sectors, which could support a relatively sizable ruling class to manage the economy and more artisans who specialized in making better tools and crafts. An ancient civilization entered the Bronze Age when it used bronze tools made from self-produced bronze or bronze traded from production areas elsewhere. The Bronze Age could be divided into the Early (3300–2100 BCE), Middle (2100–1550 BCE), and Late (1550–1200 BCE) Bronze Ages. South America's Moche civilization (100–700 CE) independently developed bronze (Donnan 1973).

5.1. The Transitional Chalcolithic or Copper Age The Final Neolithic or Chalcolithic (4500–3200 BCE) period, also called the Copper Age, entailed transitioning from the Neolithic to the Early Bronze Age. Although the Vinþa culture provided the earliest known example of copper metallurgy, it is not conventionally considered part of the Chalcolithic or Copper Age. The Vinþa site in Ploþnik of Serbia had produced probably the earliest copper tools and bronze in the world, and copper ores were mined at sites like Rudna Glava (Radivojeviü et al. 2010). The Chalcolithic cultures included the Leyla-Tepe culture (4350–4000 BCE) of ancient Azerbaijan; the Lengyel culture (5000–3400 BCE) in its later period; the Badarian culture, the Amratian (Naqada I) culture (4,400–3,500 BCE) in Upper Egypt, the Boian culture (4300–3500 BCE) toward its end, the Cucuteni-Trypillian culture in its Middle Period (4000–3500 BCE); and

The Manual Age

25

the Corded Ware culture (2900–2350 BCE), etc. They had acquired the technology for copper metallurgy or shown signs of using copper, such as copper axes and other tools. The Gerzean (Naqada II) culture (3500 BCE– 3200 BCE) used copper for all kinds of tools, including the first copper weaponry (Gardiner 1961). The Chalcolithic Merhgarh Period III people developed technologies for copper drills, copper melting crucibles, and metalworking (Sharif and Thapar 1992).

5.2. The Economy and Technological Progress The Early Bronze Age began during the Uruk Period (4000–3100 BCE), named after the Sumerian city of Uruk in Mesopotamia (Kramer 2010). The invention of bronze tools markedly raised agricultural labor productivity. It was estimated that 10,000 animals were kept in sheepfolds and stables, and 3,000 were slaughtered yearly in Ur, a large town with a population of 6,000 (Gorlinski 2012). During the Jemdet Nasr period (3100–2900 BCE) in southern Mesopotamia, farmers used light, unwheeled plows pulled by oxen to prepare the land. Sumerians used oxen-drawn wagons with solid wheels covered by leather tires kept in position by copper nails. They harnessed the oxen by collars, yokes, and headstalls and controlled the animals by reins, a ring through the nose or upper lip, and a strap under the jaw. The late Copper Age/early Bronze Age Maykop culture (3700–3000 BCE) in the Western Caucasus might have trained horses for riding or pulling heavy loads (Anthony et al. 1986). The horse was in use by the Sumerians around 2000 BCE. The most ancient bronze sword on record, dating to 3300–3100 BCE, was found at the Maykop site. The Kura-Araxes (3400– 2000 BCE) culture in the Caucasus was remarkable for producing wheeled vehicles, viticulture, wine-making, the wine goblet, and the large ceramic vessels used for grape fermentation (Edens 1995; Batiuk 2013). Cornwall during the British Bronze Age (2500–1700 BCE) was a major source of tin for much of Western Europe, and copper was extracted from sites such as the Great Orme mine in North Wales (Muhly 1985; Dutton et al. 1994). The agricultural system in southern Mesopotamia during the Early Dynastic period (2900–2350 BCE) was probably the most productive in the ancient Near East, allowing the development of a highly urbanized society. The conquests of Sargon and his successors marked the end of the Early Dynastic period and the beginning of the Akkadian Empire (2334–2154 BCE). The population of Akkad was entirely dependent upon agriculture. Farming relied on intensive irrigation. The high agricultural productivity enabled the growth of the highest population densities in the world at the

26

Chapter 1

time, giving Akkad its military advantage. Roads and a regular postal service bound the empire together. The irrigation ditches and drainage systems required constant maintenance, and city temple authorities recruited farmers for this work. Sumer and Akkad were short of metal ores, timber, and building stone, all of which had to be imported (Crawford 2004). During the Aegean Bronze Age (3200–1050 BCE), tin and charcoal were imported to Cyprus, where copper was mined and alloyed with tin to produce bronze. In Knossos, the Minoan civilization (2600–1100 BCE) was a mercantile people engaged in overseas trade. They developed Mediterranean polyculture, the practice of growing more than one crop at a time. The simplest plow (an ard) appeared and spread throughout Europe between 2500 and 2000 BCE. The wooden plow was pulled by pairs of donkeys or oxen (Dickinson 1994). The Mycenaean (1600–1100 BCE) network of roads in the Peloponnese and their megalithic Cyclopean fortifications at Mycenae and Tiryns demonstrated the zenith of Greek engineering. Mycenaean trade routes reached Cyprus, Amman in the Near East, Apulia in Italy, and Spain. A central administration managed the Mycenaean economy. Its palaces monitored various industries and commodities, the organization of land management, and the rations given to the dependent personnel. The workforce was specialized; each worker was assigned to a specific task in the stages of production. The palatial centers organized their workforce and resources to construct large-scale agriculture, industry, and infrastructure projects, such as a large dam near Tiryns and the drainage systems of the Kopais basin in Boeotia and the Nemea valley (Castleden 2005). The oldest bronze object in China was a knife found at a Majiayao site in Dongxiang, Gansu, dated 2900–2740 BCE. More copper and bronze objects have been found at Machang sites in Gansu. Metallurgy spread to the middle and lower Yellow River region in the late third millennium BCE (Liu and Chen 2012). The copper pieces found at Dengjiawan within the Shijiahe culture (2500–2000 BCE) cluster were the earliest copper objects discovered in southern China (Zhang 2013). The Qijia culture (2200–1600 BCE) around the upper Yellow River region produced some of China's earliest bronze and copper mirrors. Their metal knives and axes from copper and copper-tin alloys pointed to early contact with Central Asia (Chen 2013). In Shanxi and Henan, the Erlitou culture (1900–1500 BCE) produced ritual bronze vessels, including the earliest recovered dings (cooking vessels with three or four legs). Chinese archaeologists generally identify the Erlitou culture as the Xia Dynasty (2070–1600 BCE), the first dynasty in Chinese

The Manual Age

27

history. The Erligang culture (1600–1400 BCE) across the Yellow River Valley and the Yangtze regions developed bronzes from the Erlitou style and techniques and started using bronze castings. Bronze vessels became more widely used and uniform in style. The chariot first appeared in China around 1200 BCE (Liu and Xu 2007). The Indus Valley Civilization (IVC) (3300–1300 BCE), also known as the Harappan Civilization, developed metallurgy (copper, bronze, lead, and tin) during the Early Harappan phase (3300–2600 BCE) and animal-drawn plows by 2500 BCE. Harappans used wheeled transport such as bullock carts. Their plank-built watercraft with a central mast and a sail of cloth or woven rushes were also suitable for sea trade. They developed a system of uniform weights and measures and achieved great accuracy in measuring length, mass, and time. Harappans built cities with elaborate drainage and water supply systems. The coastal city of Lothal in Gujarat had a massive, dredged canal and a docking facility. An extensive maritime trade network had operated between the Harappan and Mesopotamian civilizations since the Middle Harappan Phase (2600–1900 BCE) (Wright 2010).

5.3. Writing, Literature, Philosophy, Science, and Art The invention of writing made it possible to transmit and spread information across space and time efficiently, which was an extension and externalization of human communication and memory functions. Before the invention of writing, knowledge could only exist in human brains as a memory. Communities relied on a few intelligent people with a good memory, revered as sages, to keep their collective knowledge. The cuneiform script emerged during the late Uruk period (34th to 32nd centuries BCE). By the time of the Jemdet Nasr period, the script had undergone significant changes, moving from original pictographs to simpler and more abstract designs and acquiring its iconic wedge-shaped appearance. The language of the texts on these clay tablets was thought to be Sumerian, and they dealt without exception with administrative matters such as the rationing of foodstuffs or listing objects and animals. In the Early Dynastic IIIa period (2600– 2500/2450 BCE), syllabic writing began, and the full flow of human speech was first recorded around 2600 BCE (Kramer 2010). Enheduanna, the high priestess of Nanna (the Sumerian moon god) of the temple of Sin at Ur and the daughter of Sargon, was the first poet with a name known in history. Her works included hymns to the goddess Inanna and the Temple Hymns. The Perforated Relief and the Plaque of Ur Nanshe of King Ur-Nanshe, the first king of the First Dynasty of Lagash (c. 2500

28

Chapter 1

BCE), his grandson Eannatum's Stele of the Vultures, and the silver vase and calcite vase dedicated by Eannatum’s nephew Entemena to his god, demonstrated a high degree of excellence in art and techniques. The copper Bassetki Statue, cast with the lost wax method, showed the artisans’ high skill level. A cadastral survey seemed to have been instituted, and the first collection of astronomical observations and terrestrial omens was made for a library established by Sargon (Kramer 2010; Crawford 2004). Some symbols on Gerzean pottery resemble traditional hieroglyph writing. The earliest hieroglyphs appeared just before the Early Dynastic Period of Egypt. By the end of the Third Dynasty (2686–2613 BCE), Egyptian writing had been expanded to include more than 200 symbols, both phonograms and ideograms (Allen 2000; Ray 1986). King Djoser (2667–2648 BCE) was the second king of the Third Dynasty. His architect Imhotep built the Step Pyramid at Saqqara. Under the Fourth Dynasty (2613–2494 BCE), Khufu (2589–2566 BCE) built the Great Pyramid of Giza, his son Khafra (2558– 2532 BCE) the second pyramid and possibly the Sphinx in Giza, and his grandson Menkaure (2532–2504 BCE) the third pyramid. The Old Kingdom was one of the most dynamic periods in the development of Egyptian art. Sculptors created the earliest portraits of individuals and the first lifesize statues, perfected the art of carving intricate relief decorations, and produced detailed images of animals, plants, and landscapes (Gardiner 1961; Shaw 2003). The Cretan hieroglyphs were found on artifacts of the early Bronze Age during the Minoan period and still need to be deciphered. They were used during 2100–1700 BCE. The Minoan language Linear A used during 2500– 1450 BCE still needs to be deciphered. The Mycenaean syllabic script, Linear B, was used during 1450–1200 BCE and has been deciphered. The preserved Linear B records mainly dealt with administrative issues of Mycenean palatial states, indicating that the administration was uniform with the same language, terminology, and system of taxation and distribution. The Mycenean religion already included several deities that could be found in the Olympic Pantheon (Castleden 2005; Dickinson 1994). The inscribed symbols on pottery shards from around 1600 BCE of the Wucheng culture might be a form of writing, and about 120 inscriptions have been found (Liu and Chen 2012). The Shang people had a fully developed writing system, the predecessor of modern Chinese script, preserved on inscriptions on bronze, pottery, jade and other stones, horn, and most prolifically on oracle bones (Qiu 1985). The discovery of oracle bones at Yinxu (the Ruins of Yin) in Anyang established the site as the last

The Manual Age

29

capital of the Shang Dynasty. The Shang people invented many musical instruments (Tong 1983) and made observations of planets and various comets (Xu, Stephenson, and Jiang 1995). The cast inscriptions on many large bronzes form much of the surviving early Chinese writing. They have helped historians and archaeologists piece together the history of China (Shaughnessy 1992). The Maykop people developed the most ancient stringed instrument, dating from the late fourth millennium BCE (Rezepkin 2000). The earliest examples of “Indus script,” which might not have formed a writing system, were dated around 3000 BCE. The Harappan people appeared to have used stringed musical instruments (Singh 2008). Minoan pottery, palace frescos, stone carvings, and intricately carved seal stones are the best-preserved surviving examples of Minoan art. The frescos include many depictions of people, often with bead necklaces, bracelets, and hair ornaments. The most famous fresco is the bull-leaping fresco. The Mycenaeans also painted entire scenes on their vessels depicting warriors, chariots, horses, and deities reminiscent of events in Homer's Iliad (Higgins and Morgan 1967; Dickinson 1994; Castleden 2005).

5.4. Governance and Empire Building The invention of writing and the increased productivity and better or new products brought about by the bronze tools made it possible to have more complex social structures and even larger empires. The expansion of Uruk trade networks from around 3600 BCE led to the establishment of many Uruk enclaves in Syria, Iran, Mesopotamia, and Anatolia. These enclaves had strong signs of government organization and grew into city-states that covered up to 250 acres (1 km²) and supported up to 10,000–20,000 people by the end of the period. The subsequent Jemdet Nasr period featured centralized buildings, administrative cuneiform tablets, and cylinder seals. Gilgamesh, the famous king of Uruk, is believed to have reigned in the Early Dynastic II period (2750–2600 BCE). The third Sumerian king of the First Dynasty of Lagash (approximately 2500–2300 BCE), Eannatum, established one of the first verifiable empires in history by conquering all of Sumer and some cities outside Sumer (Crawford 2004; Kramer 2010). In Egypt, the Naqada III culture (3200 to 3000 BCE) or Protodynastic Period was the period of state formation, culminating at the beginning of the Early Dynastic Period (3150–2686 BCE). King Narmer of Upper Egypt unified Lower and Upper Egypt c. 3100 BCE (Grimal 1994), established a national administration, and appointed royal governors. Cereal agriculture

30

Chapter 1

and centralization contributed to the success of the state for the next 800 years (Gardiner 1961; Shaw 2003). The Babylonian king Hammurabi (1792–1750 BCE) established a centralized government with an efficient bureaucracy and taxation, expanding Babylonian dominance over southern Mesopotamia. Hammurabi conquered in the east the pre-Iranic Elamites, Gutians, Lullubi, and Kassites, and in the west the Amorite Semitic states of the Levant (modern Syria and Jordan). One of the most important works of this “First Dynasty of Babylon” was compiling a code of laws, the Code of Hammurabi (Beaulieu 2018). The Minoans constructed their first palaces at the end of the Early Minoan period at Malia in the third millennium BCE. The palaces were centers of government, administrative offices, shrines, workshops, and storage spaces. The Palace of Knossos was the largest palace the Minoan people had ever constructed. It was about 150 meters across, with an area of some 20,000 square meters. The palace was laid out to surround the central court of the Minoans. The space surrounding the court contained rooms and hallways (Dickinson 1994). Toward the end of the Middle Bronze Age (1600 BCE), Mycenaeans formed several centers of power in southern mainland Greece, dominated by an elite warrior society. Around 1450 BCE, Mycenaeans from mainland Ancient Greece controlled Crete and colonized several other Aegean islands, reaching as far as Rhodes. Imposing palaces were built in the main Mycenaean centers. The focal point of a Mycenaean palace was the megaron, the throne room. The palatial territory was divided into several sub-regions (provinces) and smaller districts. The Mycenaean king, known as wanax, was the head of this society, the leading landlord, the spiritual and military leader, and an entrepreneur and trader aided by a network of high officials (Castleden 2005; Dickinson 1994). Erlitou in Yanshi, Henan, covers 100 ha (250 acres), with palaces built over different phases, bronze smelting workshops, and a population of between 18,000 and 30,000 at its peak (Liu and Xu 2007). The Erligang site in Zhengzhou was part of an ancient city surrounded by a roughly rectangular rammed earth wall with a perimeter of about seven kilometers. It was estimated that the walls would have been 20 meters wide at the base and eight meters high. Workshops for bone, pottery, and bronze vessels were found outside the city walls. In 1983, another Erligang culture site, a walled city built around 1600ௗBCE with an area of nearly 200 ha (490 acres) and pottery characteristic of the Erligang culture, was found six km northeast of

The Manual Age

31

the Erlitou site. It is now known as Yanshi Shang City and is identified by some scholars as the first Shang Dynasty capital, Western Bo (Liu and Chen 2012). Yinxu was the site of the last capital of China's Shang Dynasty, “Yin” (⇧) (1300–1046 BCE), which existed for 255 years and through the reign of 12 kings. It was located near the modern city of Anyang in Henan. At the beginning of the fourteenth century BCE, King Pangeng moved his capital from Yan (present-day Qufu, Shandong Province) to “Yin.” Yinxu is the largest archaeological site in China, covering an area of 30 km², and excavations have uncovered over 80 rammed-earth foundation sites, including palaces, shrines, tombs, and workshops. The Chariot Pits have the earliest samples of animal-driven carts discovered in China. Each of the six pits contains the remains of a carriage and two horses (Yang and Liu 1999). It is believed that the Shang Dynasty controlled and influenced a territory extending to the sea in the east, the present Shaanxi in the west, Hubei in the south, and Liaoning in the north, over more than one million km² (Turchin, Adams, and Hall 2006).

6. The Iron Age and the Beginning of Ancient History The Iron Age saw the prevalent use of iron. Its later arrival was due to the difficulty of extracting iron from oxidized iron ores. Iron smelting is more difficult due to the higher melting temperature required than tin and copper smelting. Conventionally the Iron Age is taken to end with the beginning of the historiographical records. Archeologists previously thought that iron production took place in Anatolia by 1200 BCE and then spread to other parts of the world. In the ancient Near East, the Iron Age was thought to last from 1200 BCE to 550 BCE, when the Achaemenid Empire, the First Persian Empire, was established by Cyrus the Great (Beaujard 2010; Stremlin 2008). The Iron Age started and ended much later in other regions, such as Central and Northern Europe. In China, recorded history appeared before iron tools replaced bronze tools. More recent findings indicate that iron’s systematic production and use in Anatolia began around 2000 BCE. Iron production technology might have been invented earlier and originated from multiple sites (Waldbaum 1978). Agriculture was still the dominant production technology in the Iron Age and Classical Antiquity. The main difference between the Bronze and Iron Ages was the widespread use of iron tools. In the Bronze Age, most farming tools were still made of stone and wood because bronze was rare and expensive. In the Iron Age, iron was widely used to make farming tools,

32

Chapter 1

which massively increased labor productivity, supported even larger state apparatus and war machines, and enabled larger wars and empires. At the same time, more people could make a living as intellectuals to spread knowledge, teach students, and advise secular rulers and religious leaders, which led to the emergence of different schools of thought. Different from the end of the Stone Age or Bronze Age, the end of the Iron Age did not lead to the adoption of tools made from new materials. It only marked the end of prehistory and the beginning of recorded human history. Iron, and later steel, continued to be the primary material for making tools up to modern times. The use of iron significantly increased labor productivity and enabled more division of labor and prosperity in literature, art, and technology. The historical record with reliable dates began during the Iron Age. Historians usually divide recorded human history into ancient history, the postclassical era, and modern history. Ancient history was the period from the beginning of recorded human history to the beginning of the postclassical era. Classical antiquity is a term for the period beginning with Homer's earliest-recorded epic Greek poetry (eighth–seventh centuries BCE) and ending with the close of late antiquity (250–750 AD).

6.1. The Spread of Iron-Making Technology The earliest use of iron in the Near East and Egypt was from meteoric iron. A bead found in Iran from the 5th millennium BCE (Photos 1989) and nine tiny beads found in Gerzeh in Egypt dated to 3200 BCE (Rehren et al. 2013) were made of meteoric iron. In the Mesopotamian states of Sumer, Akkad, and Assyria, iron smelting might have started as early as perhaps 3000 BCE. A dagger with a smelted iron blade was found in a Hattic tomb in Anatolia, dating from 2500 BCE (Bani-Hani, Abd-Allah, and El-Khouri 2012). The bloomery smelting of iron was an early method for producing iron from its ore. A bloomery is a furnace consisting of a pit or chimney with heatresistant walls made of earth, clay, or stone; near the bottom, one or more pipes enter through the sidewalls. The earliest bloomery smelting of iron was found at Tell el-Hammeh, Jordan, dating to around 930 BCE. The product from bloomery smelting was a porous mass of iron and slag called a bloom (sponge iron), which was iron produced from the direct reduction of iron ore (iron oxides) by a reducing gas (often carbon monoxide). The sponge iron was usually consolidated (shingled) and further forged into wrought iron. Shingling hammered a heated bloom into a more regularshaped piece with less slag. The spread of iron-working technology from the Late Bronze Age IIB to the Early Iron Age coincided with the Late

The Manual Age

33

Bronze Age Collapse when almost every city between Pylos and Gaza was violently destroyed (Knapp and Manning 2016). Iron metal was singularly scarce in Egyptian antiquities collections until Assyria's conquest. A sword bearing the name of Pharaoh Merneptah and a battle-ax with an iron blade and gold-decorated bronze shaft was excavated at Ugarit (Stech-Wheeler et al. 1981). An iron dagger with a golden hilt, among other iron objects found in the tomb of Tutankhamun, an Egyptian pharaoh of the 18th Dynasty, was meteoric iron (Comelli et al. 2016). Many iron-edged weapons were found at sites dating to the Greek Dark Ages (1100–800 BCE), the period from the Mycenaean civilization's end to the Greek poleis' first signs (Knapp and Manning 2016). The Iron Age began in Eastern Europe with the Koban, Chernogorovka, and Novocherkassk cultures from c. 900 BCE. In central Europe, it included the early Iron Age Hallstatt culture (800–450 BCE) and the late Iron Age La Tène culture (beginning in 450 BCE) (Collis 2003). Iron metallurgy was known in China by the ninth century BCE, but meteoric iron was used much earlier. A meteoric iron-bladed bronze tomahawk found at Taixicun in the Gaocheng of Hebei was dated to the middle Shang dynasty around the fourteenth century BC (Chung 1976). A sword excavated from the tomb of the Duke of Guo around 800 BCE at Sanmenxia city of Henan was made from smelted iron that had been shingled and carburized to become steel (Zhang and Tang 2017). The oldest objects made of smelted iron are two iron bars excavated from the Siwang culture at the Mogou site in Lintan of Gansu, dated between 1510 and 1310 BCE (Mei et al. 2015). Some researchers suggest that iron metallurgy in China came from the Near East and the Caucasus. Archeologists usually thought that the North India Iron Age succeeding the Late Harappan (Cemetery H) culture (1900–1300 BCE) lasted from 1200 to 300 BCE. Still, recent findings suggest that iron-making emerged there as early as 1800 BCE. Iron implements were found at many archaeological sites, such as Malhar, Dadupur, Raja Nala Ka Tila, and Lahuradewa, dated 1800–1200 BCE (Tewari 2003). The Kodumanal site (300 BCE–200 CE) in the Erode district of Tamil Nadu developed the crucible technique to produce high-quality steel by mixing and heating high-purity wrought iron, charcoal, and glass in the crucible (Juleff 1996). Metalsmithing in the Iron Age expanded from casting, the Bronze Age’s primary form, to forging. Weapons, utensils, and other implements were hammered into shape. Pure iron is softer than bronze, so pure or wrought

34

Chapter 1

iron tools wear out faster. The wide use of iron before the invention of steelmaking was due to cheaper production and the wide availability of iron ore. During the Iron Age, the best tools and weapons were those made from steel with a carbon content between 0.30% and 1.2% by weight. The carbon steel fragments found at Kaman-Kalehöyük and dated to 1800 BCE are the earliest known evidence of steelmaking worldwide (Akanuma 2008). The wide adoption of iron and steel coincided with the agricultural practice of slash-and-burn shifting cultivation. This agrarian technique involved cutting and burning plants in forests to create fields. The plots of land were cultivated temporarily, then abandoned, and allowed to revert to their natural vegetation while the cultivator moved on to another plot. The forest in Southern Europe had less capacity to regenerate after slash and burn because it mainly consisted of open evergreen leaves and pine forests. Most of the forests in the Mediterranean had disappeared by classical times (Hughes 2011).

6.2. Phoenician Alphabet and Written Languages Following the Bronze Age Collapse, the Phoenicians, originally from Canaan ports, dominated trade in the Mediterranean during 1200–800 BCE. Carthage (814–146 BCE) developed from a Phoenician colony into the center of an empire dominating the Mediterranean during the first millennium BCE (Miles 2011). The Proto-Canaanite alphabet (the Phoenician alphabet after 1050 BCE), derived from Egyptian hieroglyphs, was the predecessor of later Greek and Latin alphabets. People in Sinai and Canaan used it during the late Middle and Late Bronze Ages. The script became widely used with the rise of the Neo-Hittite kingdoms in the thirteenth and twelfth centuries BCE. The Ahiram epitaph engraved on the sarcophagus of Ahiram, a Phoenician king of Byblos, from 1000 BCE, was the earliest known example of the fully developed Phoenician alphabet. The earliest Aramaic alphabet was derived from the Phoenician alphabet. Aramaic developed its distinctive “square” style, and the ancient Israelites and other Canaan peoples adopted the alphabet to write their languages. It became the Hebrew alphabet of today. Aramaic language became the lingua franca of the Assyrian Empire from the eighth century BCE and the Achaemenid Empire (Howard 2014). The introduction of alphabetic characters and the consequent development of written language during the Iron Age enabled literature and historical records. The seventh-century BCE Neo-Assyrian copies of the Epic of Gilgamesh and the Enûma Eliš from Ashurbanipal's library in Nineveh and

The Manual Age

35

the Atra-Hasis are the best preserved most ancient works of Mesopotamian literature. Assyro-Babylonian literature written in the cuneiform Akkadian language continued into the Iron Age until the sixth century BCE (Fiore 1965). The Neo-Assyrian cuneiform script was the final stage, with the number of glyphs reduced and the glyph shapes standardized and simplified. Ancient Chinese writing script (the oracle bone script and its successor, the large seal script well preserved in the bronze inscriptions and less well in stone carvings) helped preserve ancient Chinese literature. The Book of Odes, the Book of Documents, and the Book of Changes (I Ching) were all believed to be works before the sixth century BCE. The Book of Odes has tremendously impacted Chinese literature and social communication up to modern times. The Book of Documents and the Book of Changes have also influenced later Chinese philosophy and political thinking.

6.3. Classical Antiquity Classical antiquity was a period of thriving innovations in production and intellectual activities. Its economic foundation was the increased productivity and new products following the widespread use of iron in making tools and goods. From Draco’s code and Solon’s reform to the death of Alexander the Great in 323 BCE, the classical period of Ancient Greece witnessed the thriving of philosophy, art, and Athenian democracy, the Greco-Persian Wars (499–449 BCE), and the Peace of Callias. The philosophies of the Milesian School, Xenophanes of Ionia, Pythagoreanism, the Eleatic School, Pluralism and Atomism, Sophists, Socrates, Plato, and Aristotle fundamentally influenced Western and modern philosophical thinking. The conquests of Alexander the Great (330–323 BCE) marked the beginning of the Hellenistic period, during which significant advances were made in science, arts, and architecture (Fox 2006). The economy of the Roman Republic was largely agrarian. The ancient trade in agricultural goods was well established, and ancient Rome had been a major center for agricultural trade. Rome received extensive grain shipments as tax payments by the second century BCE, and vast amounts of grain were transported, mainly by sea. The economy of the Roman Empire became monetized, and a basic banking system was formed. Trade routes stretched from Britain and Scandinavia in the west to India and China in the east, and major crops, spices, fabrics, and medicines were included (Fox 2006). Improved horse harnesses and the whippletree (to distribute force evenly) allowed draught animals to work longer, harder, and more efficiently

36

Chapter 1

(Brownrigg and Crouwel 2017). They became the primary power source for agriculture, transport, and warfare. Watermills were probably invented in the third century BCE. The Roman engineer Vitruvius had the first technical description (dated to 40 BCE) of a watermill. Water-driven grain pounders were used in the Greco-Roman world by the first century CE (Wilson 2002). Windmills were first invented in eastern Persia, as recorded by the Persian geographer Estakhri in the ninth century, although wind-driven wheels had been known to the Hellenic world and other countries much earlier (Lucas 2006). The historical record in China with reliable dates began in 841 BCE, marked by the period called the Gonghe Regency in the Zhou Dynasty (1046–256 BCE). During the Spring-Autumn period (770–476 BCE), hegemons arose, and Confucius (c. 551–479 BCE) started his school of thought while the first canals were built, and irrigation was used extensively. Guan Zhong (Kuan Chung, c. 720–c. 645 BCE), a politician and philosopher with a Legalist color, and Laozi (Lao Tse, sixth to fifth century BCE), the founder of Taoist philosophy, also had long-lasting influences on Chinese culture and political thinking. The ensuing Warring States period (475–221 BCE) was an era of intensive warfare and the development of many philosophical doctrines, later known as the Hundred Schools of Thought. Dujiangyan, built in 256 BCE by the State of Qin in present-day Sichuan, is still used to irrigate over 5,300 square kilometers of land. The Zhengguo Canal, built in 246 BCE in present-day Shaanxi to irrigate the Guanzhong plain, contributed to the strength of the Qin State in its wars to unify China (Willmott 1989). Quenching had been used to harden steel in the Warring States period. Though it was too expensive for mass production, multiplerefined steel produced by repeated hammering was an early steelmaking method (Wagner Donald 1993). The Qin Wars of Conquest led to the first unified Chinese empire, the Qin Dynasty (221–206 BCE). The Lingqu Canal, built in 214 BCE, connects the Xiang River (which flows north into the Yangtze) with the Li River (which flows south into the Gui River and Xijiang River). Locks were an essential component for shipping in canal systems. Flash locks, which consisted of an opening that could be quickly opened and closed in the dam, were used extensively in Ancient China (Li 2018; Needham 1965). The Qin Dynasty abolished the system of enfeoffment and adopted the centralized government system. It also standardized carriages, roads, Chinese script, and the units of length, volume, and weight, profoundly impacting China’s later economic development.

The Manual Age

37

The following Han Dynasty (206 BCE–220 CE) was considered a golden age in Chinese history. By the first century BCE, the Chinese had innovated the hydraulic-powered trip hammer for pounding, decorticating, and polishing grains. The square-pallet chain pump was used in China by the first century AD, which was usually powered by a waterwheel, pedals, or oxen pulling on a system of mechanical wheels. The chain pumps were mainly used to lift water from a lower to a higher elevation to fill irrigation canals and channels and provide water for urban and palatial pipe systems. During the Eastern Han Dynasty, the engineer Bi Lan constructed a series of square-pallet chain pumps outside the capital city of Luoyang to service Luoyang’s palaces and living quarters (Needham 1965). By the Han period, the entire plowshare was cast iron (Liu 2010). These heavy moldboard iron plows would slowly spread west and revolutionize farming in Northern Europe by the tenth century CE (White 1964). The invention of pulp papermaking had been credited to the Han court eunuch Cai Lun during the early second century CE. However, archaeological findings indicate much earlier paper use in China (Gunaratne 2001). The first century AD saw a method to create carbon-intermediate steel by melting wrought iron with cast iron (Needham 1971). Wheelbarrows were depicted in second-century tomb murals and brick tomb reliefs, and one of the tombs at Chengdu of Sichuan dated precisely to 118 AD. There were further innovations in the wheelbarrow during the Three Kingdoms period (Needham 1965). The births of Mahavira (c. 480–c. 408 BCE) and Buddha (c. 480–c. 400 BCE) marked the beginning of recorded history (with disputed dates) in South Asia (Dowling and Scarlett 2006). The “golden age” of classical Indian culture, as reflected in Sanskrit literature, began around 500 BCE. The great epics of Ramayana and Mahabharata were rooted in this classical period. The Gupta Empire (c. 320–550 AD) period saw extensive inventions and discoveries in science and technology and scholars such as Kalidasa, Aryabhata, Varahamihira, Vishnu Sharma, and Vatsyayana. Art, literature, religion, philosophy, and education thrived, creating magnificent architecture, sculptures, and paintings (Sen 1999). South Asia was estimated to have one of the largest economies in the world between the first and fifteenth centuries AD (Maddison 2001). The early classical antiquity (800–200 BCE) saw the emergence of great thinkers in different societies across various regions, Confucius, Laozi, Mahavira, Buddha, Socrates (c. 470–399 BCE), Pythagoras (c. 570–c. 495), Hebrew prophets, etc. The German philosopher Karl Jaspers coined the term

38

Chapter 1

the Axial Age (from the German Achsenzeit) to refer to this period during which most spiritual and religious traditions emerged in Eurasian Societies (Jaspers 1948, 1949). Although the concept has been widely debated and discussed (Mullins et al. 2018; Baumard, Hyafil, and Boyer 2015), we may view this period as an epoch with not only enough recorded knowledge for those great thinkers to reflect on the purpose of humanity as well as individual life but also the necessary wealth and technologies to spread their thoughts. Those great thinkers generally advocated a virtuous life by individuals, which was consistent with the ruling class’s interest and subsequently promoted by the rulers because honest people were easy to govern. The great thoughts and religions were externalizing and alienating people’s routine ideas and thinking processes. They were often used by the ruling classes to indoctrinate people in ways that the great thinkers themselves probably would disapprove of. They usually exerted a significant influence on people’s decisions and behavior.

7. From the Postclassical Era to the British Agricultural Revolution The development of the great world religions and the growth of trade networks between civilizations characterized the postclassical era. The early modern period continued the increase in global trade networks and witnessed the exploration and colonization of the Americas (Bentley 1996), which stimulated commerce, armaments, and shipbuilding, and increased the scope and variety of crops, raw materials, and consumer goods. Agriculture was still the dominant production technology, but iron-making, handicraft, cottage industry, etc., were improving their technologies and growing. Technological progress in these periods was incremental rather than revolutionary (Ma 2020); no advances were comparable to bronze and iron tools. The exchange of goods, plants, animals, and food crops between the Old and New Worlds greatly affected the human environment.

7.1. The Postclassical Era By the end of classical antiquity, the Manual Age's essential production technologies had largely been in place, so the postclassical era was a period of gradual improvement. The Islamic world was crucial in exchanging crops and technology between the European, Asian, and African continents. The extensive trade routes covered by the Muslim traders enabled the diffusion of many crops, plants, and farming techniques beyond and across the Islamic world. These factors contributed to the transformation of the

The Manual Age

39

agricultural practices in the medieval Islamic world from the eighth century, described by some as the “Arab Agricultural Revolution.” Several crops of major importance were introduced to Europe, along with the techniques for their cultivation. The concept of summer irrigation was also introduced to Europe from the Islamic world (Decker 2009). Intensive irrigation systems, crop rotation systems, and agricultural manuals were widely adopted in Medieval Europe. Norias, water mills, water-raising machines, dams, and reservoirs were used in irrigation systems (Lucas 2006). By 900 CE, developments in iron melting allowed for the increased production of iron, which was used to produce agricultural implements such as plows, hand tools, and horseshoes. The development of the moldboard plow, capable of turning over Northern Europe’s heavy, wet soils, led to the clearing of forests in that area and a significant increase in agricultural production. The increase in arable land led to a rise in population (White 1964). Crops grown by European farmers included wheat, rye, barley, oats, peas, beans, and vetches. Vetches became common from the thirteenth century as a fodder crop for animals and its nitrogen-fixation fertilizing properties. The Medieval Warm Period between 900 and 1300 AD led to increased European harvests. Crop yields peaked in the thirteenth century and stayed more or less steady until the eighteenth century. The technology of medieval agriculture was generally sufficient to meet the population’s needs under normal circumstances (Campbell and Overton 1993). The Chinese classical era was thought to have ended with the Han Dynasty. By the fourth century CE, as a medium for writing in China, paper had replaced silk sheets and bamboo and wooden slips. The earliest book copied on paper excavated so far was sections of The Records of the Three Kingdoms and dated to the Jin Dynasty (265–420 CE) (Guo 1972). The stirrup was invented in China, and the first use of paired stirrups was credited to the Jin Dynasty, but it might have been invented much earlier (Mahdihassan 1975). A stirrup is a light frame or ring that holds a rider’s foot, attached to the saddle by a strap, to aid in mounting and as a support while riding an animal. They increase the rider's ability to control the animal and thus its usefulness. Some have viewed the stirrup as one essential tool used to create and spread modern civilization, possibly as important as the wheel or printing press. It was spread westward through the nomadic peoples of Central Eurasia and to Europe during the Middle Ages(White 1964; Baber 1996). The Tang Dynasty (618-907 CE) is regarded as a high point in Chinese civilization and cosmopolitan culture. Its capital Chang'an, the present-day

40

Chapter 1

Xi'an, was the most populous city in the world. The Tang era is also considered the greatest age for Chinese poetry. Woodblock printing, which originated in the Han Dynasty (before 220 CE) as a method to print initially on textiles and later on paper, was used for printing books. A relief image of an entire page was carved into blocks of wood, which were then inked and used to print copies of that page. The Diamond Sutra, printed in 868 CE, was considered the oldest book by this method. Still, the Dharani Sutra, found in the Bulguk-sa Temple compound in Kyongju, Korea, was printed between 684 and 751 CE and is now considered the oldest (Park 2014). The Song Dynasty (960-1279) was the first to issue real paper money nationally. It also saw the first known use of gunpowder, the first discernment of true north using a compass, and the first use of the more efficient pound locks. The Song economy was one of the most prosperous and advanced in the medieval world. A predecessor to the Bessemer process of making steel was described by the prolific scholar and polymath Shen Kuo (1031–1095) in 1075. It involved a “partial decarbonization” method of repeated cast iron forging under a cold blast (Hartwell 1966). Bi Sheng made a movable type of earthenware c. 1045. Metal movable type was first used in Korea, and the oldest extant movable metal print book, Jikji, was printed in Korea in 1377. Around 1450 Johannes Gutenberg made a mechanical metal movable-type printing press in Europe (Park 2014). Printing made knowledge more accessible to people than when books were hand-copied. When books were rare and expensive, only those with a good memory could become a scholar. Books were an externalization and scaleup of human memory and speech. After nearly 100 years of Mongolian rule under the Yuan Dynasty (12711368), the Ming Dynasty (1368–1644) in China was described by some as “one of the greatest eras of orderly government and social stability in human history” (Fan 2016). The Ming fleet led by Admiral Zheng He (1371–1433 or 1435) visited Southeast Asia, South Asia, Western Asia, and East Africa during his seven expeditions from 1405 to 1433. His larger ships stretched 120 meters in length and carried hundreds of sailors on four tiers of decks (Bowring 2018). Some scholars think Zheng He’s fleet passed the Cape of Good Hope and arrived in the Americas (Steele 2005; Lee 2011, 2015).

7.2. The Early Modern Era The early modern era of modern history spanned the period from the end of the postclassical era (c. 1500) to the beginning of the Age of Revolutions (c. 1800). The Renaissance in the 15th and 16th centuries marked the transition

The Manual Age

41

between the post-classical era and the modern era, characterized by an effort to revive and surpass ideas and achievements of classical antiquity. The Age of Discovery, with the voyages of Christopher Columbus beginning in 1492, started the exchange between the Old World and the New World. This process caused misery and tragedy for humanity while significantly increasing the wealth of many societies. After 1492, a global exchange of previously local crops and livestock breeds occurred, and Europeans found more arable lands and a cheaper labor force for their agricultural and commercial explorations. The early modern era prepared the way for the commercial foundation of the Industrial Revolution (Bentley 1996). The initial drive for global exploration in the fifteenth century was to find new routes for trading with the Far East. From the eighth century until then, the Republic of Venice and neighboring maritime republics monopolized European trade with the Middle East. At the same time, Muslim traders dominated maritime routes throughout the Indian Ocean and trade between the Middle East and the Far East. The silk and spice trade made these Mediterranean city-states phenomenally rich. With the rise of the Ottoman Empire that eventually led to the fall of Constantinople in 1453, Europeans were barred from the combined-land-sea routes between Europe and the Far East (Arnold 2002). The Portuguese began systematically exploring the Atlantic coast of Africa in 1418, under the sponsorship of Prince Henry the Navigator, which led to the Portuguese discovery of the Atlantic archipelagos of Madeira in 1419 and the Azores (which appeared in the Atlas Catalan in the fourteenth century) in 1427, and the coast of Africa. Eventually, in 1488, Bartolomeu Dias rounded the southern tip of Africa and reached the mouth of the Great Fish River, proving that the Indian Ocean was accessible from the Atlantic. Vasco da Gama reached India by sailing around Africa, opening up direct trade with Asia in 1498 (Diffie and Winius 1977). The Portuguese dominated this spice trade route for the next 100 years. Pedro Álvares Cabral found South America on his way to Asia in 1500. The Portuguese sailed to the Spice Islands in 1512 and landed in China one year later. In 1513, the Spaniard Vasco Núñez de Balboa reached the Pacific Ocean from the New World by crossing the Isthmus of Panama. In 1522, a Castilian (Spanish) expedition, led by Portuguese navigator Ferdinand Magellan and later by the Spanish-Basque navigator Juan Sebastián Elcano, completed the world’s first circumnavigation by sailing westward from Europe (Diffie and Winius 1977). The route for direct trade between Asia and the New World opened up.

42

Chapter 1

Several crops, including maize, potatoes, sweet potatoes, and manioc (cassava) from the New World, have played critical roles in the social development of the Old-World societies. The high yields and adaptability to less fertile lands of those crops caused population growth worldwide even before the Industrial Revolution and had a lasting effect on many cultures (Crosby 2003). The potato had become an important staple crop throughout Europe by the late 1700s. The potato allowed farmers to produce more food, and the population increased food consumption. The nutrition boost caused by increased potato consumption lowered disease rates, increased birth rates, and decreased mortality rates throughout the British Empire, North America, and Europe (Chapman 2000). The first intensive use of fertilizer, guano, imported to Europe from Peru, followed the introduction of the potato. The first artificial pesticide, an arsenic compound, was used to fight Colorado potato beetles (Mann 2011). Portuguese traders introduced maize and cassava from Brazil into Africa in the sixteenth century, which have become critical staple foods for Africans. Spanish colonizers introduced maize and sweet potatoes to Asia from the Americas, contributing to population growth in Asia. Tomatoes came to Europe from the New World via Spain. From the nineteenth century, tomato sauces became a typical ingredient of Italian food, especially in Neapolitan cooking. Peanuts from the New World became an important source of protein and oil in Old World countries. Chili peppers introduced to the Old World had transformed cuisines in many cultures. Several varieties of wheat, barley, rice and soybeans, and turnips went from the Old World to the New World. Coffee from the Old World became an essential commodity in the New World (Crosby 2003). The international trade between continents also resulted in new forms of economic organization. Because voyages to get commodities from other continents were high-risk ventures with potentially high returns, investors formed companies to spread the risk. Initially, it was customary to set up a company only for a single voyage and liquidate the company upon the fleet’s return. Since pooling risk over time might further spread and better manage the risk of mismatch between an inelastic demand and an elastic supply, it implied a need to form a cartel to control the supply. The English were the first to adopt this approach by creating their monopoly enterprise, the English East India Company, as a joint-stock company in 1600. English trade expeditions to the Far East started on 10 April 1591 with three ships, one of which reached the Malay Peninsula. The English East India Company obtained its royal charter on 31 December 1600 (Farrington 2002).

The Manual Age

43

The direct Dutch involvement in spice trade with the Far East began in 1595 with Cornelis de Houtman’s four-ship exploratory expedition to Banten, the main pepper port of West Java, followed by an increasing number of fleets sent out by competing merchant groups from 1598 onwards. The United East Indian Company, usually known as the Dutch East India Company, was established as a chartered company in 1602. The Dutch government granted it a 21-year monopoly on the Dutch spice trade. It set up the Amsterdam Stock Exchange for dealing printed stocks and bonds and, in 1602, issued the first tradable shares. This invention enhanced the ability of joint-stock companies to attract capital from investors as they could now quickly dispose of their claims. It became the first multinational corporation and megacorporation (Gelderblom, De Jong, and Jonker 2013).

7.3. The British Agricultural Revolution The British Agricultural Revolution was the unprecedented increase in agricultural production in Britain between the mid-seventeenth and late nineteenth centuries, which subsequently affected agriculture worldwide. Major underlying innovations included enclosure, improvement in farming tools, crop rotation, land conversion, and selective breeding. England’s wheat output per acre went up from about 19 bushels in 1720 to 21–22 bushels by the middle of the century and around 30 bushels by 1840 (Mingay 1977). One bushel in the British Imperial System equals 26.37 liters. The enclosure of farmlands, which occurred sparsely from the thirteenth century, was accelerated in the eighteenth century with special Acts of Parliament to expedite the legal process (Neeson 2000). The consolidation of private ownership encouraged land improvement. The four-course crop rotation (wheat, turnips, barley, and clover) opened up fodder and grazing crops, allowing livestock to be bred year-round. Cereal crop yields increased as nitrogen-rich manure and nitrogen-fixing crops increased the available nitrogen in the soil (Mingay 1977). Improving farming tools also played a vital role in the British Agricultural Revolution. The Dutch acquired the iron-tipped, curved moldboard adjustable depth plow from the Chinese in the early seventeenth century, which one or two oxen could pull compared to the six or eight needed for the heavy-wheeled northern European plow. The Dutch plow was highly successful in Britain on wet, boggy soil and ordinary land. Joseph Foljambe's cast iron plow (patented in 1730) combined an earlier Dutch design with several innovations, making it easier to pull and more

44

Chapter 1

controllable. Land maintenance advancements in Flanders and the Netherlands were introduced to Britain. Land conversion and improved plows increased farming productivity and the amount of arable land. Jethro Tull's 1731 invention of a horse-drawn seed drill and horse hoe (the small plow to hoe between crop rows) revolutionized planting in Britain. Andrew Meikle patented the first practical threshing machine in 1784 (Fussell 1981). Robert Bakewell and Thomas Coke introduced selective livestock breeding as a scientific practice. Using native stock, Bakewell could quickly select for large yet fine-boned sheep with long lustrous wool. The average weight of a bull sold for slaughter at Smithfield was reported around 1700 as 370 pounds (170 kg), but bulls with a weight of 840 pounds (380 kg) were recorded by 1786 (Thomas 2005). The agricultural revolution in Britain proved to be a milestone in history. The increased output per acre hence per farmer, enabled farmers to exchange for more non-farm goods, enhancing the aggregate demand for textiles and other commodities, stimulating innovation and investment in the textile industry and other relevant sectors. The improved production per farmer increased the population and the available workforce, creating the labor force needed by the Industrial Revolution to sustain the country's rise to industrial pre-eminence.

8. Summary The progress of human society is reflected and underlaid by productivity increases, which in turn are caused by new tools and techniques in production and new products. During the Manual Age, technological progress was primarily to increase the efficiency of manual labor with tools that would enhance or save forces exerted by muscle power. Draught animals, sails, windmills, and watermills could replace some human manual work. Because of inefficient communication and transport tools, new tools and new techniques often took hundreds of thousands of years to spread across continents during the Paleolithic Age. The Neolithic agricultural revolution brought farming and husbandry to human society and enabled humanity to form large communities while new technologies could spread across continents in a few hundred years. The Bronze Age witnessed a sharp increase in productivity because of the more efficient bronze tools, and agricultural output became sufficient to support a more extensive bureaucracy and artisans. Thus, states formed to manage many laborers so that nonfarming workers could produce more tools, crafts, arts, and knowledge.

The Manual Age

45

The wide availability of iron tools further increased labor productivity and provided the economic foundation for the thriving period of classical antiquity, from which many of our contemporary philosophies and knowledge originated. From the end of classical antiquity to the beginning of the British Agricultural Revolution, human society underwent a normal growth stage to fully utilize iron-based technologies before the next revolutionary change in the dominant production technology (Ma 2019, 2020). Together with opening up the New World and the Far East for resources and markets during the Age of Discovery, the inventions of complicated tools and their applications to utilize non-biological power sources and increase productivity during this period prepared for the Industrial Revolution.

References Adams, R Mc. 1983. “The Jarmo stone and pottery vessel industries.” In Prehistoric Archaeology along the Zagros Flanks, edited by Linda S. Braidwood, 209-232. Chicago, IL: The University of Chicago Oriental Institute. Adovasio, James M, Olga Soffer, and Bohuslav Klima. 1996. “Upper Palaeolithic fibre technology: interlaced woven finds from Pavlov I, Czech Republic, c. 26,000 years ago.” Antiquity 70 (269):526-534. Akanuma, Hideo. 2008. “The Significance of Early Bronze Age Iron Objects from Kaman-Kalehöyük, Turkey.” Anatolian Archaeological Studies 17:313-320. Allen, James P. 2000. Middle Egyptian: An introduction to the language and culture of hieroglyphs. Cambridge: Cambridge University Press. Allen, Jim. 1970. “Prehistoric agricultural systems in the Wahgi Valley-a further note.” The Australian Journal of Anthropology 7 (3):177. Ammerman, Albert J, and Luigi L Cavalli-Sforza. 1971. “Measuring the rate of spread of early farming in Europe.” Man 6 (4):674-688. Anthony, David W, Peter Bogucki, Eugen Comúa, Marija Gimbutas, Borislav Jovanoviü, JP Mallory, and Sarunas Milisaukas. 1986. “The ‘Kurgan Culture, ’ Indo-European origins, and the domestication of the horse: a reconsideration [and comments and replies].” Current Anthropology 27 (4):291-313. Anthony, David W, and Dorcas R Brown. 2011. “The secondary products revolution, horse-riding, and mounted warfare.” Journal of World Prehistory 24 (2-3):131. Antón, Susan C. 2003. “Natural history of Homo erectus.” American Journal of Physical Anthropology 122 (S37):126-170.

46

Chapter 1

Armand, J. 1985. “The emergence of the handaxe tradition in Asia, with special reference to India.” In Recent advances in Indo-Pacific prehistory, edited by V. N. Misra and Peter S. Bellwood, 3-8. Delhi: Oxford & IBH Publishing Co. Arnold, David. 2002. The Age of Discovery, 1400-1600. London and New York: Routledge. Aubert, Maxime, Adam Brumm, Muhammad Ramli, Thomas Sutikna, E Wahyu Saptomo, Budianto Hakim, Michael J Morwood, Gerrit D van den Bergh, Leslie Kinsley, and Anthony Dosseto. 2014. “Pleistocene cave art from Sulawesi, Indonesia.” Nature 514 (7521):223-227. Audouze, Françoise. 1987. “The Paris basin in Magdalenian times.” In The Pleistocene Old World, edited by Olga Soffer, 183-200. New York and London: Springer. Aurenche, Olivier, Philippe Galet, Emmanuelle Régagnon-Caroline, and Jacques Évin. 2001. “Proto-Neolithic and Neolithic cultures in the Middle East—the birth of agriculture, livestock raising, and ceramics: A calibrated 14 C chronology 12,500-5500 cal BCE.” Radiocarbon 43 (3):1191-1202. Avari, Burjor. 2016. India: The ancient past: A history of the Indian subcontinent from C. 7000 BCE to CE 1200. London and New York: Routledge. Baber, Zaheer. 1996. The science of empire: Scientific knowledge, civilization, and colonial rule in India. Albany, NY: SUNY Press. Backwell, Lucinda, Justin Bradfield, Kristian J Carlson, Tea Jashashvili, Lyn Wadley, and Francesco d'Errico. 2018. “The antiquity of bow-andarrow technology: evidence from Middle Stone Age layers at Sibudu Cave.” Antiquity 92 (362):289-303. Backwell, Lucinda, Francesco d'Errico, and Lyn Wadley. 2008. “Middle stone age bone tools from the Howiesons Poort layers, Sibudu Cave, South Africa.” Journal of Archaeological Science 35 (6):1566-1580. Bani-Hani, Manar, Ramadan Abd-Allah, and Lamia El-Khouri. 2012. “Archeaometallurgical finds from Barsinia, Northern Jordan: Microstructural characterization and conservation treatment.” Journal of Cultural Heritage 13 (3):314-325. Bar-Yosef, Ofer. 1989. “The PPNA in the Levant–an overview.” Paléorient 15 (1):57-63. BarဨYosef, Ofer. 1998. “The Natufian culture in the Levant, threshold to the origins of agriculture.” Evolutionary Anthropology: Issues, News, and Reviews 6 (5):159-177.

The Manual Age

47

Barber, Elizabeth Jane Wayland. 1991. Prehistoric textiles: the development of cloth in the Neolithic and Bronze Ages with special reference to the Aegean. Princeton, NJ: Princeton University Press. Batiuk, Stephen D. 2013. “The fruits of migration: Understanding the ‘longue duree’ and the socio-economic relations of the Early Transcaucasian Culture.” Journal of Anthropological Archaeology 32 (4):449-477. Baumard, Nicolas, Alexandre Hyafil, and Pascal Boyer. 2015. “What changed during the axial age: Cognitive styles or reward systems?” Communicative & Integrative Biology 8 (5):e1046657. Beaujard, Philippe. 2010. “From three possible Iron-Age world-systems to a single Afro-Eurasian world-system.” Journal of World History 21 (1):1-43. Beaulieu, Paul-Alain. 2018. A History of Babylon, 2200 BC-AD 75. Chichester, UK: John Wiley & Sons. Bednarik, Robert G. 2014. “The beginnings of maritime travel.” Advances in Anthropology 4 (04):209. Bellwood, Peter. 1991. “The Austronesian dispersal and the origin of languages.” Scientific American 265 (1):88-93. Bellwood, Peter. 2004. First Farmers: The Origins of Agricultural Societies. Malden, MA: Wiley-Blackwell. Bentley, Jerry H. 1996. “Cross-cultural interaction and periodization in world history.” The American Historical Review 101 (3):749-770. Blackham, Mark. 1996. “Further investigations as to the relationship of Samarran and Ubaid ceramic assemblages.” Iraq 58:1-15. Boaretto, Elisabetta, Xiaohong Wu, Jiarong Yuan, Ofer Bar-Yosef, Vikki Chu, Yan Pan, Kexin Liu, David Cohen, Tianlong Jiao, and Shuicheng Li. 2009. “Radiocarbon dating of charcoal and bone collagen associated with early pottery at Yuchanyan Cave, Hunan Province, China.” Proceedings of the National Academy of Sciences 106 (24):9595-9600. Boaz, Noel Thomas, T Noel, Russell L Ciochon, Russell Ciochon, and L Russell. 2004. Dragon bone hill: an Ice-Age saga of Homo Erectus. Oxford: Oxford University Press. Boëda, Eric, Jean-Michel Geneste, Christophe Griggo, Norbert Mercier, Sultan Muhesen, Jean Louis Reyss, Ahmed Taha, and Hélene Valladas. 1999. “A Levallois point embedded in the vertebra of a wild ass (Equus africanus): hafting, projectiles and Mousterian hunting weapons.” Antiquity 73 (280):394-402. Böhme, Madelaine, Nikolai Spassov, Martin Ebner, Denis Geraads, Latinka Hristova, Uwe Kirscher, Sabine Kötter, Ulf Linnemann, Jérôme Prieto,

48

Chapter 1

and Socrates Roussiakis. 2017. “Messinian age and savannah environment of the possible hominin Graecopithecus from Europe.” PLoS One 12 (5). Bowring, Philip. 2018. Empire of the Winds: The Global Role of Asia’s Great Archipelago. London: Bloomsbury Publishing. Boyd, Brian, and Jill Cook. 1993. “A reconsideration of the ‘Ain Sakhri’ figurine.” Proceedings of the Prehistoric Society 59:399-405. Brownrigg, Gail, and Joost Crouwel. 2017. “Developments in Harnessing and Draught in the Roman World.” Oxford Journal of Archaeology 36 (2):197-220. Brunton, Guy, and Gertrude Caton-Thompson. 1928. The Badarian civilisation and predynastic remains near Badari. London: British School of Archaeology in Egypt, University College. Butlin, Noel George. 1983. Our original aggression: Aboriginal populations of southeastern Australia, 1788-1850. Sydney and Boston: Allen & Unwin. Campbell, Bruce MS, and Mark Overton. 1993. “A new perspective on medieval and early modern agriculture: six centuries of Norfolk farming c. 1250-c. 1850.” Past & Present 141 (1):38-105. Carbone, Lucia, R Alan Harris, Sante Gnerre, Krishna R Veeramah, Belen Lorente-Galdos, John Huddleston, Thomas J Meyer, Javier Herrero, Christian Roos, and Bronwen Aken. 2014. “Gibbon genome and the fast karyotype evolution of small apes.” Nature 513 (7517):195-201. Carter, Robert A. 2006. “Boat remains and maritime trade in the Persian Gulf during sixth and fifth millennia BC.” Antiquity. 80 (307):52-63. Castleden, Rodney. 2005. The Mycenaeans. London and New York: Routledge. Cauvin, Jacques. 2000. The Birth of the Gods and the Origins of Agriculture. Translated by Trevor Watkins. Cambridge: Cambridge University Press. Cauvin, Jacques. 2002. “The symbolic foundations of the Neolithic Revolution in the Near East.” In Life in Neolithic farming communities, edited by Ian Kuijt, 235-252. New York: Springer. Chang, Kwang-chih. 1986. The archaeology of ancient China. New Haven, CT: Yale University Press. Chapman, Jeff. 2000. “The impact of the potato.” History Magazine 1 (2):41-44. Chen, Honghai. 2013. “The Qijia culture of the upper Yellow River valley.” In A Companion to Chinese Archaeology, edited by Anne P. Underhill, 105-24. Oxford: Wiley-Blackwell. Chung, Li. 1976. “Studies on the iron blade of a Shang dynasty bronze axe unearthed at Kao-cheng.” Acta Archaeologia Sinica (2):17-34.

The Manual Age

49

Colledge, Sue, James Conolly, Stephen Shennan, Peter Bellwood, Laurent Bouby, Julie Hansen, DavidR Harris, Konstantinos Kotsakis, Mehmet zdoan, and Edgar Peltenburg. 2004. “Archaeobotanical evidence for the spread of farming in the Eastern Mediterranean.” Current Anthropology 45 (S4): S35-S58. Collis, John. 2003. The European iron age. London and New York: Routledge. Comelli, Daniela, Massimo D'orazio, Luigi Folco, Mahmud ElဨHalwagy, Tommaso Frizzi, Roberto Alberti, Valentina Capogrosso, Abdelrazek Elnaggar, Hala Hassan, and Austin Nevin. 2016. “The meteoritic origin of Tutankhamun's iron dagger blade.” Meteoritics & Planetary Science 51 (7):1301-1309. Copeland, Lorraine. 2000. “Forty-six Emireh points from the Lebanon in the context of the Middle to Upper Paleolithic transition in the Levant.” Paléorient 62:73-92. Cote, Susanne M. 2004. “Origins of the African hominoids: an assessment of the palaeobiogeographical evidence.” Comptes Rendus Palevol 3 (4):323-340. Crawford, Harriet. 2004. Sumer and the Sumerians. Cambridge: Cambridge University Press. Crochet, Jean-Yves, Jean-Loup Welcomme, Jerome Ivorra, Gilles Ruffet, Nicolas Boulbes, Ramon Capdevila, Julien Claude, Cyril Firmat, Gregoire Metais, and Jacques Michaux. 2009. “A new vertebrate fauna associated with lithic artefacts from the Early Pleistocene of the Herault Valley (southern France) dated around 1.57 Ma.” Comptes Rendus Palevol 8 (8):725-736. Crosby, Alfred W. 2003. The Columbian exchange: biological and cultural consequences of 1492. Vol. 2. West Point, CT and London: Praeger. Darvill, Timothy. 2003. The Concise Oxford Dictionary of Archaeology. Oxford: Oxford University Press. Dayan, Tamar. 1994. “Early domesticated dogs of the Near East.” Journal of Archaeological Science 21 (5):633-633. De Heinzelin, Jean, J Desmond Clark, Tim White, William Hart, Paul Renne, Giday WoldeGabriel, Yonas Beyene, and Elisabeth Vrba. 1999. “Environment and behavior of 2.5-million-year-old Bouri hominids.” Science 284 (5414):625-629. Decker, Michael. 2009. “Plants and progress: rethinking the Islamic agricultural revolution.” Journal of World History 20 (2):187-206. Demoule, Jean-Paul, and Catherine Perlès. 1993. “The Greek Neolithic: a new review.” Journal of World Prehistory 7 (4):355-416.

50

Chapter 1

Dennell, Robin W, H Rendell, and Ernie Hailwood. 1988. “Early toolmaking in Asia: two-million-year-old artefacts in Pakistan.” Antiquity 62 (234):98-106. Dickinson, Oliver. 1994. The Aegean Bronze Age. Cambridge: Cambridge University Press. Diffie, Bailey Wallys, and George D. Winius. 1977. Foundations of the Portuguese empire, 1415-1580. Vol. 1. Minneapolis, MN: University of Minnesota Press. DiMaggio, Erin N, Christopher J Campisano, John Rowan, Guillaume Dupont-Nivet, Alan L Deino, Faysal Bibi, Margaret E Lewis, Antoine Souron, Dominique Garello, and Lars Werdelin. 2015. “Late Pliocene fossiliferous sedimentary record and the environmental context of early Homo from Afar, Ethiopia.” Science 347 (6228):1355-1359. Donnan, Christopher B. 1973. “A Precolumbian smelter from Northern Perú.” Archaeology 26 (4):289-297. Dowling, Elizabeth M, and W George Scarlett. 2006. Encyclopedia of religious and spiritual development. Thousand Oaks, London and New Dehli: Sage. Druzhkova, Anna S, Olaf Thalmann, Vladimir A Trifonov, Jennifer A Leonard, Nadezhda V Vorobieva, Nikolai D Ovodov, Alexander S Graphodatsky, and Robert K Wayne. 2013. “Ancient DNA analysis affirms the canid from Altai as a primitive dog.” PLoS One 8 (3):e57754. Dutton, Andrew, Peter J Fasham, DA Jenkins, AE Caseldine, and Sheila Hamilton-Dyer. 1994. “Prehistoric copper mining on the Great Orme, Llandudno, Gwynedd.” Proceedings of the Prehistoric Society 60 (1):245-286. Edens, Christopher. 1995. “Transcaucasia at the end of the Early Bronze Age.” Bulletin of the American Schools of Oriental Research 299 (1):5364. Edwards, Phillip C. 2016. “The chronology and dispersal of the Pre-Pottery Neolithic B cultural complex in the Levant.” Paléorient 42 (2):53-72. Edwards, Phillip C, John Meadows, Ghattas Sayej, and Michael Westaway. 2004. “From the PPNA to the PPNB: new views from the Southern Levant after excavations at Zahrat adh-Dhra'2 in Jordan.” Paléorient 30 (2):21-60. Elliott, Benjamin. 2015. “Facing the chop: redefining British antler mattocks to consider larger-scale maritime networks in the early fifthmillennium cal BC.” European Journal of Archaeology 18 (2):222-244. Engel, Frederic. 1970. “Exploration of the Chilca Canyon, Peru.” Current Anthropology 11 (1):55-58.

The Manual Age

51

Evans, Arthur. 2013. The Palace of Minos: a comparative account of the successive stages of the early Cretan civilization as illustrated by the discoveries at Knossos. Cambridge: Cambridge University Press. Fan, C Simon. 2016. Culture, institution, and development in China: the economics of national character. London and New York: Routledge. Farrington, Anthony. 2002. Trading Places: The East India Company and Asia 1600-1834. London: British Library. Fiore, Silvestro. 1965. Voices from the clay: the development of AssyroBabylonian literature. Norman, OK: University of Oklahoma Press. Fleisch, Henri. 1966. “Notes de préhistoire libanaise.” Bulletin de la Société Préhistorique Française. Comptes Rendus des Séances Mensuelles 63 (7):239-242. Fox, Robin Lane. 2006. The Classical World: An Epic History of Greece and Rome. London: Penguin UK. Fuss, Jochen, Nikolai Spassov, David R Begun, and Madelaine Böhme. 2017. “Potential hominin affinities of Graecopithecus from the Late Miocene of Europe.” PLoS One 12 (5):e0177127. Fussell, George Edwin. 1981. The Farmer's Tools: The History of British Farm Implements, Tools and Machinery AD 1500-1900. London: Orbis Publishing. Gabunia, Leo, Abesalom Vekua, David Lordkipanidze, Carl C Swisher, Reid Ferring, Antje Justus, Medea Nioradze, Merab Tvalchrelidze, Susan C Antón, and Gerhard Bosinski. 2000. “Earliest Pleistocene hominid cranial remains from Dmanisi, Republic of Georgia: taxonomy, geological setting, and age.” Science 288 (5468):1019-1025. Garcia, Tristan, Gilbert Féraud, Christophe Falguères, Henry de Lumley, Christian Perrenoud, and David Lordkipanidze. 2010. “Earliest human remains in Eurasia: new 40Ar/39Ar dating of the Dmanisi hominidbearing levels, Georgia.” Quaternary Geochronology 5 (4):443-451. Gardiner, Alan Henderson. 1961. Egypt of the Pharaohs: an Introduction. Vol. 165. Oxford: Oxford University Press. Garfinkel, Yosef. 1993. “The Yarmukian Culture in Israel.” Paléorient 19 (1):115-134. Garfinkel, Yosef. 1999a. “Radiometric dates from eighth millennium BP Israel.” Bulletin of the American Schools of Oriental Research 315 (1):113. Garfinkel, Yosef. 1999b. The Yarmukians: Neolithic Art from Shaұar Hagolan. Jerusalem: Bible Lands Museum. Garfinkel, Yosef, and Dani Nadel. 1989. “The Sultanian flint assemblage from Gesher and its implications for recognizing early Neolithic entities in the Levant.” Paléorient 15 (2):139-151.

52

Chapter 1

Gelderblom, Oscar, Abe De Jong, and Joost Jonker. 2013. “The formative years of the modern corporation: The Dutch East India Company VOC, 1602–1623.” The Journal of Economic History 73 (4):1050-1076. Germonpré, Mietje, Mikhail V Sablin, Rhiannon E Stevens, Robert EM Hedges, Michael Hofreiter, Mathias Stiller, and Viviane R Després. 2009. “Fossil dogs and wolves from Palaeolithic sites in Belgium, Ukraine and Russia: osteometry, ancient DNA and stable isotopes.” Journal of Archaeological Science 36 (2):473-490. Gorlinski, Gini. 2012. The history of agriculture. New York: Britannica Educational Publishing. Grimal, Nicolas. 1994. A history of ancient Egypt. Oxford: Wiley-Blackwell. Gunaratne, Shelton A. 2001. “Paper, printing and the printing press: A horizontally integrative macrohistory analysis.” Gazette (Leiden, Netherlands) 63 (6):459-479. Guo, Moruo. 1972. “The remnant of the Three Kingdoms written by Jin people newly unearthed in Xinjiang.” Cultural Relics (8):2-6. Gupta, Anil K. 2004. “Origin of agriculture and domestication of plants and animals linked to early Holocene climate amelioration.” Current Science-Bangalore 87 (1):54-59. Harmand, Sonia, Jason E Lewis, Craig S Feibel, Christopher J Lepre, Sandrine Prat, Arnaud Lenoble, Xavier Boës, Rhonda L Quinn, Michel Brenet, and Adrian Arroyo. 2015. “3.3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenya.” Nature 521 (7552):310-315. Hartwell, Robert. 1966. “Markets, technology, and the structure of enterprise in the development of the eleventh-century Chinese iron and steel industry.” The Journal of Economic History 26 (1):29-58. Haviland, William A, Harald EL Prins, Dana Walrath, and Bunny McBride. 2012. The essence of anthropology. Belmont, CA: Cengage Learning. Hedges, S Blair, and Sudhir Kumar. 2003. “Genomic clocks and evolutionary timescales.” Trends in Genetics 19 (4):200-206. Higgins, Reynold Alleyne, and Lyvia Morgan. 1967. Minoan and Mycenaean art. London: Thames and Hudson Hill, Andrew, and Steven Ward. 1988. “Origin of the Hominidae: the record of African large hominoid evolution between 14 My and 4 My.” American Journal of Physical Anthropology 31 (S9):49-83. Hiscock, Peter. 2007. Archaeology of ancient Australia. London and New York: Routledge. Hoffmann, Dirk L, Chris D Standish, Marcos García-Diez, Paul B Pettitt, James A Milton, João Zilhão, José Javier Alcolea-González, Pedro Cantalejo-Duarte, Hipólito Collado, and Rodrigo De Balbín. 2018. “U-

The Manual Age

53

Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art.” Science 359 (6378):912-915. Horwitz, Liora Kolska, Eitan Tchernov, Pierre Ducos, Cornelia Becker, Angela Von Den Driesch, Louise Martin, and Andrew Garrard. 1999. “Animal domestication in the southern Levant.” Paléorient 25 (2):6380. Hou, Yamei, Richard Potts, Baoyin Yuan, Zhengtang Guo, Alan Deino, Wei Wang, Jennifer Clark, Guangmao Xie, and Weiwen Huang. 2000. “MidPleistocene Acheulean-like stone technology of the Bose basin, South China.” Science 287 (5458):1622-1626. Howard, Michael C. 2014. Transnationalism in ancient and medieval societies: The role of cross-border trade and travel. Jefferson, NC: McFarland. Hublin, Jean-Jacques, Abdelouahed Ben-Ncer, Shara E Bailey, Sarah E Freidline, Simon Neubauer, Matthew M Skinner, Inga Bergmann, Adeline Le Cabec, Stefano Benazzi, and Katerina Harvati. 2017. “New fossils from Jebel Irhoud, Morocco and the pan-African origin of Homo sapiens.” Nature 546 (7657):289-292. Hughes, J Donald. 2011. “Ancient deforestation revisited.” Journal of the History of Biology 44 (1):43-57. Indriati, Etty, Carl C Swisher III, Christopher Lepre, Rhonda L Quinn, Rusyad A Suriyanto, Agus T Hascaryo, Rainer Grün, Craig S Feibel, Briana L Pobiner, and Maxime Aubert. 2011. “The age of the 20 meter Solo River Terrace, Java, Indonesia and the survival of Homo erectus in Asia.” PLoS One 6 (6). Isaac, Erich. 1962. “On the domestication of cattle.” Science 137 (3525):195-204. James, Steven R, RW Dennell, Allan S Gilbert, Henry T Lewis, JAJ Gowlett, Thomas F Lynch, WC McGrew, Charles R Peters, Geoffrey G Pope, and Ann B Stahl. 1989. “Hominid use of fire in the Lower and Middle Pleistocene: A review of the evidence [and comments and replies].” Current Anthropology 30 (1):1-26. Jarrige, Jean-Francois. 1975. “Excavations at Mehrgarh, Pakistan.” In South Asian Archaeology, edited by J. E. Van Lohuizen-De Leeuw, 76-87. Leiden: Brill. Jaspers, Karl. 1948. “The Axial Age of Human History.” Commentary 6:430. Jaspers, Karl. 1949. “Vom Ursprung und Ziel der Geschichte, München: R.” Piper & Co Verlag. Ji, Xueping, Xiujuan Ao, Xinglan Xu, Xiaoling Zhang, and Huizhong Tan. 2006. “A batch of Mousterian characteristic stone products unearthed in

54

Chapter 1

Dahe, Fuyuan, Yunnan Province.” China Cultural Relics News, 18 August 2020, 2. Jolie, Edward A, Thomas F Lynch, Phil R Geib, and James M Adovasio. 2011. “Cordage, textiles, and the late Pleistocene peopling of the Andes.” Current Anthropology 52 (2):285-296. Jordan, Peter, and Marek Zvelebil, eds. 2009. Ceramics before farming: the dispersal of pottery among prehistoric Eurasian hunter-gatherers. Walnut Creek, CA: Left Coast Press. Juleff, Gill. 1996. “An ancient wind-powered iron smelting technology in Sri Lanka.” Nature 379 (6560):60-63. Klein, Richard G. 2009. The human career: human biological and cultural origins. Chicago and London: University of Chicago Press. Knapp, A Bernard, and Sturt W Manning. 2016. “Crisis in context: The end of the Late Bronze Age in the eastern Mediterranean.” American Journal of Archaeology 120 (1):99-149. Kramer, Samuel Noah. 2010. The Sumerians: Their history, culture, and character. Chicago, IL: University of Chicago Press. Kuijt, Ian, and Nathan B Goodale. 2006. “Chronological frameworks and disparate technology: An exploration of chipped stone variability and the forager to farmer transition at 'Iraq Ed-Dubb, Jordan.” Paléorient 32 (1):27-45. Kumar, Sudhir, and S Blair Hedges. 1998. “A molecular timescale for vertebrate evolution.” Nature 392 (6679):917-920. Kurt, Amélie. 1995. The Ancient Near East. Vol. 1, History of the Ancient World. London and New York: Routledge Kuzmin, Yaroslav V. 2006. “Chronology of the earliest pottery in East Asia: progress and pitfalls.” Antiquity 80 (308):362-371. Lazuén, Talía. 2012. “European Neanderthal stone hunting weapons reveal complex behaviour long before the appearance of modern humans.” Journal of Archaeological Science 39 (7):2304-2311. Lee, Siu-Leung. 2011. “Zheng He’s Voyages Revealed in Matteo Ricci’s World Map.” Journal of Ancient Civilizations 2 (8):306-334. Lee, Siu-Leung. 2015. “Maps that turn world history upside down.” Midwestern Epigraphic Society Journal 32:24-29. Leick, Gwendolyn. 2002. Mesopotamia: The invention of the city. London: Penguin UK. Lepre, Christopher J, Hélène Roche, Dennis V Kent, Sonia Harmand, Rhonda L Quinn, Jean-Philippe Brugal, Pierre-Jean Texier, Arnaud Lenoble, and Craig S Feibel. 2011. “An earlier origin for the Acheulian.” Nature 477 (7362):82-85.

The Manual Age

55

Li, Bozhong. 2018. “Water and the History of China.” Social Sciences In China 39 (1):120-131. Liu, Li. 2005. The Chinese Neolithic: trajectories to early states. Cambridge: Cambridge University Press. Liu, Li, and Xingcan Chen. 2012. The archaeology of China: from the late Paleolithic to the early Bronze Age. Cambridge: Cambridge University Press. Liu, Li, and Hong Xu. 2007. “Rethinking Erlitou: legend, history and Chinese Archaeology.” Antiquity 81 (314):886-901. Liu, Xinglin. 2010. “The Installation and Use of Iron Plough in the Han Dynasty and Relevant issues [J].” Archaeology and Cultural Relics (4):59-62. Liverani, Mario. 2013. The Ancient Near East: History, society and economy. London and New York: Routledge. Lombard, Marlize, and Laurel Phillipson. 2010. “Indications of bow and stone-tipped arrow use 64 000 years ago in KwaZulu-Natal, South Africa.” Antiquity 84 (325):635-648. Lu, PJ, N Yao, JF So, GE Harlow, JF Lu, GF Wang, and PM Chaikin. 2005. “The earliest use of corundum and diamond in prehistoric China.” Archaeometry 47 (1):1-12. Lucas, Adam. 2006. Wind, water, work: ancient and medieval milling technology. Vol. 8. Leiden: Brill. Ma, Qing-Ping. 2019. What drives China's economy - economic, sociopolitical, historical and cultural factors. London, Singapore: Routledge. Ma, Qing-Ping. 2020. “The China Miracle in a Technological and sociopolitical Framework—The Role of Institutions.” Journal of Chinese Economic and Business Studies 18 (3):253-272. Maddison, Angus. 2001. The world economy: a millennial perspective Paris. Paris: OECD. Mahdihassan, S. 1975. “Stirrup in China of the second century.” Annals of the Bhandarkar Oriental Research Institute 56 (1/4):234-235. Maisels, Charles Keith. 2001. Early civilizations of the old world: the formative histories of Egypt, the Levant, Mesopotamia, India and China. London and New York: Routledge. Mann, Charles. 2011. How the potato changed the world. Smithsonian Magazine (November 2011). Marlowe, Frank W. 2005. “Hunterဨgatherers and human evolution.” Evolutionary Anthropology: Issues, News, and Reviews 14 (2):54-67. Mazar, A. 1990. Archaeology of the Land of the Bible 10,000-586 BCE. New York: Doubleday.

56

Chapter 1

McClellan III, James E, and Harold Dorn. 2015. Science and technology in world history: an introduction. Baltimore, MD: Johns Hopkins University Press. McDougall, Ian, Francis H Brown, and John G Fleagle. 2005. “Stratigraphic placement and age of modern humans from Kibish, Ethiopia.” Nature 433 (7027):733-736. Mei, Jianjun, Pu Wang, Kunlong Chen, Lu Wang, Yingchen Wang, and Yaxiong Liu. 2015. “Archaeometallurgical studies in China: some recent developments and challenging issues.” Journal of Archaeological Science 56:221-232. Mellars, Paul. 2006. “Archeology and the dispersal of modern humans in Europe: Deconstructing the “Aurignacian”.” Evolutionary Anthropology: Issues, News, and Reviews 15 (5):167-182. Miles, Richard. 2011. Carthage Must Be Destroyed: The Rise and Fall of an Ancient Civilization. London: Penguin. Milner, George R, Scott W Hammerstedt, and Kirk D French. 2010. “Chert hoes as digging tools.” Antiquity 84 (323):103-113. Mingay, Gordon Edmund. 1977. The Agricultural Revolution: Changes in Agriculture, 1650-1880. London: A. and C. Black. Mithen, Steven. 2011. After the ice: a global human history, 20,000-5000 BC. London: Weidenfeld & Nicolson. Muhly, James D. 1985. “Sources of tin and the beginnings of bronze metallurgy.” American Journal of Archaeology 89 (2):275-291. Mullins, Daniel Austin, Daniel Hoyer, Christina Collins, Thomas Currie, Kevin Feeney, Pieter François, Patrick E Savage, Harvey Whitehouse, and Peter Turchin. 2018. “A systematic assessment of “axial age” proposals using global comparative historical evidence.” American Sociological Review 83 (3):596-626. Muttoni, Giovanni, Giancarlo Scardia, Dennis V Kent, Enrico Morsiani, Fabrizio Tremolada, Mauro Cremaschi, and Carlo Peretto. 2011. “First dated human occupation of Italy at~ 0.85 Ma during the late Early Pleistocene climate transition.” Earth and Planetary Science Letters 307 (3-4):241-252. Needham, Joseph. 1965. Science and Civilization in China: Volume 4, Part 2. Cambridge: Cambridge University Press. Needham, Joseph. 1971. Science and Civilization in China: Volume 4, Part 3. Cambridge: Cambridge University Press. Neeson, JM. 2000. “English enclosures and British peasants: current debates about rural social structure in Britain c. 1750-1870.” Jahrbuch für Wirtschaftsgeschichte/Economic History Yearbook 41 (2):17-32.

The Manual Age

57

Nelson, Sarah M. 1997. “Hongshan: an early complex society in Northeast China.” Bulletin of the Indo-Pacific Prehistory Association 16:57-62. Nishiaki, Yoshihiro, and Marie Le Mière. 2005. “The oldest pottery Neolithic of Upper Mesopotamia: new evidence from Tell Seker alAheimar, the Khabur, northeast Syria.” Paléorient 31 (2):55-68. Pappu, Shanti, Yanni Gunnell, Kumar Akhilesh, Régis Braucher, Maurice Taieb, François Demory, and Nicolas Thouveny. 2011. “Early Pleistocene presence of Acheulian hominins in south India.” Science 331 (6024):1596-1599. Park, Hye Ok. 2014. “The history of pre-Gutenberg woodblock and movable type printing in Korea.” International Journal of Humanities and Social Science 4 (9):9-17. Perri, Angela R, Tatiana R Feuerborn, Laurent AF Frantz, Greger Larson, Ripan S Malhi, David J Meltzer, and Kelsey E Witt. 2021. “Dog domestication and the dual dispersal of people and dogs into the Americas.” Proceedings of the National Academy of Sciences 118 (6):e2010083118. Photos, Effie. 1989. “The question of meteoritic versus smelted nickelဨrich iron: Archaeological evidence and experimental results.” World Archaeology 20 (3):403-421. Pike, Alistair WG, Dirk L Hoffmann, Marcos García-Diez, Paul B Pettitt, Jose Alcolea, Rodrigo De Balbin, César González-Sainz, Carmen de las Heras, Jose A Lasheras, and Ramón Montes. 2012. “U-series dating of Paleolithic art in 11 caves in Spain.” Science 336 (6087):1409-1413. Pilbeam, David, and Nathan Young. 2004. “Hominoid evolution: synthesizing disparate data.” Comptes Rendus Palevol 3 (4):305-321. Piperno, Dolores R. 2011. “The origins of plant cultivation and domestication in the New World tropics: patterns, process, and new developments.” Current Anthropology 52 (S4): S453-S470. Piperno, Dolores R., Anthony J. Ranere, Irene Holst, Jose Iriarte, and Ruth Dickau. 2009. “Starch grain and phytolith evidence for early ninth millennium BP maize from the Central Balsa River Valley, Mexico.” Proceedings of the National Academy of Sciences 106 (13):5019-5024. Pitts, Michael W, and Mark Roberts. 1997. Fairweather Eden: life in Britain half a million years ago as revealed by the excavations at Boxgrove. London: Century. Possehl, Gregory L. 2002. The Indus civilization: a contemporary perspective. Lanham, MD: Rowman Altamira. Qian, Fang, Qingchen Li, Peizhu Wu, Sixun Yuan, Rulian Xing, Huailu Chen, and Hucai Zhang. 1991. “Lower Pleistocene, Yuanmou Formation.” In Quaternary Geology and Paleoanthropology of

58

Chapter 1

Yuanmou, Yunnan, China, edited by Fang Qian and Guoxing Zhou, 1750. Beijing: Science Press. Qiu, Xigui. 1985. “On the Methods of Studying Ancient Chinese Script.” Early China 11:301-316. Radivojeviü, Miljana, Thilo Rehren, Ernst Pernicka, Dušan Šljivar, Michael Brauns, and Dušan Boriü. 2010. “On the origins of extractive metallurgy: new evidence from Europe.” Journal of Archaeological Science 37 (11):2775-2787. Ray, John D. 1986. “The emergence of writing in Egypt.” World Archaeology 17 (3):307-316. Rehren, Thilo, Tamás Belgya, Albert Jambon, György Káli, Zsolt Kasztovszky, Zoltán Kis, Imre Kovács, Boglárka Maróti, Marcos Martinón-Torres, and Gianluca Miniaci. 2013. “5,000 years old Egyptian iron beads made from hammered meteoritic iron.” Journal of Archaeological Science 40 (12):4785-4792. Renfrew, Colin, and Bin Liu. 2018. “The emergence of complex society in China: the case of Liangzhu.” Antiquity 92 (364):975-990. Rezepkin, Aleksej D. 2000. Das frühbronzezeitliche Gräberfeld von Klady und die Majkop-Kultur in Nordwestkaukasien. Rahden, Westfalia: Leidorf. Richards, Michael P, and Erik Trinkaus. 2009. “Isotopic evidence for the diets of European Neanderthals and early modern humans.” Proceedings of the National Academy of Sciences 106 (38):1603416039. Riede, Felix. 2009. “The loss and re-introduction of bow-and-arrow technology: a case study from the Northern European Late Paleolithic.” Lithic Technology 34 (1):27-45. Rightmire, G Philip, David Lordkipanidze, and Abesalom Vekua. 2006. “Anatomical descriptions, comparative studies and evolutionary significance of the hominin skulls from Dmanisi, Republic of Georgia.” Journal of Human Evolution 50 (2):115-141. Rischkowsky, Barbara, and Dafydd Pilling. 2007. The state of the world's animal genetic resources for food and agriculture. Rome: Food & Agriculture Organization. Roberts, Richard G, Rhys Jones, Nigel A Spooner, M John Head, Andrew S Murray, and Michael A Smith. 1994. “The human colonisation of Australia: optical dates of 53,000 and 60,000 years bracket human arrival at Deaf Adder Gorge, Northern Territory.” Quaternary Science Reviews 13 (5-7):575-583. Roche, Hélène, Jean-Philip Brugal, Anne Delagnes, Craig Feibel, Sonia Harmand, Mzalendo Kibunjia, Sandrine Prat, and Pierre-Jean Texier.

The Manual Age

59

2003. “Les sites archéologiques plio-pléistocènes de la formation de Nachukui, Ouest-Turkana, Kenya: bilan synthétique 1997–2001.” Comptes Rendus Palevol 2 (8):663-673. Roux, Georges. 1992. Ancient Iraq. London: Penguin UK. Sablin, Mikhail V, and GennadyA Khlopachev. 2002. “The earliest Ice Age dogs: evidence from Eliseevichi.” Current Anthropology 43 (5):795-799. Salamini, Francesco, Hakan Özkan, Andrea Brandolini, Ralf Schäfer-Pregl, and William Martin. 2002. “Genetics and geography of wild cereal domestication in the near east.” Nature Reviews Genetics 3 (6):429-441. Salem, Abu. 1976. “Type Site of the Harifian industry of the Southern Levant.” Journal of Field Archaeology 3 (1):43-60. Schick, Tamar. 1988. “Nahal Hemar Cave: cordage, basketry and fabrics.” Atiqot 18:31-43. Semaw, Sileshi. 2000. “The world's oldest stone artefacts from Gona, Ethiopia: their implications for understanding stone technology and patterns of human evolution between 2· 6–1· 5 million years ago.” Journal of Archaeological Science 27 (12):1197-1214. Semaw, Sileshi, Michael J Rogers, Jay Quade, Paul R Renne, Robert F Butler, Manuel Dominguez-Rodrigo, Dietrich Stout, William S Hart, Travis Pickering, and Scott W Simpson. 2003. “2.6-Million-year-old stone tools and associated bones from OGS-6 and OGS-7, Gona, Afar, Ethiopia.” Journal of Human Evolution 45 (2):169-177. Sen, Sailendra Nath. 1999. Ancient Indian history and civilization. New Dehli, Bangalore and Calcutta: New Age International. Shan, Ming Chao, Gaowa Naren, Xingyi Zhou, and Fuyou Chen. 2017. “Discovery of Paleolithic sites 50000 years ago in Sanlongdong cave, Chifeng, Inner Mongolia.” China Cultural Relics News, 20 October 2017, 8. Sharif, M, and BK Thapar. 1992. “Food producing communities in Pakistan and northern India.” In History of civilizations of Central Asia, edited by A. H. Dani and V. M. Masson, 127-151. Dehli: Motilal Banarsidass publishers. Shaughnessy, Edward L. 1992. Sources of Western Zhou history: inscribed bronze vessels. Berkeley, Los Angeles and Oxford: University of California Press. Shaw, Ian. 2003. The Oxford History of Ancient Egypt. Oxford: Oxford University Press. Shaw, Ian, and Robert Jameson. 2008. A dictionary of archaeology. Oxford: John Wiley & Sons. Simmons, Alan H. 2016. Stone Age sailors: Paleolithic seafaring in the Mediterranean. London and New York: Routledge.

60

Chapter 1

Simpson, Elizabeth. 2018. “Luxury Arts of the Ancient Near East.” In The Adventure of the Illustrious Scholar, 662-694. Leiden: Brill. Singh, LK. 2008. Indian cultural heritage perspective for tourism. Delhi: Isha Books. Šljivar, DUŠKO. 2006. “The earliest copper metallurgy in the Central Balkans.” Metalurgija 12 (2–3):93-104. Smith, Philip EL. 1976. “Stone-age man on the Nile.” Scientific American 235 (2):30-41. Snir, Ainit, Dani Nadel, Iris Groman-Yaroslavski, Yoel Melamed, Marcelo Sternberg, Ofer Bar-Yosef, and Ehud Weiss. 2015. “The origin of cultivation and proto-weeds, long before Neolithic farming.” PLoS One 10 (7). Stavrianos, Leften Stavros. 1991. A global history: From prehistory to the present. Upper Saddle River, NJ: Prentice Hall. Stech-Wheeler, Tamara, JD Muhly, KR Maxwell-Hyslop, and R Maddin. 1981. “Iron at Taanach and early iron metallurgy in the Eastern Mediterranean.” American Journal of Archaeology 85 (3):245-268. Steele, Tracy. 2005. “1421: The Year China Discovered America.” The Sixteenth Century Journal 36 (1):276-278. Stremlin, Boris. 2008. “The Iron Age WorldဨSystem.” History Compass 6 (3):969-999. Svoboda, JiĜí, Miroslav Králík, VČra ýulíková, Šárka Hladilová, Martin Novák, Miriam Nývltová Fišáková, Daniel Nývlt, and Michaela Zelinková. 2009. “Pavlov VI: an Upper Palaeolithic living unit.” Antiquity 83 (320):282-295. Swisher, Carl C, Garniss H Curtis, Teuku Jacob, Ann G Getty, and Agus Suprijo. 1994. “Age of the earliest known hominids in Java, Indonesia.” Science 263 (5150):1118-1121. Swisher III, Carl C, Garniss H Curtis, and Roger Lewin. 2001. Java man: how two geologists changed our understanding of human evolution. Chicago, IL: University of Chicago Press. Tewari, Rakesh. 2003. “The origins of iron working in India: new evidence from the Central Ganga Plain and the Eastern Vindhyas.” Antiquity 77 (297):536-544. Thomas, Richard. 2005. “Zooarchaeology, improvement and the British agricultural revolution.” International Journal of Historical Archaeology 9 (2):71-88. Tong, Kin-woon. 1983. “Shang musical instruments: Part one.” Asian Music 14 (2):17-182. Toth, Nicholas, and Kathy Schick. 2007. “Overview of paleolithic archeology.” In Handbook of Paleoanthropology, edited by H. C.

The Manual Age

61

Winfried Henke, Thorolf Hardt and Ian Tattersall, 1943-1963. Berlin, Heidelberg and New York: Springer. Toth, Nicholas, and Kathy Schick. 2009. “The Oldowan: the tool making of early hominins and chimpanzees compared.” Annual Review of Anthropology 38:289-305. Turchin, Peter, Jonathan M Adams, and Thomas D Hall. 2006. “East-west orientation of historical empires and modern states.” Journal of WorldSystems Research 12 (2):219-229. Tylor, EB. 1881. “On the Origin of the Plough and Wheel-Carriage.” The Journal of the Anthropological Institute of Great Britain and Ireland 10:74-84. Verhoeven, Marc. 2000. “The small finds.” In Tell Sabi Abyad II. The prepottery Neolithic B settlement edited by M. Verhoeven and P.M.M.G Akkermans, 91-122. Leiden and Istanbul: Nederlands HistorischArchaeologisch Instituut. Vermeersch, Pierre M, Marcel Otte, Etienne Gilot, Etienne Paulissen, G Guselings, and D Drappier. 1982. “Blade technology in the EgyptianNile Valley: some new evidence.” Science 216 (4546):626-628. Vigne, Jean-Denis, Antoine Zazzo, Thomas Cucchi, Isabelle Carrère, François Briois, and Jean Guilaine. 2014. “The transportation of mammals to Cyprus sheds light on early voyaging and boats in the Mediterranean Sea.” Eurasian Prehistory 10 (1-2):157-176. Villmoare, Brian, William H Kimbel, Chalachew Seyoum, Christopher J Campisano, Erin N DiMaggio, John Rowan, David R Braun, J Ramón Arrowsmith, and Kaye E Reed. 2015. “Early Homo at 2.8 Ma from LediGeraru, Afar, Ethiopia.” Science 347 (6228):1352-1355. Wagner Donald, B. 1993. Iron and Steel in Ancient China: Second Impression, with Corrections. Leiden: Brill. Wahida, Ghanim. 1981. “The re-excavation of Zarzi, 1971.” Proceedings of the Prehistoric Society 47:19-40. Waldbaum, Jane C. 1978. From bronze to iron. The transition from the Bronze Age to the Iron Age in the Eastern Mediterranean. Vol. 54, Studies in Mediterranean archaeology. Göteborg: Coronet Books. Wang, Shejiang, Huayu Lu, and Luda Xing. 2014. “Chronological and typotechnological perspectives on the Palaeolithic archaeology in Lantian, central China.” Quaternary International 347:183-192. Weiss, Ehud, Mordechai E Kislev, Orit Simchoni, Dani Nadel, and Hartmut Tschauner. 2008. “Plant-food preparation area on an Upper Paleolithic brush hut floor at Ohalo II, Israel.” Journal of Archaeological Science 35 (8):2400-2414.

62

Chapter 1

Weisskopf, Alison, Ling Qin, Jinglong Ding, Pin Ding, Guoping Sun, and Dorian Q Fuller. 2015. “Phytoliths and rice: from wet to dry and back again in the Neolithic Lower Yangtze.” Antiquity 89 (347):1051-1063. Wendrich, Willeke, and Philippa Ryan. 2012. “Phytoliths and Basketry Materials at Çatalhöyük (Turkey): Timelines of growth, harvest and objects life histories.” Paléorient 38 (1-2):55-63. White, Lynn Townsend. 1964. Medieval technology and social change. Vol. 163. London, Oxford and New York: Oxford University Press. Willmott, WE. 1989. “Dujiangyan: irrigation and society in Sichuan, China.” The Australian Journal of Chinese Affairs 22:143-153. Wilson, Andrew. 2002. “Machines, power and the ancient economy.” The Journal of Roman Studies 92:1-32. Wittfogel, Karl A. 1957. Oriental Despotism: A Comparative Study of Total Powe. New Haven, CT: Yale University Press. Wrangham, Richard, and NancyLou Conklin-Brittain. 2003. “Cooking as a biological trait.” Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 136 (1):35-46. Wright, Rita P. 2010. The ancient Indus: Urbanism, economy, and society. Cambridge: Cambridge University Press Wu, Xiaohong, Chi Zhang, Paul Goldberg, David Cohen, Yan Pan, Trina Arpin, and Ofer Bar-Yosef. 2012. “Early pottery at 20,000 years ago in Xianrendong Cave, China.” Science 336 (6089):1696-1700. Xu, Guilin, Wei Wang, Christopher J Bae, Shengmin Huang, and Zhiming Mo. 2012. “Spatial distribution of Paleolithic sites in Bose Basin, Guangxi, China.” Quaternary International 281:10-13. Xu, Zhen-Tao, FR Stephenson, and Yao-Tiao Jiang. 1995. “Astronomy on oracle bone inscriptions.” Quarterly Journal of the Royal Astronomical Society 36 (4):397-406. Yang, Xiaoyan, Zhiwei Wan, Linda Perry, Houyuan Lu, Qiang Wang, Chaohong Zhao, Jun Li, Fei Xie, Jincheng Yu, and Tianxing Cui. 2012. “Early millet use in northern China.” Proceedings of the National Academy of Sciences 109 (10):3726-3730. Yang, Xizhang, and Yiman Liu. 1999. “Major Achievements of 70 Years Archaeological Works in Yinxu.” Cultural Relics of Central China (2):17-27. Yasuda, Yoshinori. 2013. “Discovery of the Yangtze River Civilization in China.” In Water Civilization, edited by Yoshinori Yasuda, 3-45. Tokyo, Heidelberg, New York, Dordrecht and London: Springer. Yellen, John E, Alison S Brooks, Els Cornelissen, Michael J Mehlman, and Kathlyn Stewart. 1995. “A middle stone age worked bone industry from Katanda, Upper Semliki Valley, Zaire.” Science 268 (5210):553-556.

The Manual Age

63

Yu, Pei. 2015. “Unearthed jade in Hubei Province shows the highest achievement of prehistoric Chinese jade culture.” Institute of Archeology, Chinese Academy of Social Sciences, accessed 2 June 2020. http://www.kaogu.cn/cn/xccz/20151225/52553.html. Yuan, Jing. 2010. “Zooarchaeological study on the domestic animals in ancient China.” Quaternary Sciences 30 (2):298-306. Zhang, Chi. 2013. “The Qujialing-Shijiahe Culture in the Middle Yangzi River Valley.” In A Companion to Chinese Archaeology, edited by Anne P. Underhill, 510-534. Chichester, UK: John Wiley & Sons. Zhang, Guoshuo, and Jiejuan Tang. 2017. “A Study of Early Iron Casting Discovered in the Central Plains.” Cultural Relics of Central China (2):64-69. Zhu, RX, Richard Potts, Fei Xie, Kenneth A Hoffman, Chenglong L Deng, CD Shi, Yongxin X Pan, Hui Qiu Wang, Rui Ping Shi, and YC Wang. 2004. “New evidence on the earliest human presence at high northern latitudes in northeast Asia.” Nature 431 (7008):559-562. Zhu, Zhao-Yu, Robin Dennell, Wei-Wen Huang, Yi Wu, Zhi-Guo Rao, ShiFan Qiu, Jiu-Bing Xie, Wu Liu, Shu-Qing Fu, and Jiang-Wei Han. 2015. “New dating of the Homo erectus cranium from Lantian (Gongwangling), China.” Journal of Human Evolution 78:144-157.

CHAPTER 2 THE MACHINE AGE: FROM THE INDUSTRIAL REVOLUTION TO THE 1970S STAGFLATION

The Industrial Revolution was the transition to new manufacturing processes from about 1760 to sometime between 1820 and 1840 (Crafts 1996). The most important feature of this transition was production going from manual-powered to machine-powered, which brought humanity into the Machine Age. This book defines machines as instruments powered by non-animal power sources, although animals could and have powered some machines. Simply put, tools are instruments powered by animal power sources, including humans. In a broad sense, tools include machines. Accompanying this manual-to-machine transition, new chemical manufacturing and iron-making processes emerged, machine tools were further developed, and the factory system became the dominant organizational form of production. The Industrial Revolution was the most important event in the history of humanity since the Neolithic Agricultural Revolution. It marked a major turning point in history; the advent of the Machine Age. The Industrial Revolution started a sustained growth in the standard of living for the general population, although meaningful improvements for low-income classes might only have appeared in the late nineteenth and early twentieth centuries (Feinstein 1998; Szreter and Mooney 1998; Voth 2003). The Second Industrial Revolution in the late nineteenth and early twentieth centuries further increased production and the standard of living (Jevons 1931). Technologies developed during and immediately following the Second Industrial Revolution provided most of the material wealth in modern society. The digital revolution in the second half of the twentieth century, which has fundamentally changed how people communicate and process information, appears to herald the coming of a new age (Bojanova 2014). This chapter will examine how the Industrial Revolution arose and the main achievements during and following it.

The Machine Age

65

1. The Causes of the Industrial Revolution The advent of the Industrial Revolution was the outcome of many factors. In addition to the British Agricultural Revolution that increased the labor force available for factories and farmers’ demands for non-farm goods, the most important factor was the Commercial Revolution, which prepared the market and commercial networks for goods to be supplied and cleared in large quantities. The level of technological developments by the early eighteenth century made inventions and innovations possible during the Industrial Revolution. The protectionist trade policy of the British government also played a significant role in the rise of the British textile industry.

1.1. The Commercial Revolution The term Commercial Revolution denotes the increased general commerce and the growth of financial services such as banking, insurance, and stock exchanges in Europe between the late eleventh century and the early eighteenth century. It created a European economy based on trade. After the fall of the West Roman Empire in 476 CE, wealth became land-based, money became scarce, and local fiefs were self-sufficient, which was not conducive to promoting economies of scale and labor productivity. The Crusades, which started in the eleventh century, increased the interaction between Europe and other continents and among Europeans. They led to the rediscovery by Europeans of spices, silks, and other commodities rare in Europe, which created a new desire for trade and, eventually, vast international trade networks through voyages of discovery. The Crusader states were the first experiments in European colonialism to create “Europe Overseas” or Outremer (Tyerman 2019). The raising, transportation, and supply of large armies led to flourishing trade between Europe and the Outremer and the establishment of institutions such as banks, stock exchanges, and insurance firms to facilitate trade. Existing informal methods of dealing in trade and commerce were gradually formalized. Banking and gold and silver from the New World facilitated the creation of a trade-based European economy. With the increasing importance of trade and commerce, mercantilism became a dominant economic theory and a general national economic policy designed to maximize a nation's exports. The commercial and trade network established by the Commercial Revolution allowed increased outputs by innovations to be cleared by the market and supported by raw materials produced by faraway suppliers. The

66

Chapter 2

growing importance of commerce and trade made merchants a vital political force in medieval society (Breslow 2004; Thrupp 1989).

1.2. The British Textile Industry and Trade Before the Industrial Revolution As a part of the trade-based European economy during the High and Late Middle Ages, the English wool trade was the backbone of the English economy (Lloyd 2005). English wool was exported to the emergent urban centers of cloth production in the Low Countries, France, and Italy (Munro 2003). From the mid-fourteenth century, English exports gradually changed to finished cloth. The British textile industry was based on wool and flax; spinning and weaving were done in households as a cottage industry under the putting-out system. Clothiers visited the villages with their trains of packhorses, and a large amount of the collected cloth was exported. The fabric was costly because of relatively expensive raw materials and low labor productivity. Imported cotton textiles attracted European consumers because of their low prices and high quality compared to local woolen and linen products. The East India Company introduced England and Scotland to cheap calico and chintz cloth in the 1660s as a novelty sideline from its spice trading posts in Asia. Still, the popularity of imported cotton textiles soon threatened the market share of domestic woolen and linen producers. Many cotton textiles, especially muslin, were imported from Bengal, which had manufactured fabrics for centuries (Ashmore 2012). Under the Mughal rule, Bengal was a center of the worldwide muslin, silk, and pearl trades (Richards 1993). The Portuguese started trading textiles from the Indian subcontinent in the sixteenth century. As the business expanded, European companies became interested in founding their factories in Dhaka. The Dutch established their factory in Dhaka in 1663, the English in 1669, and the French in 1682 (Prakash 2014). Until 1750, India produced about 25% of the world's industrial output, while Bengal had a 25% share of the global textile trade in the early eighteenth century (Maddison 1995; Prakash 2014; Clingingsmith and Williamson 2008). Mughalistan accounted for 95% of English imports from Asia in the late seventeenth and early eighteenth centuries. Bengal accounted for more than 50% of textiles, around 80% of silks imported by the Dutch from Asia and marketed to the world and 30% of the English trade with southern Europe in the early eighteenth century. European fashion increasingly depended on Mughalistan textiles and silks (Richards 1993; Prakash 2014).

The Machine Age

67

The imported cotton textiles competed with the domestic textile industry. In the late seventeenth and early eighteenth centuries, the British government passed a series of Calico Acts restricting imports and sales of cotton textiles to protect the domestic woolen industry (Eacott 2012). Calico was a plainwoven textile made from unbleached cotton. In 1700 an Act of Parliament was passed to prevent the importation of dyed or printed calicoes from India, China, or Persia. At the beginning of the eighteenth century, producing textiles made with wool from the large sheep-farming areas in the Midlands and across the country was a critical British industry, with woolen goods accounting for more than a quarter of British exports during most of the eighteenth century. Since cotton had become popular, a home-based cotton industry sprung up using cotton imported from the colonies. The success of the English textile industry is a counter-example of free trade based on the comparative advantages of the two countries. This partially explains why most countries still practice protectionism to some extent despite constant preaching by most economists about free trade. Without the protectionist policy, the English textile industry might have been unable to compete with cheap cotton imports from Bengal, at least not in the short term. The reward and drive for innovations in the textile industry would have been much weaker, so the Industrial Revolution might have been delayed or arisen elsewhere. The British Bengal Presidency was founded in 1765 after the East India Company conquered Bengal following the Battle of Plassey in 1757 and the Battle of Buxar in 1764. British colonization forced open the Bengali market to British goods. A heavy duty of 75% was imposed on the export of cotton textiles from Bengal, while raw cotton was imported without taxes or tariffs to British factories. British colonial economic policies led to deindustrialization in Bengal (Clingingsmith and Williamson 2008; Cypher and Cypher 2008).

1.3. Why did the Industrial Revolution Happen in Britain and Europe? There are many studies on why the Industrial Revolution happened in the eighteenth century in Britain and Europe instead of in earlier periods or other parts of the world. The timing of the eighteenth century and the location of Europe were probably attributable to four factors: 1) the commercial entrepreneurship from the Middle Ages; 2) the resources obtained from overseas colonies; 3) the agricultural revolution, technological developments, and scientific revolution during the early

68

Chapter 2

modern period, and 4) the government arrangements facilitating business and protecting domestic industries. The government arrangement in Europe was more facilitating to commerce and entrepreneurship than those in other parts of the world. The merchant class controlled the rich and powerful maritime republics during the Middle Ages and early modern era, with many innovations in managing business activities. The first statutory patent system was the Venetian Patent Statute of 1474, and patents were systematically granted in Venice from then on. New and inventive devices had to be communicated to the republic to obtain legal protection against potential infringers, and the protection period was ten years (Long 1991). As migrating Venetians sought similar patent protection in their new homes, the patent systems diffused to other countries (Frumkin 1945). The English patent system had its medieval origins in the crown's power to grant monopolies. The Statute of Monopolies passed by the English Parliament in 1624 restricted the crown's power explicitly so that the king could only issue letters patent (a type of legal instrument in the form of a published written order to grant an office, right, monopoly, title, or status) to the inventors or introducers of original inventions for a fixed number of years. Letters patent were the Statute of Monopolies that evolved into the first modern patent system recognizing intellectual property to stimulate creativity (MacLeod 2002). There were multiple reasons for England to be the birthplace of the Industrial Revolution. First, a large domestic market was formed when Henry VIII abolished internal tariffs. Second, the rule of law was generally practiced, property rights were respected, and a straightforward legal system allowed the formation of joint-stock companies; culturally, the society accepted self-interest and an entrepreneurial spirit. Third, England enjoyed extended periods of peace and stability, especially after unification with Scotland, and it had not been affected by wars raging on the continent. Fourth, natural or financial resources that Britain received from its many overseas colonies and profits from the British slave trade between Africa and the Caribbean might have added additional fuel to industrialization. Demand from its overseas territories increased aggregated demand in the British economy. Fifth, Britain had an excellent natural environment and resources (Landes 1969; Deane 1979). Britain had the highest quality coal in Europe, extensive coastlines, and many navigable rivers in an age where water was the most efficient means of transportation. In 1700, over 80% of coal mined worldwide was in Britain (Landes 1969). A lack of coal might explain why the Netherlands failed to

The Machine Age

69

industrialize in the eighteenth century, although it had Europe's best transport and most urbanized, well-paid, and literate people (Mokyr 2000). Although Dutch people were very experienced with waterworks, the Netherlands was not rich in water power, which might explain why waterpowered textile factories did not first arise there. Their famous windmills were not adequate for powering textile factories. In Britain, local supplies of coal, iron, lead, copper, tin, limestone, and water power provided excellent conditions for the development and expansion of the industry. Moreover, the success of the Netherlands in maritime commerce and trade might have hindered its people’s efforts in industrial innovations.

2. The Industrial Revolution The Industrial Revolution began in the United Kingdom and then spread to Continental Europe and worldwide. In many countries, it involved the application of technology developed in Britain. The German, Russian, and Belgian governments provided state funding to the new industries, and British technologies were purchased. British engineers and entrepreneurs who moved abroad for new opportunities also brought technologies with them. The textile industry played a dominant role in starting the Industrial Revolution (Ashton 1997; Deane 1979).

2.1. Machines and Power Sources The scientific and technological progress during the Middle Ages and the early modern era laid the foundation for the Industrial Revolution. The watermill and windmill had been in existence well before the eighteenth century. Shipbuilding, military equipment, clock making, and many others provided the technical know-how for inventing textile industry machines and applying water power. The new era was initiated by improving manual tools for making textiles. Human society entered the Machine Age when the more efficient tools became driven by non-biological power sources, especially those not constrained by geography. 2.1.1. New Tools in the Textile Industry The beginning of the technological progress that underlay the Industrial Revolution was the improvement of manual tools. Cotton provided a cheaper raw material to meet the demand for quality cloth by the population, which called for new tools to increase labor productivity. It used to take four to eight spinners using the spinning wheel to supply one hand loom weaver (Landes 1969). The flying shuttle, patented in 1733 by clockmaker John

70

Chapter 2

Kay, doubled the output of a weaver. His son Robert's invention of the drop box in 1760 worsened the imbalance (Wood et al. 1911; Baines 2015; Sutcliffe 1843). The first known attempt to increase spinning productivity was the roller spinning frame and the flyer-and-bobbin system for drawing wool to a more even thickness, patented by Lewis Paul and John Wyatt in 1738. Before cotton and wool could be spun into yarn, carding was needed to disentangle, clean, and intermix fibers to produce a continuous web or sliver. Lewis Paul and Daniel Bourn patented carding machines in 1748 based on two rollers traveling at different speeds. Richard Arkwright 1775 took out a patent for a carding machine that converted raw cotton buds into a continuous skein of cotton fibers that could be spun into yarn (Sutcliffe 1843; Marsden 1895).

Fig.2-1 A model of the spinning jenny in the Museum of Early Industrialization, Wuppertal, Germany (by Markus Schweiß, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=195501).

The spinning jenny was the first practical spinning frame with multiple spindles (Fig.2-1), invented in 1764 by James Hargreaves and patented in 1770. The jenny was a simple, wooden framed machine, working similarly to the spinning wheel. It cost about £6 for a 40-spindle model in 1792 and was used mainly by home spinners. The jenny produced a lightly twisted cotton yarn only suitable for weft, not warp, because it did not have

The Machine Age

71

sufficient strength. The heavier fabric fustian was a cloth with flax warp and cotton weft, which was not as soft as 100% cotton (Sutcliffe 1843; Marsden 1895). The water frame (Fig.2-2) developed and patented in 1769 by Richard Arkwright and his partners was an essential step toward the Machine Age. It produced yarn suitable for warp, so 100% cotton cloth could be made in Britain. Horses powered Arkwright's first factory to use the spinning frame. His factory in Cromford, Derbyshire, was water powered in 1771, so the invention acquired the name water frame (Baines 2015; Marsden 1895). The Cromford factory employed 200 people, mainly women and children (Vance Jr 1966), as physical strength was no longer critical with waterpowered machines, and women and children were cheap to hire. Since labor productivity was still low, increased demand and output (for existing or new products) implied increased demand for labor.

Fig.2-2 A model of the Arkwright water frame in the Museum of Early Industrialization, Wuppertal, Germany (by Markus Schweiß, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=1987212).

Richard Arkwright was more an entrepreneur than an inventor. He also developed non-human power, first horse power and then water power,

72

Chapter 2

which mechanized cotton manufacturing. He installed the first steam engine in 1780 in a cotton mill at the Haarlem Mill in Wirksworth, Derbyshire, to replenish the millpond that drove the mill's waterwheel (Tann 1979). The application of steam power made it possible for places without water power to become industrial centers. Arkwright was later known as the “father of the modern industrial factory system” (Wilson 1990). The spinning mule, developed in 1779 by Samuel Crompton, was a further improvement by combining the spinning jenny and the water frame, but water power was not applied until 1792 (Catling 1978). The mule produced stronger and finer yarn than hand spinning and, at a lower cost, which was suitable for any textile (Marsden 1895; Sutcliffe 1843). In 1830, Richard Roberts patented the first self-acting mule, with which one worker could spin 100 lb of cotton in 135 hours (Griffin 2018). It would take 50,000 hours with a hand-powered spinning wheel and 300 hours with a spinning mule. Fig. 2-3 illustrates a self-acting mule made by Taylor, Lang & Co.

Fig.2-3 A model of the self-acting mule made by Taylor, Lang & Co (by unknown author–Textile Mercury Newspaper 1892, Public Domain, https://commons.wikimedia.org/w/index.php?curid=6532163).

As inventions in spinning increased the supply of spun cotton, improved tools for weaving became necessary. Edmund Cartwright patented a twoperson operated loom in 1776 to increase the weaver’s productivity, and then developed a vertical power loom in 1784 and patented it in 1785. Robert Grimshaw licensed his ideas and built a small steam-powered weaving factory in Manchester in 1790. Samuel Horrocks patented a more successful loom in 1813, which was improved by Richard Roberts in 1822

The Machine Age

73

and then produced in large numbers by Roberts, Hill & Co. These new machines dramatically increased the output of an individual laborer (Marsden 1895). The inexpensive cotton gin invented in 1793 and patented in 1794 by Eli Whitney reshaped the economy of the Antebellum South of the United States. Before its invention, removing seeds from cotton bolls (seed pods) was time-consuming, and the labor employed in removing them made cotton plantations less profitable. The cotton gin was capable of cleaning 50 lb (23 kg) of lint (fibers) per day, which would have previously taken a female laborer two months (Bates 1899). Cotton became a tremendously profitable business. New Orleans, Mobile, Charleston, and Galveston became major shipping ports due to cotton exports throughout the South. The factory model was soon copied in the United States. Thomas Somers and the Cabot Brothers founded the Beverly Cotton Manufactory in 1787, the first cotton mill in America. In 1793, Samuel Slater founded the Slater Mill at Pawtucket, Rhode Island, and he eventually owned 13 textile mills. Daniel Day established a wool carding mill in the Blackstone Valley at Uxbridge, Massachusetts, in 1809. The Blackstone River and its tributaries were the birthplaces of America's Industrial Revolution (Bagnall 1893). From Worcester, Massachusetts, to Providence, Rhode Island, covering more than 45 miles (72 km), over 1100 mills operated in this valley at its peak. American entrepreneurs also improved the factory model. Francis Cabot Lowell set up the Boston Manufacturing Company at Waltham, Massachusetts, in 1813. After he died in 1817, his associates built America's first planned factory town, Lowell, Massachusetts, using 5.6 miles (9.0 km) of canals and the 10,000 hp provided by the Merrimack River. They introduced the Waltham-Lowell system, under which workers earned more money than they could at home and lived a cultured life in company boarding houses with strict hours and a moral code. This enterprise was one of the first to use a public stock offering for capitalization in the United States. It is considered by some a significant contributor to the success of the American Industrial Revolution (Lubar 1984). 2.1.2. Steam Power Water-powered cotton mills were a crucial step in the Industrial Revolution, but water power was too restrictive geographically for carrying humanity into the Machine Age. Therefore, steam engines became the principal driver

74

Chapter 2

of the First Industrial Revolution. They convert heat from the chemical energy in coal (or other fuels) to mechanical power, opening a broader avenue to the Machine Age. The rich coal resources in Britain facilitated the Industrial Revolution. Several inventors contributed to the development of steam engines. Ancient people had long observed steam pushing objects up, probably during cooking. In the first century CE, the ancient Greek mathematician and inventor Hero of Alexandria invented the first known “steam engine,” the aeolipile, a copper sphere with nozzles opening in directions tangent to its surface. When filled with water and heated, steam spewed from the nozzles driving the aeolipile to revolve (Morley 2000). Hero’s invention was only used for entertainment, and modern steam engines originated in the late seventeenth century. Denis Papin built a model of a piston steam engine in 1690. In 1698, Thomas Savery constructed and patented in London a lowlift combined vacuum and pressure water pump powered by steam, which generated about one horsepower (hp) and was used in numerous water works and a few mines. In 1705, Papin developed a second steam engine that used steam pressure rather than atmospheric pressure, based on the invention of Thomas Savery (Kitsikopoulos 2013).

Fig.2-4 A diagram of the Newcomen steam engine (by Newton Henry Black, Harvey Nathaniel Davis, Public Domain, https://commons.wikimedia.org/w/index.php?curid=3908160).

Thomas Newcomen created the first successful piston steam engine for pumping water in 1712, the Newcomen steam engine (Fig. 2-4), which

The Machine Age

75

produced about five hp (3.7 kW). A number of the machines were successfully used in Britain for draining hitherto unworkable deep mines. It was cost-effective when used at pitheads, where coal was cheap. The Newcomen engines had spread to Hungary, Germany, Austria, and Sweden, with 110 built by 1733 and 1,454 built by 1800 (Rolt Lionel and Allen 1997). James Watt fundamentally changed the working principles of steam engines and introduced a separate steam condenser chamber in 1765. Collaborating with Matthew Boulton, he installed two machines in commercial enterprises in 1776. Watt engines closed off the upper part of the cylinder, surrounded the cylinder with a steam jacket, and used a separate condenser so that the cooling water was no longer injected directly into the cylinder. With these improvements, Watt engines used only 20–25% as much coal per hp-hour as the Newcomen machines. The cost-efficiency of the Watt steam engines made industry-wide application of steam power possible, such that Watt was often credited with inventing steam engines. Boulton and Watt opened the Soho Foundry in 1795 to manufacture the Watt steam engines (Dickinson 2011). Fig. 2-5 is a diagram of a 1784 steam engine designed by Boulton and Watt.

Fig. 2-5 A diagram of a 1784 steam engine designed by Boulton and Watt (by Robert Henry Thurston (1839–1903), Public Domain, https://commons.wikimedia.org/w/index.php?curid=3478839).

76

Chapter 2

James Watt developed a double-acting rotative engine, in which the steam acted alternately on the two sides of the piston, in 1783. It could be used to drive the rotary machinery of a factory directly. He also designed the parallel motion linkage that was essential in double-acting engines, allowing the up and down movements of the cylinder rod to be transformed into a circular motion. Robinson's Mill at Papplewick near Nottingham became the first cotton mill driven directly by a steam engine in 1785 (Tann 1979). Peter Drinkwater opened the Piccadilly Mill, the first mill powered by steam in Manchester, in 1789. By 1800 Manchester had 42 mills and became the heart of the cotton manufacturing trade (Nevell 2005). The demand from the textile industry stimulated the growth of steam engine manufacturing, whereas the growth of steam engine manufacturing enabled the expansion of the textile industry. The steam power application to the industrial printing processes facilitated the development of newspaper and popular book publishing, promoting knowledge, information transmission, and literacy. Watt steam engines were beam engines built integral to an engine house. James Sadler made the first table engine with its cylinder placed on top of a table-shaped base (Hodgson 1927). Both types were stationary steam engines. Richard Trevithick developed the world’s first high-pressure steam engine, which discharged the exhaust steam rather than condensing it, in 1800. As the engine was compact enough for use on locomotives and steamboats, he built the first full-scale working railway steam locomotive. On 21 February 1804, Trevithick's steam locomotive hauled a train along the tramway of the Penydarren Ironworks in Merthyr Tydfil, Wales, the first locomotive-hauled railway journey in the world (Dickinson and Titley 2010). The American engineer Oliver Evans independently constructed his high-pressure, non-condensing steam engines in 1801 (Sellers Jr 1886). British engineer Arthur Woolf invented a method to lessen the magnitude of energy loss to a very long cylinder in 1804 and patented his Woolf highpressure compound engine in 1805. The compound engine consisted of a (Trevithick's) high-pressure (HP) cylinder and one or more subsequent lower-pressure (LP) cylinders (of Watt's design) (Jenkins 1932). This reduced the magnitude of cylinder heating and cooling and increased the engine’s efficiency. 2.1.3. Machine Tools New machine tools and a system for making interchangeable parts were essential to the Industrial Revolution. Various artisans built pre-industrial machinery: millwrights constructed watermills and windmills, carpenters

The Machine Age

77

made wooden framing, and smiths and turners made metal parts. Metal was difficult to process when worked manually with basic hand tools, so the use of metal in machinery was kept to a minimum. The wide use of machines in the textile industry and transport created economies of scale for using more metal parts in machinery, which led to the development of machine tools for cutting metal parts. John Wilkinson invented the first large machine tool in 1775, the cylinderboring machine, used for boring large-diameter cylinders on early steam engines. Boring is the process of enlarging a hole that has already been drilled or cast to achieve greater accuracy in the diameter of a hole using a single-point cutting tool (Forward 1924). Boring can be done by generalpurpose machines such as lathes. A lathe is a machine tool that rotates the workpiece on its axis to perform various operations. Jan Verbruggen installed a horse-powered lathe in 1772 as the horizontal boring machine to produce a more accurate and robust cannon (Jackson and De Beer 1974). Later, lathes were powered by waterwheels or steam engines. Joseph Bramah, the inventor of the hydraulic press, patented a lathe similar to the slide rest lathe (Dickinson 1941). Henry Maudslay perfected the lathe in 1800. It used a lead screw as a linkage to translate turning motion into linear motion and changeable gears between the spindle and the lead screw to adjust the thread pitches to be cut in machining screws. A tool holder, into which the cutting tool would be clamped, would slide on accurately planed surfaces to allow the cutting tool to move in either direction. Maudslay’s slide rest lathe was the first industrially practical screw-cutting lathe, which allowed the standardization of screw thread sizes for the first time and the concept of interchangeable parts to be practically applied to nuts and bolts. The slide rest lathe was one of history’s most influential inventions (Gilbert 1971). Maudslay influenced a generation of men in his workshops to build on his work, such as Richard Roberts, Joseph Clement, and Joseph Whitworth (Smiles 1884). Producing accurate flat surfaces is also essential for machine making. The planing machine uses relative linear motion between the workpiece and a single-point cutting tool to cut the workpiece to generate accurate flat surfaces. Various pioneers developed it in the late 1810s (Smiles 1884). The shaping machine was developed in the early decades of the nineteenth century, and its invention was credited to James Nasmyth (Roe 1916). Its most common use was to machine straight, flat surfaces. It also used relative linear motion between the workpiece and a single-point cutting tool to cut the workpiece.

Chapter 2

78

The milling machine first appeared between 1814 and 1818 in the United States. It used rotary cutters to remove material by many minor cuts for machining parts into precise sizes and shapes. Its invention aided the development of the system of interchangeable parts. In 1854 the Waltham Watch Company developed machine tools, gauges, and assembling methods adapted to the micro precision required for watches to manufacture fully interchangeable movement parts (Carosso 1949). Following the invention of the fundamental machine tools, the machine industry gradually became the largest industrial sector by value added in the US economy, which was the basis for the rise of the US as the leading industrial nation in the world in the late nineteenth century.

2.2. Innovations for Cheaper or New Materials An essential aspect of the Industrial Revolution was the production of new materials or a massive increase in previously undersupplied materials. Before the Industrial Revolution, Britain had imported considerable amounts of iron from Sweden and Russia. Because of the new iron-making technology, imports of iron decreased from 1785, and Britain became an exporter of bar iron and manufactured wrought iron consumer goods. Innovations in chemicals and cement also played essential parts in the Industrial Revolution. 2.2.1. Iron and Steel Making Innovations in iron-making before and during the Industrial Revolution sharply reduced the cost of cast iron and wrought iron so that more goods could use iron for better quality and durability. The first significant innovation was the replacement of charcoal with coal in smelting, using coal reverberatory furnaces based on innovations by Sir Clement Clerke and others from 1678. Coal required much less labor to mine than cutting wood and converting it to charcoal (Gordon 2001) and was more abundant than wood (Landes 1969). In reverberatory furnaces, flames produced by coal acted on the ore that was separated from the fuel to reduce the oxide to metal, so impurities in the coal would not migrate into the metal (Fig.2-5). The technology was applied to iron foundry work in the 1690s (King 2005).

The Machine Age

79

Fig.2-6 A diagram of a reverberatory furnace (by Mrnatural, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=7711169).

Abraham Darby used coke to fuel his blast furnaces at Coalbrookdale in 1709 to produce cast-iron goods such as pots and kettles. In the Horsehay and Ketley furnaces built by his son Abraham Darby II in the mid-1750s, coke pig iron produced bar iron in forges. Since coke pig iron was cheaper than charcoal pig iron, cast iron became more affordable and plentiful. It became a structural material following the building of the Iron Bridge in 1778 by Abraham Darby III. However, bar iron for smiths to forge into consumer goods was still made in finery forges that used charcoal. In 1763, Charles and John Wood patented their potting and stamping process, which melted pig iron in an oxidizing atmosphere, broke up the cooled iron by stamping, washed the iron granules, and heated them in pots in a reverberatory furnace. This process could make bar iron without charcoal for smiths (King 2005, 2011). The potting and stamping method was superseded by the puddling process, also for making bar iron without charcoal, developed in 1784 by Henry Cort (Landes 1969). Puddling decarburized pig iron by slow oxidation, with molten iron manually stirred by a long rod in a reverberatory furnace. The decarburized iron was raked into globs by the puddler, who would remove it when the glob was large enough. Puddling produced a structural grade iron at a relatively low cost and enabled a significant expansion of iron production in Britain and later in North America. Henry Cort also developed the rolling process in 1783, in which metal stock was passed through one or more pairs of rolls to reduce the thickness and to make the thickness uniform. It was 15 times faster than hammering with a trip hammer for consolidating wrought iron and expelling some of the dross (Landes 1969).

80

Chapter 2

In 1828 James Beaumont Neilson patented hot blast, the preheating of air blown into a blast furnace, the most critical development of the nineteenth century for saving energy when making pig iron. Using waste exhaust heat to preheat combustion air reduced the fuel consumed in the pig iron making by one-third for coal or two-thirds for coke initially (Landes 1969); the efficiency gains continued as the technology improved. Hot blasts also increased the capacity of furnaces by raising their operating temperature and reducing impurities in the pig iron because less coal or coke was used. Lower-quality coal or anthracite could be used where coking coal was unavailable or too expensive (Rosenberg and Nathan 1982). Edward Alfred Cowper developed the Cowper stove in 1857, which used firebrick instead of iron as the regenerative heating medium and solved iron’s expansion and cracking problems. It could also produce high heat, which resulted in very high throughput (Allen 1983). The cementation process was invented before the seventeenth century, which carburized wrought iron with charcoal in stone pots inside a furnace to produce low-quality blister steel. Benjamin Huntsman developed his crucible steel technique in the 1740s, which made high-quality steel and used blister steel as raw material. It used a coke-fired furnace capable of reaching 1,600 C. The crucibles inside the furnace were charged with lumps of blister steel and a flux (a chemical cleaning agent such as carbonate of soda) to help remove impurities (Evans and Withey 2012). This technique made high-quality steel more widely available. After its introduction, steel production in Sheffield increased from about 200 tons per year to over 80,000 tons per year in one hundred years, changing Sheffield from a small town into one of Europe's leading industrial cities. The cheaper iron and steel supply aided several industries and stimulated economic growth. 2.2.2. Chemicals The large-scale production of chemicals was a significant development during the Industrial Revolution. John Roebuck invented the lead chamber process in 1746 for the large-scale production of sulfuric acid. He replaced the relatively expensive glass vessels formerly used with larger, less expensive chambers made of riveted sheets of lead (Clow and Clow 1945). He could produce around 100 pounds (50 kg) in each chamber, at least a tenfold increase. Sulfuric acid had been used for pickling iron and steel (removing rust), making other chemicals, bleaching cloth, etc. Another significant progress was the large-scale production of sodium carbonate, which had many uses in the glass, textile, soap, and paper

The Machine Age

81

industries. Nicolas Leblanc introduced the Leblanc process in 1791, which obtained a mixture of sodium carbonate and calcium sulfide by heating the sodium sulfate (produced by a reaction of sulfuric acid with sodium chloride) with limestone (calcium carbonate) and coal. The soluble sodium carbonate could be separated from the calcium sulfide by dissolving it in water (Gillispie 1957). This method was more economical than the previously dominant methods of making soda ash, i.e., burning specific plants (barilla) or kelp. The Belgian industrial chemist Ernest Solvay developed the Solvay process in 1861, which was more economical and less polluting than the Leblanc method. It produced soda ash (sodium carbonate (Na2CO3)) from brine (sodium chloride (NaCl)) and limestone (Kiefer 2002). The development of bleaching powder (calcium hypochlorite) revolutionized the bleaching processes in the textile industry by dramatically reducing the time required (from months to days). The traditional process then in use required repeated exposure to the sun for many months in so-called bleaching fields after soaking the textiles with alkali (usually stale urine) or sour milk. Claude Louis Berthollet introduced chlorine gas as a commercial bleach in 1785 and produced a modern bleaching liquid in 1789 by passing chlorine gas through a sodium carbonate solution. He made potassium chlorate (KClO3), Berthollet's Salt, a strong chlorine oxidant, and bleach. Charles Tennant combined chlorine with lime to produce a calcium hypochlorite solution with sound bleaching effects and obtained a patent in 1788. Then he made solid calcium hypochlorite, the bleaching powder, and patented it in 1799 (Gittins 1979). Charles Tennant and his partners built a factory to produce bleaching liquor and powder at St Rollox, in northern Glasgow, which once became the largest chemical plant in the world. 2.2.3. Cement Cement is a construction substance that sets, hardens, and can bind other materials together. Non-hydraulic cement sets as it dries and reacts with carbon dioxide in the air, so it will not set in wet conditions. Hydraulic cement becomes adhesive due to a chemical reaction between the dry ingredients and water, and it can set even underwater. In 1756 John Smeaton experimented with combinations of different limestones and additives, including trass and pozzolanas, for the planned construction of a lighthouse, the Smeaton's Tower (Ryan 1929). He found that the hydraulicity of the lime was directly related to the clay content of the limestone from which it was made.

Chapter 2

82

James Parker developed “Roman cement,” made by grinding burnt septaria, nodules found in certain clay deposits, into a fine powder, in the 1780s and patented it in 1796 (Hughes, Swann, and Gardner 2007). Roman cement quickly became popular, and its success led other manufacturers to develop rival products. In 1811 James Frost produced “British cement,” using the soft local chalk and alluvial clay from the Medway Estuary as raw materials (Swallow and Carrington 1995). In 1824 Joseph Aspdin patented a process for making Portland cement, which sintered a mixture of clay and limestone to about 1,400 C and then ground it into a fine powder (Ryan 1929). This product was not considered true “modern” Portland cement because it contained no alite (tricalcium silicate, Ca3SiO5). In 1843, Joseph Aspdin’s son William made a significantly different cement, which was slow-setting, high-strength, and suitable for use in concrete, by increasing the limestone content in the mixture and burning it much harder. William’s invention was considered the first “modern” Portland cement. Isaac Charles Johnson further refined its production process. He produced modern Portland cement, the most common type of cement in general use worldwide and an essential ingredient of concrete, mortar, stucco, and grout (Francis 1977). Cement was used on a large scale in constructing the London sewerage system between 1859 and 1865. 2.2.4. Gaslighting Gas lighting became a significant industry later in the Industrial Revolution. Archibald Cochrane, ninth Earl of Dundonald, had already used gas in 1789 for lighting his family estate (Clow and Clow 1942). The large-scale introduction of gas lighting was attributed to the work of William Murdoch (DiLaura 2008). By 1794 Murdoch was producing coal gas from a small retort containing heated coals. In 1802 he made a public exhibition of his lighting by illuminating the exterior of the Soho Foundry. The first gas lighting utilities were established in London between 1812 and 1820. Gas lighting utilities soon became one of the UK's significant coal consumers. Factories and stores could remain open longer with gas lighting than tallow candles or oil. Nightlife began to flourish in cities and towns as interiors and streets could be lit on a larger scale than before.

2.3. Innovations in Transportation Improved transportation infrastructure was also vital in the Industrial Revolution (Evans 1981). At the beginning of the Industrial Revolution, inland transport was by navigable rivers and roads, with coastal vessels

The Machine Age

83

employed to move heavy goods by sea. Humans and animals supplied the motive power on land, and sails provided the motive power on the sea. Transport cost was an additional cost to the production cost and a limiting factor for the geographic size of markets. 2.3.1. By Water Transportation on land used to be much more expensive than by water. In the early nineteenth century, it cost as much to transport a ton of freight 32 miles by wagon over an unimproved road as to ship the same freight 3000 miles across the Atlantic (Taylor 2015). Coastal cities or those on the banks of large rivers tended to be more developed because of their better access to low-cost transportation. Building canals was the first technology to transport bulk materials long distances inland economically. A horse could pull a barge weighing over 30 tons, while it could pull at most one ton of freight on a Macadam road, a type of multi-layer stone-covered and crowned road designed by John McAdam in 1816 (Fay 1937; Brown 2002). The high land transport costs used to limit the size of the markets. Canals began to be built in the late eighteenth century to link the major manufacturing centers and mines across the country in the UK. The Bridgewater Canal in North West England opened in 1761, commissioned by Francis Egerton, the third Duke of Bridgewater, to transport coal from his mines in Worsley to Manchester. Its construction cost £168,000 (£27,686,337 as of 2021), and the price of coal in Manchester fell by about half within a year of its opening (Nevell 2013). New canals were hastily built to replicate its commercial success. The Leeds and Liverpool Canal and the Thames and Severn Canal opened in 1774 and 1789, respectively. This period of intense canal building was known as Canal Mania, and a national network came into existence by the 1820s (Crompton 1993). Many people attempted to build machine-powered ships in the eighteenth century. Denis Papin constructed a vessel powered by a steam engine linked to paddles in 1704. Marquis Claude de Jouffroy built a 13-meter steamer with rotating paddles that sailed on the Doubs River in 1776 and another stele steamer that steamed the river Saône for 15 minutes in 1783. John Fitch and James Rumsey also successfully tested their respective steam-powered boats in 1787. William Symington designed and patented a steam engine with improvements over the Watt engine in 1787 and successfully tested a pleasure boat powered with his steam engine on Dalswinton Loch in 1788. He fitted a larger steam engine on a larger boat (60 feet long) in 1789 (Dumpleton 2002).

84

Chapter 2

In 1802, the first commercial paddle-steamer, Charlotte Dundas, built by Symington for the Forth and Clyde Canal Company, successfully hauled two 70-ton barges almost 20 miles (30 km) in six hours against a strong headwind. Robert Fulton and Robert Livingston built the first commercially successful steamboat, the North River Steamboat (Clermont), after testing their small steamboat in 1803 on the Seine in France. The Clermont went into commercial service in 1807 on the Hudson River between New York and Albany, making the 150-mile trip in 32 hours. The steamboat Phoenix, built by John Stevens, sailed from Hoboken to Philadelphia in 1809 and became the first steamship to navigate the open ocean successfully (Dumpleton 2002). The screw propeller introduced in 1835 by Francis Pettit Smith gradually replaced the paddle wheel used in early steamboats. Smith's steamship (SS) Archimedes used the first steam-driven screw propeller, demonstrating the superiority of screws against paddles on trials. The invention of the surface condenser allowed boilers to run on salt water without stopping to clean the precipitates, making long sea journeys possible (Seaton 2013). The 236 ft (72 m) long Great Western, built by the engineer Isambard Kingdom Brunel from 1836 to 1838, was the first ship to prove that transatlantic steamship services were viable. The ship carrying four sail masts was constructed mainly from wood and propelled by steam-powered paddle wheels. Great Britain, built by Brunel and launched in 1843, was made of metal rather than wood and powered by an engine via a screw propeller. It was considered the first modern ship (Buchanan 2006). Aaron Manby and Joseph Maudslay built the oscillating engine in the 1820s as a type of direct-acting engine designed to achieve further reductions in engine size and weight. The first seagoing iron steamboat powered by an oscillating engine was built by Horseley Ironworks and named the Aaron Manby. She sailed to Paris in June 1822 under Captain (later Admiral) Charles Napier. John Penn perfected the oscillating engine and replaced the Admiralty yacht Her/His Majesty’s Ship (HMS) Black Eagle's engines in 1844 with his machines, which had doubled the power without increasing the weight or space occupied. He also introduced the trunk engine for driving screw propellers in warships. It was a direct-acting engine developed to reduce an engine's height while retaining a long stroke. The trunk engine was efficient compared to competing products and was the first mass-produced, HP, high-revolution marine engine (Smith 1938).

The Machine Age

85

2.3.2. Roads The improvement of the road system in Britain began before the Industrial Revolution. By the mid-seventeenth century, basic infrastructure for stagecoaches, a covered wagon used to carry passengers and goods inside, had been established. The first stagecoach route started in 1610 and ran from Edinburgh to Leith (Merdinger 1952). Turnpike trusts were established in England and Wales from about 1706 in response to the need for better roads than the few poorly-maintained tracks. They charged passage fees to help recoup road construction and maintenance costs. The number of turnpike trusts grew fast from the 1720s, and by the 1750s, almost every main road in England and Wales was the responsibility of a turnpike trust (Bogart 2005). The major turnpikes radiated from London, enabling the Royal Mail to reach the rest of the country. Heavy goods were carried by broad-wheeled carts hauled by teams of horses on these roads and lighter goods by smaller carts or by groups of packhorses. 2.3.3. Railway Using railways to reduce friction has a long history. In Greece during the sixth century BCE, a six-kilometer railway, the Diolkos wagonway, was used to transport boats across the Corinth Isthmus. Narrow gauge railways with wooden rails were common in European mines by 1550. Wagons on these early railways were all drawn by horses or humans (Lewis 1970). When improvements in iron production made pig and wrought iron cheap for building iron rails, horse-drawn public railways were built in the early nineteenth century. A good horse on an ordinary turnpike road could draw one ton, whereas it could draw 36 tons on the railway (Flint 1868). George Stephenson built the first railway using steam locomotives; the eight-mile (13 km) Hetton Colliery railway opened in 1822. The first public railway using steam locomotives, the Stockton and Darlington Railway, was officially opened on 27 September 1825 (Cottrell and Ottley 1975). The Liverpool and Manchester Railway, the first inter-city railway in the world, was opened on 15 September 1830, linking the expanding industrial town of Manchester with the port town of Liverpool (Rees 1978). The success of the inter-city railway led to Railway Mania, which soon spread to other countries. The American railroad mania began with the Baltimore and Ohio Railroad in 1828. Its first section from Baltimore West to Ellicott's Mills (now known as Ellicott City) opened on May 24, 1830 (Flint 1868). Belgium completed its first railway from northern Brussels to

86

Chapter 2

Mechelen in May 1835 (Espion, Engels, and Provost 2017). In France, the first railway line started operation between Saint-Étienne and Andrézieux in 1827 (Desaunais 1934). Then the government stepped in to build a centralized system that radiated from Paris, plus lines that cut east to west in the south. By the 1840s, railways linked the major cities in Germany, and each German state was responsible for the lines within its borders (Heinze and Kill 2019). The first railway line was built in Russia in 1837 between Saint-Petersburg and Tsarskoye Selo (Kazanskaya and Sobor 2019). Railroads would eventually reduce the cost of land transport by over 95% (Bogart 2013), which hugely expanded the market reach of producers, increased consumers’ choices, and stimulated economic growth.

2.4. Social Changes and Reforms The effects of the Industrial Revolution on living conditions have been very controversial. Some economists consider that the Industrial Revolution caused the living standards of ordinary people to undergo sustained growth for the first time in history. Others argue that while the development of the economy was unprecedented during the Industrial Revolution, living standards for most of the population did not grow meaningfully until the late nineteenth century. Studies have shown that real wages in Britain only increased by 15% between the 1780s and 1850s and that life expectancy in Britain did not begin to increase dramatically until the 1870s (Szreter and Mooney 1998). 2.4.1. The Rise of Factories and Urbanization Factories emerged around the beginning of the Industrial Revolution. John Lombe set up a water-powered silk mill at Derby in 1721, using machines for twisting silk into thread. By 1746, an integrated brass mill at Warmley near Bristol smelted ore into brass and turned it into various goods; the mill also provided housing for workers on site. The modern factory, however, rose with the mechanization of cotton spinning, and the large numbers of workers migrating to work from rural communities in the factories contributed to the growth of urban areas. Manchester, nicknamed “Cottonopolis” and the world's first industrial city, best illustrated this process (Williams 1996). Migrant Flemish weavers started the cottage textile industry in Manchester around the fourteenth century. The region became a center for the manufacturing and trading of wool and linen. During the Industrial Revolution, Manchester experienced a nearly nine-fold increase in its population between 1760 and 1830 (Clark 2005).

The Machine Age

87

Bradford is another example of industrialization-led urban growth. In 1801, Bradford was a rural market town of 6,393 people, with wool spinning and cloth weaving in local cottages and farms (Sheeran 2005). The Industrial Revolution led to rapid growth in the production of worsted cloth in which Bradford specialized, and the town soon became known as the world’s wool capital. Bradford produced ample coal to provide the power needed for industrial growth and sandstone for building the mills (Chrystal 2018). As jobs in the textile mills attracted workers, the population reached 182,000 by 1850 (von Viebahn 1852). Industrialization made supporting a large population on a given piece of land possible. Because coal, imported cotton, brick, and slate had replaced wood, flax, and thatch, ground for producing the latter could be cultivated to feed people. Without using much land, early steam engines produced four times more mechanical energy than a workhorse that needed three to five acres for fodder. During 1750–1800, 70% of European urbanization happened in Britain. As the population increased, life expectancy also increased dramatically, which could be attributed to the marked reduction in infant mortality. The percentage of children born in London who died before age five decreased from 74.5% during 1730–1749 to 31.8% during 1810–1829 (Buer 2018). 2.4.2. Child Labor, the Luddites, Trade Unions, and Social Reforms With mechanization, since there was no need for strength, child labor became the choice for manufacturing in the early phases of the Industrial Revolution. Employers could pay a child less than an adult even though their productivity was comparable. Two-thirds of the workers were children in 143 water-powered cotton mills in England and Scotland in 1788 (Galbi 1997). Many children were forced to work in relatively bad conditions for much lower pay than adults, often earning only 10–20% of an adult male’s wage. Beatings and long hours (14 hours a day, six days a week) were common. The first factory act, the Health and Morals of Apprentices Act of 1802, tried to improve the condition of workers but was never put into practice. The 1819 Cotton Mills and Factories Act forbade the employment of children under nine in cotton mills and limited work hours for children of nine to 16 to 12 hours. The Sadler Committee in 1832 investigated the poor working conditions of child labor and some of the abuses in textile factories (Marvel 1977). The passage of the Factory Act of 1833 allowed children from ages nine to 12 not to work more than nine hours a day or 48 hours a

88

Chapter 2

week (Nardinelli 1980). The Mines and Collieries Act of 1842, following the investigation of working conditions in mines by Ashley’s Mines Commission in 1842, stipulated that children under ten could not work in mines. Also, no women or girls could work in the mines. Edwin Chadwick published his report on The Sanitary Condition of the Labouring Population in 1842 (Chadwick 1842). The Factories Act of 1844 limited working hours to 12 per day for women and children and set maximum working hours for children of nine to 13 to nine hours per day. The Factories Act of 1847 stipulated that women and young people could work ten hours a day and a maximum of 63 hours a week. Some industrialists, such as Robert Owen, tried to improve conditions for their workers. He made many efforts to improve conditions for workers at the New Lanark mills. He was often regarded as one of the key thinkers of the early socialist movement (Robertson 1971). A critical cause for restricting child labor and working hours for others was the potential oversupply of labor, with decreased demand for physical strength and increased labor productivity caused by industrialization. The aggregate demand for products would not clear the market if underage children, women, and men were all employed. Thus, without these restrictions, adult males would face severe competition from women and underage children, as demonstrated by Arkwright’s Cromford factory, which used mainly women and children (Vance Jr 1966). Industrialization increased household income, so working families might be less willing to let their underage children work in factories. Therefore, while the Industrial Revolution created more new jobs, it also reduced working hours and restricted the range of people for employment as the application of machines progressed. While the Industrial Revolution created many work opportunities for rural laborers and underage children, it also cost many craft workers their jobs. Lace and hosiery workers near Nottingham first felt the impact of mechanization because of the invention of framework knitting. Many weavers also found they could no longer compete with machines that only required workers with relatively limited or no skills. Many of those turned their animosity toward the machines and began destroying factories and machinery. They became known as Luddites, followers of Ned Ludd, a folklore figure who was supposed to have broken two stocking frames in a fit of rage in 1799. The movement began in Arnold, Nottingham, on 11 March 1811 and spread rapidly throughout England over the following two years (Byrne 2013). The British government used drastic measures to end the movement, but unrest continued in other sectors as they were industrialized.

The Machine Age

89

In the 1830s, large parts of southern Britain were affected by the Captain Swing disturbances, in which agricultural laborers smashed threshing machines and burned hayricks (Jones 2009). The Industrial Revolution concentrated labor into mills, factories, and mines, thus facilitating workers' collective actions to demand better working conditions and higher pay. The workers organized in combination or via trade unions could demand better terms. When employers refused their demands, the unions took strike action by withdrawing all labor and causing a consequent cessation of production. To protect the interests of employers, the Combination Act of 1799 in Britain forbade workers to form any trade union until its repeal in 1824 because of sympathy in society for the plight of the workers. Even then, trade union activities and collective actions were severely restricted. In 1832, James Brine, James Hammett, George Loveless, James Loveless, Thomas Standfield, and John Standfield from Tolpuddle in Dorset founded the Friendly Society of Agricultural Laborers to protest against the gradual lowering of wages in the 1830s. They refused to work for less than ten shillings a week. They were arrested, found guilty, and transported to Australia under the Unlawful Oaths Act 1797 for swearing a secret oath as members of the Friendly Society of Agricultural Laborers (Griffiths 1997). The social changes caused by the Industrial Revolution strengthened the call for democratic reforms. The radical politician John Wilkes developed the belief that every man had the right to vote in the 1760s. John Cartwright (1740–1824) published his pamphlet Take Your Choice! in 1776 and advocated annual parliaments, the secret ballot, and manhood suffrage (Miller 1968). Radical organizations sprang up in the 1790s, demanded universal male suffrage with yearly elections, and expressed their support for the principles of the French Revolution. The government passed the Seditious Meetings Act of 1795, prohibiting unlicensed gatherings of over fifty people. The Spa Fields Riots of 1816 and the Derbyshire Rising of 1817 were followed by the Peterloo Massacre of 1819 (Ellis and Fender 2009). The Six Acts of 1819 limited the right to demonstrate or hold public meetings. The Radical War (the Scottish Insurrection of 1820) of strikes and unrest was repressed by government forces (Donnelly 1976). The Reform Act of 1832 granted seats in the House of Commons to large cities that had sprung up during the Industrial Revolution and removed seats from the “rotten boroughs,” increasing the electorate from about 500,000 to 813,000 in a population of 14 million (Phillips and Wetherell 1995). The Chartist Movement from 1838 to 1858 was the first large-scale working-

Chapter 2

90

class political movement that campaigned for political equality and social justice. Trade unions slowly overcame the legal restrictions on the right to strike. After the second Chartist Petition was presented to parliament in April 1842 and rejected, a general strike involving cotton workers and colliers was organized, stopping production across Great Britain (Jenkins 1980). Reform Act of 1867 extended the voting franchise to men in urban areas who paid rates in person. Reform Act 1884 extended it to all men paying an annual rental of £10 or holding land valued at £10. The British electorate then totaled over 5,500,000 (McKenna 2019).

3. The Second Industrial Revolution The Industrial Revolution ended in the early-mid-nineteenth century, which was followed by a transition period between 1840 and 1870 when technological and economic progress continued with the increasing adoption of steam-powered transport, the large-scale manufacture of machine tools, and the growing use of machinery in steam-powered factories (Landes 1969; Taylor 2015). During these transition years, the Industrial Revolution evolved into the Second Industrial Revolution, a phase of rapid industrialization generally dated between 1870 and 1914 (Mokyr 1998). Several new areas characterize the Second Industrial Revolution, including large-scale steelmaking, petroleum, internal combustion engines, automobiles, electrification for industrial mass production, and electrical communications.

3.1. Energy and Power Sources The (First) Industrial Revolution was powered by waterwheels and steam engines, whereas internal combustion engines and electricity powered the Second Industrial Revolution in addition to steam engines. Internal combustion engines are fueled by petrol and diesel oil from petroleum. Automobiles powered by internal combustion engines have become essential consumer goods in modern society. 3.1.1. Petroleum The first application of petroleum was to produce kerosene for lamp oil and then petrol or diesel oil for internal combustion engines and jet fuel for jet engines. Petroleum was used as early as the third and fourth centuries. China’s earliest oil wells were drilled in 347 CE (Totten 2004). Oil fields were claimed to have been exploited from the ninth century around modern Baku, Azerbaijan, to produce naphtha. In 1806, the Russian Empire occupied Baku Khanate and monopolized oil production. Oil extraction

The Machine Age

91

methods in those times were very primitive: wells were mainly hand-dug and drilled to very shallow depths. In 1846, a well was drilled with percussion tools to 21 meters for oil exploration in Baku. In 1846, the Canadian geologist Abraham Gesner distilled a clear thin fluid from coal that he showed as an excellent lamp fuel and named it kerosene. As the cost of extracting kerosene from coal was high, he used a naturally occurring asphaltum called albertite. Gesner obtained his first kerosene patent in 1854 and formed the North American Gas Light Company to manufacture kerosene from bituminous coal and oil shale in New York (Beaton 1955). In 1847, James Young found that by slow distillation of a natural petroleum seepage in the Riddings colliery at Alfreton, Derbyshire, he could obtain a light, thin oil suitable for use as lamp oil which he named “paraffin oil” and a thicker oil suitable for lubricating machinery. He set up a small business refining crude oil in 1848. On October 17, 1850, he patented a method for extracting paraffin oil from cannel coal (a type of oil shale) (Hassan 1978; Dickey 1959). In 1850 Young set up the Bathgate chemical works, the first genuinely commercial oil works and oil refinery in the world, using oil extracted from locally mined torbanite, shale, and bituminous coal to manufacture paraffin oil (kerosene) and lubricating oil. Samuel Martin Kier experimented with several distillates of the crude oil from his salt wells around Livermore and nearby Saltsburg. He began selling kerosene made from crude oil to local miners in 1851. He also invented a new lamp to burn his product. Kerosene lighting was much more efficient and less expensive than vegetable oils, tallows, and whale oil. Kier established America's first oil refinery in Pittsburgh in 1853 (Cadman 1959). Jan Józef Ignacy àukasiewicz distilled kerosene from seep oil and invented the modern kerosene lamp in 1853. He opened the oil “mine” at Bóbrka, near Krosno, in 1854 and several other oil wells in the subsequent years. In 1856 in Ulaszowice, near Jasáo, he opened the world's first industrial oil refinery, producing artificial asphalt, machine oil and lubricants, and kerosene (Pabis-Braunstein 1989). Canada’s first commercial oil well became operational in 1858 at Oil Springs, Ontario (then Canada West) (Crain and Eng 2004). On August 27, 1859, Edwin L. Drake struck oil in Cherrytree, Pennsylvania (Black and Ladson 2010), using cable tool drilling developed in ancient China for drilling brine wells and introduced to Europe in 1828 (Temple, Needham, and Biochemiker 2007). The Drake Well attracted the first wave of oil drilling, refining, and marketing investments. The increased supply of

92

Chapter 2

petroleum in North America allowed oil refiners to produce illuminating oil from it without paying royalties to kerosene patent holders. Kerosene produced a brighter light than gas lighting, and the demand for kerosene made it one of the leading petroleum products (Cadman 1959). 3.1.2. Internal Combustion Engines The compactness and higher energy efficiency of internal combustion engines invented during the Second Industrial Revolution created more applications that were impossible or difficult to use steam engines. Isaac de Rivaz patented in 1807 probably the world's first internal combustion engine, using hydrogen gas as a fuel (Guarnieri 2011). Jean Joseph Etienne Lenoir created the first commercially successful internal combustion engine using illuminating gas around 1859 and patented it in 1860 (Payen 1963). In 1863, he built the Hippomobile with a hydrogen gas-fueled internal combustion engine (Boretti 2011). Nikolaus August Otto and Eugen Langen improved on Lenoir’s machine and built an atmospheric engine in 1864. In 1876, Otto developed the Otto cycle engine, in which the fuel and air were mixed after compression, and a spark plug initiated the combustion process. It could run on many different fuels, including coal gas and gasoline, and became the first widely used internal combustion engine (Boretti 2011). It was used initially in small shops because small steam engines were inefficient and later more successful in automobiles. The mass production of cars after 1914 led to gasoline shortages during World War I. The Burton process for thermal cracking was invented to increase gasoline yield, which helped alleviate the shortages (Wilson 1928). The diesel engine is an internal combustion engine and was independently designed by Rudolf Diesel and Herbert Akroyd Stuart in the 1890s (Cummins Jr 1976). Fuel ignition, which is injected into the combustion chamber, is caused by the elevated temperature of the air in the cylinder due to mechanical compression (adiabatic compression). Compressing the atmosphere increases the air temperature inside the cylinder to such a high degree that the atomized diesel fuel ignites spontaneously. The diesel engine is more energy efficient than the Otto engine and has the highest engine efficiency of any practical internal or external combustion engine. It is run on diesel oil distilled from petroleum at a higher yield and is widely used in factories, ships, locomotives, and automobiles. Diesel engines made ocean shipping the cheapest mode of long-distance transport and promoted economic globalization (Smil 2007).

The Machine Age

93

3.1.3. Electricity The development of electrical industries was closely related to the scientific findings in electromagnetism. In 1600, William Gilbert published his study of electricity and magnetism (Pumfrey and Tilley 2003). In 1663, Otto von Guericke invented the first electric generator, which produced static electricity by applying friction in the machine (Dibner 1954). In 1799 Alessandro Volta invented Volta’s battery, made from alternating layers of zinc and copper (Volta 1800). Hans Christian Ørsted discovered that electric currents created magnetic fields on 21 April 1820, while André-Marie Ampère developed a mathematical and physical theory on the relationship between electricity and magnetism (Rautio 2014; Al-Khalili 2015). The scientist and experimentalist Michael Faraday discovered the principles of electromagnetic induction, diamagnetism, and electrolysis laws, among others. He also constructed the electric dynamo during 1831–1832 and the electric motor in 1821 based on his findings (Al-Khalili 2015). Electric lighting was an important early use of electricity. Probably in the first decade of the nineteenth century, Humphry Davy invented the carbon arc light, which consisted of an arc between carbon electrodes in the air. He used a 2000-cell battery to power the lamp (Harris 1993). The lack of a constant electricity supply hindered its wide application. In the midnineteenth century, electrical engineers began focusing on improving Faraday's dynamo to provide continuous electricity. Antoine Hyppolite Pixii presented an electromechanical generator capable of higher electromotive forces in 1832 (Guarnieri 2018). The Woolrich Electrical Generator, designed by John Stephen Woolrich and built in February 1844 at the Magneto Works of Thomas Prime and Son, Birmingham, was the earliest electrical generator used in an industrial process (commercial electroplating) (Guarnieri 2018; Hunt 1973). Ányos Jedlik found the principle of dynamo self-excitation while building an electric motor that did not have a permanent magnet in the 1850s, but his findings were overlooked (Halacsy 1971). Antonio Pacinotti invented a direct-current electrical generator using a ring armature wrapped around a coil of wire in 1864 (Guarnieri 2018). In 1867, Werner von Siemens also described a dynamo without permanent magnets, and Siemens became the first company to build such devices (Feldenkirchen 1994). Zénobe Gramme introduced the first dynamo capable of producing smooth high currents and voltages in 1869 and put it into commercial production in 1871, marking the beginning of the large-scale production and use of electric power (Guarnieri 2013a, 2018).

94

Chapter 2

In 1870, water from one of the lakes in Lord Armstrong’s Cragside estate was used to drive a Siemens dynamo to provide electricity to the building on the estate (Benn 2021). The electricity generated by this first hydroelectric power station in the world was used to power an arc lamp installed in the picture gallery in 1878 (Irlam 1989). Charles F. Brush developed a dynamo system in 1877, which performed best in a comparative test conducted by the Franklin Institute in the United States. His improved dynamo provided electricity to arc lighting in Public Square in Cleveland, Ohio, on April 29, 1879 (Lamoreaux, Levenstein, and Sokoloff 2006). Sir Joseph Swan invented the first feasible incandescent light bulb, demonstrated on 3 February 1879 before an audience at the lecture theater of the Literary and Philosophical Society of Newcastle upon Tyne. Thomas Edison produced his first feasible light bulb on October 21, 1979 (Broad 1979). The arc lamp in Lord Armstrong’s Cragside estate was replaced in 1880 by incandescent lamps. In 1881, the Savoy Theatre in London became the first theater and the first public building in the world lit entirely by electricity, with 824 lamps on the stage and 370 more in parts of the house. HMS Inflexible was also lit by the incandescent light bulb in 1881, and James Coxon, a draper in Newcastle, had the first shop to be lit by electricity in the world (Chirnside 1979; McKenzie 2018). In September 1882, Edison’s American company put into service in Pearl Street in Manhattan, New York, the first system for the commercial distribution of electric light from a centralized power station. Such direct current (DC) systems proved more suitable for the short-distance distribution of electricity (Guarnieri 2013a). Alternating current (AC) systems can exploit the capability of transformers to step voltages up and down and transmit electric power over long distances. One of the first operative AC systems was made for public lighting in 1885 in Rome and powered by two Siemens & Halske alternators rated 30 hp (22 kW), 2 kV at 120 Hz. Sebastian Ziani de Ferranti established Ferranti, Thompson, and Ince in 1882 to market his Ferranti-Thompson Alternator (Williams 1987). He designed the Deptford Power Station for the London Electric Supply Corporation in 1887 using an AC system on an unprecedented scale. On its completion in 1891, it generated 800 kilowatts and supplied central London with high-voltage (10,000V) AC power that was then transformed to low-voltage for consumer use on each street (Guarnieri 2013b). The steam turbine developed by Sir Charles Parsons in 1884 made cheap and plentiful electricity possible (Scaife 1985). The turbine produced rotary

The Machine Age

95

power rather than reciprocating power, and its large number of stages allowed for high efficiency and reduced its size by 90%. His first model was connected to a dynamo that generated 7.5 kW (10 hp) of electricity (Williams 2018). In 1889, he founded C. A. Parsons and Company in Newcastle to produce turbo generators. He set up the Newcastle and District Electric Lighting Company, which opened Forth Banks Power Station in 1890, the first power station in the world to generate electricity using turbo generators (Gibb 1947). By Parson’s death in 1931, his turbine had been adopted for all major power stations worldwide. William Sturgeon invented the first commutator DC electric motor capable of turning machinery in 1832 (Fleming 1925). Moritz von Jacobi created the first actual rotating electric motor in 1834 with a remarkable mechanical output power. His second motor was powerful enough to drive a boat with 14 people across a wide river in 1839 (Zhang and Yang 2020; Michalowicz 1948). Thomas Davenport also built a rotating electric motor in 1834, patented it in 1837, and used it to power the first electric printing machine in 1839 (Michalowicz 1948). The Scottish inventor Robert Davidson developed another electrical motor suitable for practical operation between 1837 and 1842. He used it to power the first full-scale electric locomotive, Galvani, 16-ft long (5 m) and weighed 6 tons (Guarnieri 2018). Zénobe Gramme showed that Pacinotti's dynamo could be used as a motor in 1873, and the Gramme machine, based on such a design, was the first industrially successful electrical motor (Guarnieri 2018). Fyodor Pirotsky, in 1875, invented and tested the world's first electric tram line in Sestroretsk, Russia, and the first public electric tramway in St. Petersburg in 1880. The second line was built by Werner von Siemens in Lichterfelde, Germany, and opened in 1881. In 1883, Volk's electric railway was opened in Brighton, Britain, and Mödling and Hinterbrühl Tram opened in Austria (Guarnieri 2020). Frank J. Sprague founded the Sprague Electric Railway & Motor Company in 1884 and developed the first DC motor that maintained constant speed under varying loads in 1886. He also invented regenerative braking, a method of braking that used the drive motor to return power to the main supply system. He improved the designs of electric streetcars collecting electricity from overhead lines. Sprague installed the first successful extensive electric street railway system, the Richmond Union Passenger Railway in Richmond, Virginia, which began passenger operation on February 2, 1888. Sprague’s system spread to around 100 large cities in the world in subsequent years (Guarnieri 2020). The electric street railway became a major infrastructure before 1920.

96

Chapter 2

Galileo Ferraris independently invented an AC electric motor (Induction motor) in 1885, and Nikola Tesla in 1887 (Jarvis 1969; Ahmadinia 2014). They soon began to be used in the electrification of industry. Electric motors were more convenient to arrange on the factory floors than steam engines and internal combustion engines so that they could be installed according to the needs of the production processes. They were also easy to maintain. Electrification enabled the assembly line and mass production because it allowed the placement of machine tools in the order of the workflow. Electric lighting in factories would reduce the heat, pollution, and fire hazard caused by gas lighting, such that the reduction in fire insurance premiums often offset the cost of electricity for lighting. Electrification also allowed the inexpensive production of electrochemicals; for example, aluminum, chlorine, sodium hydroxide, and magnesium began to be produced by electrolysis (Grotheer et al. 2006).

3.2. Innovations in Materials The Second Industrial Revolution saw new steelmaking technologies, chemicals, and substances invented. The increased supply of existing materials and the invention of novel materials created new industrial sectors, which provided more employment. They contributed to economic growth during and following the Second Industrial Revolution. 3.2.1. Steelmaking Steelmaking is often cited as the first of those new areas. The innovation in steelmaking occurred before 1870. Sir Henry Bessemer patented the Bessemer process in 1856, in which partial decarburization of the pig iron by oxidation was achieved by blowing air through the molten iron (Gale 1973). Firms using his method, however, could not get good quality steel because of impurities in the iron. The problem could be solved by turning off the airflow to let the impurities burn off, and the right amount of carbon would remain, but the solution was difficult to implement. Robert Forester Mushet found a simple solution in 1856 of burning off all the impurities and carbon and then reintroducing carbon and manganese by adding an exact amount of spiegeleisen, a ferromanganese alloy containing approximately 15% manganese and small quantities of carbon and silicon (Bishop 1959). Mushet's solution had the effect of improving the quality of the finished product and increasing its malleability. Before the invention of the Bessemer process, crucible steel rolled into bars was sold at £50 to £60 (approximately £6,640 to £7,968 in 2023) per long ton (1,016 kg). The

The Machine Age

97

Bessemer process incorporating Mushet's solution sharply reduced the cost of steel to £7 per long ton, making steel cheaper and more widely available in various applications (Warren 1998). Mushet invented “R Mushet's Special Steel” (RMS) in 1868 which was both the first true tool steel and the first air-hardening steel (Bishop 1959; Warren 1998). The refractory lining of the converter used in the original or acid Bessemer process used clay, which required low phosphorus in the raw material, such as relatively scarce hematite ore. Sidney Gilchrist Thomas developed the “basic Bessemer process” to eliminate the phosphorus from iron and patented his process in 1878 (Habashi 2013). His converter used dolomite, limestone, or magnesite linings, and the oxygen in the blast of air oxidized carbon and other impurities. The addition of lime at this stage caused the oxides to separate as a slag on the surface of the molten metal, which could be profitable as a phosphate fertilizer. The “basic Bessemer process” was especially valuable on the continent of Europe, where the proportion of phosphoric iron was much larger than in England. Sir Charles William Siemens developed a regenerative furnace in the 1850s, using regenerative preheating of fuel and air for combustion (Siemens 1862). He claimed the method could save 70–80% of the fuel, and an open-hearth furnace could reach temperatures high enough to melt steel. Pierre-Émile Martin was the first to apply it to steel production in 1865. The open-hearth furnace was easier to control and permitted the melting and refining of scrap steel, which lowered steel production costs and recycled otherwise troublesome waste material. It displaced the Bessemer process and became the leading steelmaking process by the early twentieth century (Husson 1963). The open-hearth process provided the sheet steel that enabled large HP boilers and high-tensile steel for machinery. The availability of cheap steel allowed the building of large bridges, railroads, skyscrapers, and large ships with steel, which in turn stimulated economic growth. Steelmaking became a pillar industry of the US economy. 3.2.2. New Chemicals The late nineteenth century saw a significant increase in the quantity and variety of chemicals manufactured. Prominent among the chemicals were synthetic dyes and fertilizers. The massive growth of the textile industry during the Industrial Revolution stimulated the demand for dyes. Still, at the time, all dyes used for coloring cloth were natural substances, many of which were expensive and labor-intensive to extract. The discovery or invention of a synthetic dye by William Henry Perkin in 1856 started a new

98

Chapter 2

field of the chemical industry (Robinson 1957). Sir John Bennet Lawes pioneered the production of artificial, manufactured fertilizer, which evolved into another essential branch of the chemical industry and profoundly impacted agriculture (Warington 1900). Perkin accidentally found a purple dye, later named “mauveine,” in 1856 when he tried to synthesize quinine as instructed by his professor, August Wilhelm von Hofmann, by oxidizing aniline with potassium dichromate. He found that the reaction product, when extracted with alcohol, produced a substance with an intense purple color. Before that, purple cloth had used the expensive Tyrian purple, made from a mucous secretion from some predatory sea snails. Perkin commercialized it as the world’s first synthetic dye (Robinson 1957). Aniline used to be expensive. The Béchamp reduction devised by Antoine Béchamp in 1854 reduced aromatic nitro compounds to their corresponding anilines (Béchamp 1854). This inexpensive method to produce aniline permitted Perkin to launch the synthetic dye industry. After the discovery of mauveine, many new aniline dyes appeared, and factories producing them were built across Europe, especially in Germany. BASF (Badische Anilin und Soda Fabrik) was founded on 6 April 1865 in Mannheim to make soda, acids, and other chemicals necessary for dye production (Kreimeyer et al. 2015). Heinrich Caro working for BASF, developed the synthesis of alizarin in 1869, eosin in 1871, and methylene blue in 1876 (Travis 1991). He and Adolf von Baeyer synthesized the first indigo dye in 1878 (Schmidt 1997). Alizarin is commonly used for the derivatization of other dyes. The shade produced by alizarin depends on the metal oxide used to mordant a textile: aluminum yields a red; iron, a dark violet; and chromium, a reddishbrown. Eosin is used mainly for producing a blood-red color in silk, wool, paper, leather, and cotton. Indigo dye is the blue often associated with blue jeans. Friedrich Bayer et Compagnie was founded on August 1, 1863, in Barmen as a general partnership by dye salesman Friedrich Bayer and master dyer Johann Friedrich Weskott to manufacture and sell synthetic dyestuffs (Schadewaldt 1975). It experienced impressive growth and became a jointstock company in 1881. Later, Bayer added pharmaceuticals to its product portfolio and produced aspirin and other drugs. Hoechst AG (now part of the Sanofi-Aventis pharmaceuticals group) was founded in 1863 to produce synthetic dyes. The German chemical industry quickly began to dominate the field of synthetic dyes. By 1913, the German industry produced almost

The Machine Age

99

90 percent of the world’s supply of dyestuffs and sold about 80 percent of their production abroad (Aftalion 1991). Justus von Liebig, in the 1830s and 1840s, identified the chemical elements of nitrogen (N), phosphorus (P), and potassium (K) as essential to plant growth. He promoted using inorganic minerals for plant nutrition (Scharrer 1949). He attempted to commercialize a fertilizer created by treating the phosphate of lime in a bone meal with sulfuric acid. Sir James Murry performed the first commercial superphosphate production by treating crushed bones with sulfuric acid. John Bennet Lawes successfully commercialized it, the first product of the nascent artificial fertilizer industry. The English patents for the manufacture of superphosphate were granted to Murray and Lawes on May 23, 1842 (Ivell 2012). Later, Edward Packard, James Fison, and other entrepreneurs also started their businesses in phosphate fertilizer. The commercial phosphate fertilizer industry was launched (Ford and O’Connor 2009). Guano had been an essential natural fertilizer with high nitrogen, phosphate, and potassium content and was eclipsed by saltpeter as a source of nitrogen fertilizer after 1870. Guano and saltpeter were important exports from Peru and Chile (Sutton et al. 2013). In 1903, Kristian Birkeland and Sam Eyde developed the Birkeland-Eyde process to fix atmospheric nitrogen (N2) into nitric acid (HNO3) and provide a way of nitrogen-based fertilizer production (Eyde 1909). It was soon replaced by the much more efficient Haber or Haber-Bosch process, developed by the Nobel prize-winning chemists Carl Bosch and Fritz Haber in Germany (Modak 2002). The process converts atmospheric nitrogen (N2) to ammonia (NH3) by a reaction with hydrogen (H2) using a metal catalyst under high temperatures and pressures. The annual world production of synthetic nitrogen fertilizer is over 100 million tons. The food supply of the current world population depends on the HaberBosch process. 3.2.3. Rubber, Synthetic Fibers, and Plastics The vulcanization of rubber converts natural rubber into more durable materials by adding sulfur or other curatives. This invention by the American Charles Goodyear and the Briton Thomas Hancock in the 1840s paved the way for a growing rubber industry (Guise-Richardson 2010). Vulcanized rubber is an effective sealing material for small gaps between moving machine parts. It is also essential for manufacturing rubber tires, especially for the automobile and modern bicycle industries. Harry John Lawson introduced the chain drive to the bike, connecting the frame-

100

Chapter 2

mounted crank pedals to the rear wheel in 1876 (Malizia and Blocken 2020). John Kemp Starley produced the first commercially successful safety bicycle, Rover, in 1885. Solid rubber tires were used initially, and the introduction of pneumatic (inflatable air-filled) bicycle tires facilitated the popularity of the modern bicycle. Robert William Thomson patented pneumatic tires in Britain in 1845 (Shearer 1977). John Boyd Dunlop developed the first practical pneumatic tire in 1887 in South Belfast (Jaura 2013). Alexander Parkes patented Parkesine, a celluloid based on nitrocellulose treated with various solvents, in 1856. In 1865, Paul Schützenberger discovered cellulose reacts with acetic anhydride to form cellulose acetate. Improving on Parkes' invention, John Wesley Hyatt developed celluloid in 1868 by plasticizing the nitrocellulose with camphor, which was used for making a photographic film by Kodak and other suppliers (Rasmussen 2021). Hilaire de Chardonnet invented the first artificial silk made from nitrocellulose in the 1880s. In 1894 Charles Frederick Cross, Edward John Bevan, and Clayton Beadle created the fiber “viscose” made from xanthate, the reaction product of carbon disulfide and cellulose in basic conditions. The first commercial viscose rayon was produced by the UK company Courtaulds Fibers in 1905. Camille Dreyfus and his younger brother Henri were producing films for the motion picture industry and acetate lacquer for coating fabrics from cellulose acetate by 1910. They made excellent laboratory samples of continuous filament yarn in 1913 (Morgan 1981). Viscose and cellulose acetate are called artificial fibers, as cellulose is from plants. Bakelite, the first fully synthetic thermoset, was invented in 1907 by Leo Baekeland using phenol and formaldehyde and patented in 1909 (Crespy, Bozonnet, and Meier 2008). Polyvinyl chloride (PVC), first created in 1872 but commercially produced in the late 1920s, and polystyrene (PS), first produced by BASF in the 1930s, were the earliest examples of polymer plastics. Durite Plastics Inc. first manufactured phenol-furfural resins in 1923. Imperial Chemical Industries (ICI) researchers Reginald Gibson and Eric Fawcett discovered polyethylene in 1933. Polypropylene was found by Giulio Natta in 1954 and began to be manufactured in 1957. Dow Chemical invented expanded polystyrene (used for building insulation, packaging, and cups) in 1954 (Andrady and Neal 2009). Nylon, the first synthetic fiber, was developed by Wallace Carothers at the chemical firm DuPont in the 1930s. It soon made its debut in the United States as a replacement for silk, just in time for the introduction of rationing

The Machine Age

101

during World War II. Its use as a material for women's stockings overshadowed more practical benefits, such as replacing silk in parachutes and other military uses. Polyester was patented in Britain in 1928 by the International General Electric company (Loasby 1951). John Rex Whinfield and James Tennant Dickson produced the first polyester in 1941, polyethylene terephthalate (PET) fiber, which they named terylene, also known as Dacron, and patented it (Vogler 2016). The fiber equals or surpasses nylon in toughness and resilience, and PET is also used as plastic. The world production of synthetic fibers was 76.5 million tons in 2019 (Fernández 2021). Synthetic materials have greatly expanded the range of materials and products for humanity to improve their lives and strengthen their ability to make more improvements.

3.3. Communications and Transportation The Second Industrial Revolution laid the foundations of modern society in telecommunications and transportation. Telegraphy (invented during the transition years after the First Industrial Revolution), the telephone, and wireless communications shortened the time for communities and individuals to share information and learn new knowledge. Steel railway networks, automobiles, modern ships, and airplanes shortened the distance for people and commodities to move around the world, so people could quickly get resources from farther distant places, accelerating economic growth and social progress. These innovations transformed the ways people live, work, and interact with each other. 3.3.1. Telecommunications Telegraphy is the long-distance transmission of textual or symbolic (as opposed to verbal or audio) messages without the physical exchange of an object bearing the message. Electrical telegraphy is telegraphy that uses electrical signals. Sir William Fothergill Cooke and Charles Wheatstone developed and patented the first electrical telegraph in May 1837. It used several needles to point to letters of the alphabet, and a four-needle system installed between Euston and Camden Town in London was successfully demonstrated on 25 July 1837 (Guarnieri 2019). They installed a fiveneedle, six-wire system on the Great Western Railway over the 13 miles from Paddington station to West Drayton in 1838, the first commercial telegraph in the world (Huurdeman 2003). Samuel Morse developed an electrical telegraph using the Morse code signaling alphabet with his assistant Alfred Vail and patented it in the

102

Chapter 2

United States in 1837. The first telegram in the United States was sent by Morse on 11 January 1838 across two miles of wire at Speedwell Ironworks, New Jersey (Guarnieri 2019). The Morse/Vail telegraph was quickly deployed in the following two decades; the overland telegraph connected the west coast to the east coast by 24 October 1861. John Watkins Brett laid the first undersea cable between France and the UK in August 1850 (unsuccessfully) and September 1851. In 1853 cables were laid successfully to link Britain with Ireland, Belgium, and the Netherlands and across the Belts in Denmark. The ship SS Great Eastern completed the transatlantic cables on 18 July 1866. By 1890, an international telegraph network allowed orders to be placed by merchants in the UK or the US to suppliers in India and China for goods to be transported in efficient new steamships (Headrick and Griset 2001). The telephone was patented in 1876 by Alexander Graham Bell and was initially used mainly to speed up business transactions (Gorman 1995). The first telephones were directly connected to users in pairs, quickly replaced by manually operated central switchboards, making it possible for subscribers to call people at different locations from one telephone line. The first experimental telephone exchange was based on the ideas of Tivadar Puskás and was built by the Bell Telephone Company in Boston in 1877. George W. Coy designed and made the first commercial US telephone exchange in New Haven, Connecticut, in January 1878. In 1887 Puskás introduced the multiplex switchboard. On March 10, 1891, Almon Brown Strowger patented the stepping switch, which led to the automation of telephone circuit switching (Patil 2015). James Clerk Maxwell's electromagnetic theory unified light, electricity, and magnetism and provided the scientific foundation for electromagnetic wave applications. David Edward Hughes and Heinrich Hertz demonstrated electromagnetic waves (Susskind 1964). The Italian inventor Guglielmo Marconi founded the Wireless Telegraph & Signal Company in Britain in 1897 and sent the first wireless communication over the open sea on 13 May 1897. The first transatlantic transmission was made in 1901 from Poldhu, Cornwall, to Signal Hill, Newfoundland. Marconi built high-powered stations on both sides of the Atlantic to communicate with ships at sea and began commercial service to transmit nightly news summaries to subscribing ships in 1904 (Brittain 2004). In 1904, Sir John Ambrose Fleming developed the vacuum tube, underpinning the development of modern electronics and radio broadcasting (Dylla and Corneliussen 2005). Lee De Forest's subsequent invention of the triode allowed the amplification

The Machine Age

103

of electronic signals (Delogne 1998), which paved the way for radio broadcasting in the 1920s. 3.3.2. A Railway with Steel Rails The increase in steel production and the decrease in price from the 1860s meant that rails could finally be made from steel at a competitive cost. In 1857 Robert Forester Mushet at the Darkhill Ironworks in Gloucestershire was the first to make durable rails of steel (Swank 1904). Before that, rails were made from wrought iron, which was soft and often contained flaws such that iron rails could not support heavy locomotives. Steel is a much more durable material with greater strength, so rails of longer lengths could be rolled, and they need less maintenance than wrought iron ones. The Mushet’s steel rails were first sent to Derby Midland railway station and laid at part of the station approach, where the iron rails had to be renewed at least every six months. These rails still seemed as perfect as ever in late 1863, after six years with some 700 trains passing over it daily (Rolt 1970). This showed that steel rails lasted over ten times longer than iron rails. Since then, steel rails steadily replaced iron ones as the standard for railway rails and became the basis for the development of rail transportation worldwide. With the falling cost of steel, heavier-weight rails were used, allowing the use of more powerful locomotives. Longer trains and longer rail cars pulled by more powerful locomotives significantly increased the productivity of railways, making railways the dominant form of transport throughout the industrialized world. 3.3.3. Automobiles Steam-powered road vehicles were developed between 1867 and 1869 by Sylvester H. Roper (Limebeer and Sharp 2006) and Louis-Guillaume Perreaux and Pierre Michaux (Falco 2003). Gottlieb Wilhelm Daimler and Wilhelm Maybach built their first Standuhr (Grandfather Clock) engine, an Otto engine, and produced the first motorcycle with it, the Daimler Reitwagen (“riding car”) in 1885 (Alford and Ferriss 2007). Karl Benz patented the world's first automobile, the three-wheeled Motorwagen with a rear-mounted four-stroke engine, in 1886 (Nübel 1987). He began to sell the Motorwagen in the late summer of 1888, making it the first commercially available automobile in history. Daimler and Maybach built the Stahlradwagen (“steel wheeled car”) in 1889 using the engine they developed (Nübel 1987), and Peugeot, a steel foundry, led by Armand

104

Chapter 2

Peugeot began building cars based on the Stahlradwagen design by 1890 (Sievers 2005). Henry Ford built his first car in 1896 and founded the Ford Motor Company in 1903. He and others at the company struggled to scale up production to produce affordable vehicles for the average worker. Ford Motor completely redesigned the factory with machine tools and special-purpose machines systematically positioned in the work sequence (Ford and Crowther 1922). All work and instruments were placed within easy reach, and unnecessary human movements were eliminated wherever necessary on conveyors to form the assembly line. The Ford Model T production used 32,000 machine tools, most of which were powered by electricity (Hounshell 1985). Ford Motor developed this mass production process for the first time in history, in which a large, complex product consisting of 5000 parts had been produced on a scale of hundreds of thousands per year (Ford and Crowther 1922). The mass production method allowed the Model T price to decline from $780 in 1910 to $360 in 1914 and $290 in 1924. By 1927, more than two million Model Ts were produced (McCarter 2011). Automobiles further increased people’s mobility, became a necessity for American households, and made the US a nation on wheels. Automobile manufacturing was one of the pillar industries of the US economy in the twentieth century. 3.3.4. Shipbuilding Advances in steam engines and metallurgy led to the emergence of ironclad warships in the late 1850s and 1860s. An ironclad ship is a steam-propelled warship protected by iron or steel armor plates. The first ironclad battleship, Gloire, was launched by the French Navy in November 1859. In early 1859 the Royal Navy started building two iron-hulled armored frigates. The Devastation-class turret ships built in the 1870s for the British Royal Navy were the first modern battleships. They were the first class of ocean-going capital ships that did not carry sails and the first warships whose entire main armament was mounted on top of the hull (Sandler 1970). Improvements in steam efficiency allowed ships to carry much more freight than coal, making transoceanic shipping economically viable and significantly increasing international trade volumes. The multiple expansion steam engines with a series of cylinders of progressively increasing diameter were designed to divide the work into equal shares for each expansion stage. The first successful commercial use was a triple-expansion engine built by Alexander C. Kirk for the SS Aberdeen in 1881 (Broeze 1989). The four-cylinder triple-expansion engine was popular with large

The Machine Age

105

passenger liners because it was smoother, faster-responding, and with less vibration. Parsons’ steam turbine revolutionized marine transport and naval warfare (Harris 1984). The turbine-powered yacht, Turbinia, built in 1894, demonstrated its superiority with 34 knots at the Spithead Navy Review in 1897 and set the standard for the next generation of steamships. HMS Viper and Cobra destroyers were built and powered within two years with Parsons’ turbines. The first turbine-powered passenger ship, Clyde turbine steamer (TS) King Edward launched in 1901; the first turbine transatlantic liners, Royal Mail Ship (RMS) Victorian and Virginian, in 1905; and the first turbine-powered battleship, HMS Dreadnought, in 1906, were all driven by Parsons' turbine engines. Diesel engines were first used to power a French canal boat and a much larger vessel on the Volga in 1903. The first oceangoing ship with diesel engines, Selandia, was launched in November 1911, but it was until the 1920s did diesel engines begin to make serious commercial inroads in powering ships. Oceanic shipping powered by diesel engines became the cheapest long-distance transport mode, facilitating international trade growth (Smil 2007). 3.3.5. Airplanes In 1799, George Cayley began to work on the concept of modern airplanes (Ackroyd 2011). Between 1867 and 1896, the German aviation pioneer Otto Lilienthal studied heavier-than-air flight (Lilienthal 1896). Clement Ader constructed his first of three flying machines in 1886, the Éole. It was a batlike design run by a lightweight steam engine driving a four-blade propeller. On 9 October 1890, Ader attempted to fly the Éole. Aviation historians credit this effort as a powered take-off and uncontrolled hop of approximately 50 m (160 ft) at a height of roughly 20 cm (Gibbs-Smith 1959). The Wright brothers invented and flew the first airplane in 1903, recognized as “the first sustained and controlled heavier-than-air powered flight” (Repperger 2003). They built on the works of George Cayley. Following its limited use in World War I, airplanes were present in all the major battles of World War II. The first jet aircraft was the German Heinkel He 178 in 1939. The first jet airliner, the de Havilland Comet, was introduced in 1952. The Boeing 707, the first widely successful commercial jet, was in commercial service from 1958 for over 50 years. Gas turbines, developed

106

Chapter 2

in the 1930s and used to power airplanes, made fast, inexpensive, and massscale intercontinental travel possible (Smil 2007).

3.4. The Industrialization of Agriculture The application of machines in agriculture significantly increased labor productivity. In the mid-nineteenth century, horse-drawn machinery revolutionized harvesting, such as the McCormick Reaper (Ankli 1976). Farmers began to use steam-powered threshers and tractors, but those machines were expensive, dangerous, and a fire hazard. The first gasolinepowered tractors were successfully developed around 1900, and mechanized portable power machines had replaced draft animals (particularly horses) by the early 1920s (Dieffenbach and Gray 1960). Selfpropelled mechanical harvesters (combines), planters, transplanters, and other equipment have been developed, notably by Fordson, John Deere, and the International Harvester Farmall company, further revolutionizing agriculture. Farming tasks could now be done quickly and on a previously impossible scale, and modern farms generate much greater volumes of highquality produce per land unit (Reid 2011). Farmers and farm workers nowadays are only a tiny fraction of employment in developed countries.

3.5. Innovations in Management Complex and big firms emerged during the Second Industrial Revolution. Railroads employed vast amounts of capital and ran a more complicated business than before. They needed better ways to track costs to calculate the rates they needed to charge, and they also needed to keep track of locomotives and cars. Telegraph lines were built along the railroads to keep track of trains when telegraph became available. This need to track costs and resources led to “railroad accounting,” later adopted by other industries and eventually became modern accounting. A collision on the Western Railroad in the US in 1841 led to a call for safety reform and, consequently, the reorganization of railroads into different departments with clear lines of management authority. Frederick Winslow Taylor and others in America developed the concept of scientific management or Taylorism in the 1880s and 1890s (Blake and Moseley 2011). Taylor’s study on scientific management initially concentrated on reducing the steps taken in performing work such as bricklaying or shoveling. The concepts later evolved into fields such as industrial engineering and business management that helped to completely restructure the operations of factories or even

The Machine Age

107

entire segments of the economy. Management became not only a profession but also an academic discipline.

4. The Digital Revolution and the Third or Fourth Industrial Revolution After World War II, the impact of industrialization spread to all walks of life, especially with the invention of semiconductor electronics. Home appliances significantly reduced labor intensity at home, and women began to have more time for paid work. Automation in the economy raised labor productivity in manufacturing and telecommunications. By the 1970s, the manufacturing sector in developed countries had largely met the demand for physical goods. The period from the end of World War II to the early 1970s was a golden time for developed countries and many emerging economies. A new revolution, the digital revolution, was brewing during this period. The digital revolution was the change from mechanical and analog electronic technology to digital technology characterized by the adoption and proliferation of digital computers and digital record-keeping. It began between the 1950s and 1970s and continued to the early twenty-first century. For many commentators, analogous to the Agricultural Revolution and Industrial Revolution, the digital revolution marked the beginning of an Information Age (Sterling 1997), also called the information revolution. The mass production and widespread use of digital logic circuits and their derived technologies, including the computer, digital cellular phone, and the internet, are central to this digital revolution. It has also been called the Third Industrial Revolution in recent years. Around 1970, many researchers noted that the service sector was becoming the largest sector, and they predicted a new era in human history (Bell 1973). According to Toffler (1981), the Neolithic Agricultural Revolution was the first wave; the Industrial Revolution was the second wave, and the rise of the service sector, including the application of computers and databases, was the third wave. The current usage of the term the Third Industrial Revolution means information revolution. Jeremy Rifkin published the book The Third Industrial Revolution in 2011 and argued that fundamental economic change would occur when new communication technologies converged with new energy regimes (Rifkin 2016). In his conceptualization, the Third Industrial Revolution is not the digital revolution. Instead, it is the imminent convergence of information and communication technologies with distributed new energy sources after the digital revolution. Rifkin

108

Chapter 2

appears to be the first to use the term Third Industrial Revolution, though in a different sense from the more recent usage. If the Information Age started in the 1970s, it coincided with the stagflation in the 1970s, especially after the first oil shortages following the Fourth Arab-Israeli War in 1973. Stagflation, i.e., slow growth with high inflation, challenged previous views on inflation, which thought high inflation rates would not coexist with high unemployment rates and slow growth. In contrast with the previous demand-pulled inflation, economists proposed cost-pushed inflation to explain the phenomenon of stagflation. With the wide application of computers and information technology (IT)in the 1980s and 1990s, it had been expected that productivity growth would accelerate. However, economic studies failed to find an increase in productivity growth, dubbed the “productivity puzzle” of IT applications (David 1990). The impacts of the digital revolution are most noticeable in communications, information acquisition, entertainment, and other intangible processes such as money transfer, ordering goods and services, and payments. In recent years, the progress in artificial intelligence (AI), robotics, and IT has led to the concept of the Fourth Industrial Revolution. Unlike the previous conceptualization of industrial revolutions, which economic and technological historians proposed after the completion or at least in the middle of the revolution, the Fourth Industrial Revolution is conceptualized at a time point where we are not sure whether we are in the middle of, or we are about to enter the process. The Fourth Industrial Revolution has been described as applying new technologies fusing the physical, digital, and biological worlds and impacting all disciplines, economies, and industries. In the following chapters, it will be demonstrated that human society is entering the Robotic Age, in which robots with advanced AI liberate humans from production and service jobs.

5. Summary The First Industrial Revolution arose from the improvement of manual tools for textile production and resulted in the widespread use of steam engines to drive machines that replaced human physical power. The Middle Ages’ Commercial Revolution and the incremental technological progress prepared the foundation of the Industrial Revolution. The improved transport and communications by the eve of the Industrial Revolution facilitated the diffusion of technology so that innovation could benefit businesses worldwide in a few years.

The Machine Age

109

The Second Industrial Revolution, built upon the technological progress of the First Industrial Revolution, continued to expand industrial production regarding the ranges and varieties of raw materials and finished products and management practices, which provided the foundation for modern human society. The appearance of new communication and transport tools made the diffusion of new technology much faster, and existing technology could also facilitate new inventions and innovations. The much-touted Third and Fourth Industrial Revolutions seem to be transitional periods before human society enters the Robotic Age.

References Ackroyd, JAD. 2011. “Sir George Cayley: The invention of the aeroplane near Scarborough at the time of Trafalgar.” Journal of Aeronautical History (6):130-181. Aftalion, Fred. 1991. A history of the international chemical industry: University of Pennsylvania Press. Ahmadinia, Nahid. 2014. “The linear induction motor (LIM) & single linear induction motor (SLIM).” American Journal of Electrical Power and Energy Systems 3 (4):71-75. Al-Khalili, Jim. 2015. “The birth of the electric machines: a commentary on Faraday (1832)‘Experimental researches in electricity’.” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 373 (2039):20140208. Alford, Steven E, and Suzanne Ferriss. 2007. Motorcycle. London: Reaktion Books. Allen, Robert C. 1983. “Collective invention.” Journal of Economic Behavior & Organization 4 (1):1-24. Andrady, Anthony L, and Mike A Neal. 2009. “Applications and societal benefits of plastics.” Philosophical Transactions of the Royal Society B: Biological Sciences 364 (1526):1977-1984. Ankli, Robert E. 1976. “The coming of the reaper.” Business and Economic History:1-24. Ashmore, Sonia. 2012. Muslin. London: V & A Publishing. Ashton, Thomas Southcliffe. 1997. The industrial revolution 1760-1830, Oxford: Oxford University Press. Bagnall, William R. 1893. The textile industries of the United States: including sketches and notices of cotton, woolen, silk, and linen manufacturers in the colonial period. Vol. 1. Cambridge, MA: Riverside Press.

110

Chapter 2

Baines, Edward. 2015. History of the cotton manufacture in Great Britain. Cambridge: Cambridge University Press. Bates, Edward C. 1899. The Story of the Cotton Gin. Westborough, MA: The Westborough Historical Society. Beaton, Kendall. 1955. “Dr. Gesner's kerosene: the start of American oil refining.” Business History Review 29 (1):28-53. Béchamp, Antoine. 1854. “De l'action des protosels de fer sur la nitronaphtaline et la nitrobenzine;: nouvelle méthode de formation des bases orgnaiques artificielles de zinin.” Annales de Chimie et de Physique 42:186-196. Bell, Daniel. 1973. The coming of post-industrial society: A venture in social forecasting New York: Basic Books. Benn, Jeremy. 2021. “Lord Armstrong and the lakes of Cragside, Northumberland.” Dams and Reservoirs 31 (1):2-9. Bishop, Philip W. 1959. “The Beginnings of Cheap Steel.” Bulletin of the United States National Museum 218:27-47. Black, Brian, and Marcy Ladson. 2010. “Oil at 150: Energy past and future in Pennsylvania.” Pennsylvania Legacies 10 (1):6-13. Blake, Anne M, and James L Moseley. 2011. “Frederick Winslow Taylor: One hundred years of managerial insight.” International Journal of Management 28 (4):346. Bogart, Dan. 2005. “Turnpike trusts and the transportation revolution in 18th century England.” Explorations in Economic History 42 (4):479508. Bogart, Dan. 2013. “The Transportation Revolution in Industrializing Britain: A Survey.” University of California-Irvine, Department of Economics Working Papers (121306). Bojanova, Irena. 2014. “The Digital Revolution: What's on the Horizon?” IT Professional 16 (1):8-12. Boretti, Alberto. 2011. “Advantages of the direct injection of both diesel and hydrogen in dual fuel H2ICE.” International Journal of Hydrogen Energy 36 (15):9312-9317. Breslow, Boyd. 2004. “A London Merchant and the Governance of Medieval England.” Medieval Prosopography 25:101-113. Brittain, James E. 2004. “Electrical engineering hall of fame: Guglielmo Marconi.” Proceedings of the IEEE 92 (9):1501-1504. Broad, William J. 1979. “Rival Centennial Casts New Light on Edison: It's the 100th birthday of the light bulb, and some Britons say that Edison didn't invent it.” Science 204 (4388):32-36.

The Machine Age

111

Broeze, Frank. 1989. “Distance tamed: steam navigation to Australia and New Zealand from its beginnings to the outbreak of the Great War.” The Journal of Transport History 10 (1):1-21. Brown, Richard. 2002. Society and economy in modern Britain 1700-1850. London and New York: Routledge. Buchanan, R Angus. 2006. Brunel: the life and times of Isambard Kingdom Brunel. London and New York: Hambledon Continuum. Buer, Mabel Craven. 2018. Revival: Health, wealth, and population in the early days of the industrial revolution (1926). London and New York: Routledge. Byrne, Richard. 2013. “A Nod to Ned Ludd.” The Baffler (23):120-128. Cadman, WK. 1959. “Kier's 5-Barrel Still: A Venerable Industrial Relic.” Western Pennsylvania History 42 (4):351-362. Carosso, Vincent P. 1949. “The Waltham Watch Company: A Case History.” Bulletin of the Business Historical Society 23 (4):165-187. Catling, Harold. 1978. “The development of the spinning mule.” Textile History 9 (1):35-57. Chadwick, Edwin. 1842. Report to Her Majesty's principal secretary of state for the Home Department, from the Poor Law Commissioners on an inquiry into the sanitary condition of the labouring population of Great Britain: with appendices. London: W. Clowes. Chirnside, RC. 1979. “Sir Joseph Swan and the invention of the electric lamp.” Electronics and Power 25 (2):96. Chrystal, Paul. 2018. Bradford at work: people and industries through the years. Stroud, Gloucestershire: Amberley Publishing. Clark, Gregory. 2005. The British Industrial Revolution, 1760-1860. Davis, CA: University of California Davis. Clingingsmith, David, and Jeffrey G Williamson. 2008. “Deindustrialization in 18th and 19th century India: Mughal decline, climate shocks and British industrial ascent.” Explorations in Economic History 45 (3):209234. Clow, Archibald, and Nan L Clow. 1942. “Lord Dundonald.” The Economic History Review 12 (1/2):47-58. Clow, Archibald, and Nan L Clow. 1945. “Vitriol in the industrial revolution.” The Economic History Review 15 (1/2):44-55. Cottrell, PL, and G Ottley. 1975. “The Beginnings of the Stockton & Darlington Railway.” The Journal of Transport History ss-3 (2):86-93. Crafts, Nicholas FR. 1996. “The first industrial revolution: A guided tour for growth economists.” The American economic review 86 (2):197-201. Crain, ER, and P Eng. 2004. “A true history of oil and gas development.” Canadian Well Logging Society 1 (13):9-12.

112

Chapter 2

Crespy, Daniel, Marianne Bozonnet, and Martin Meier. 2008. “100 Years of Bakelite, the Material of 1000 Uses.” Angewandte Chemie International Edition 47 (18):3322-3328. Crompton, GW. 1993. “Canals and the industrial revolution.” The Journal of Transport History 14 (2):93-110. Cummins Jr, C Lyle. 1976. “Early IC and automotive engines.” SAE Transactions 85 (3):1960-1971. Cypher, James, and James M Cypher. 2008. The process of economic development. London and New York: Routledge. David, Paul A. 1990. “The dynamo and the computer: an historical perspective on the modern productivity paradox.” The American economic review 80 (2):355-361. Deane, Phyllis M. 1979. The first industrial revolution. Cambridge: Cambridge University Press. Delogne, Paul. 1998. “Lee de Forest, the inventor of electronics: A tribute to be paid.” Proceedings of the IEEE 86 (9):1878-1880. Desaunais, A. 1934. “L’exploitation fluviale du bassin de Saint-Etienne.” Géocarrefour 10 (1):5-45. Dibner, Bern. 1954. “Ten founding fathers of the electrical science: II. Otto von Guericke: And the first electric machine.” Electrical Engineering 73 (5):396-397. Dickey, Parke A. 1959. “The first oil well.” Journal of Petroleum Technology 11 (1):14-26. Dickinson, Henry Winram. 1941. “Joseph Bramah and his Inventions.” Transactions of the Newcomen Society 22 (1):169-186. Dickinson, Henry Winram. 2011. A short history of the steam engine. Cambridge: Cambridge University Press. Dickinson, Henry Winram, and Arthur Titley. 2010. Richard Trevithick: the engineer and the man. Cambridge: Cambridge University Press. Dieffenbach, EM, and RB Gray. 1960. “The development of the tractor.” Agricultural Yearbook 1960:24-45. DiLaura, David. 2008. “A brief history of lighting.” Optics and Photonics News 19 (9):22-28. Donnelly, Frederick Kerr. 1976. “The Scottish Rising of 1820: A Reinterpretation.” International Review of Scottish Studies 6:27-37. Dumpleton, Bernard. 2002. The Story of the Paddle Steamer. Bristol: Intellect Books. Dylla, HF, and Steven T Corneliussen. 2005. “John Ambrose Fleming and the beginning of electronics.” Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films 23 (4):1244-1251.

The Machine Age

113

Eacott, Jonathan P. 2012. “Making an imperial compromise: The Calico Acts, the Atlantic Colonies, and the structure of the British Empire.” The William and Mary Quarterly 69 (4):731-762. Ellis, Christopher J, and John Fender. 2009. “The economic evolution of democracy.” Economics of Governance 10 (2):119-145. Espion, Bernard, Robin Engels, and Michel Provost. 2017. “Civil engineering heritage: country profile–Belgium.” Proceedings of the Institution of Civil Engineers-Engineering History and Heritage 170 (4):197-205. Evans, Chris, and Alun Withey. 2012. “An enlightenment in steel? Innovation in the steel trades of eighteenth-century Britain.” Technology and Culture 53 (3):533-560. Evans, Francis T. 1981. “Roads, railways, and canals: technical choices in 19th-century Britain.” Technology and Culture 22 (1):1-34. Eyde, Herr Sam. 1909. “The manufacture of nitrates from the atmosphere by the electric arc—Birkeland-Eyde process.” Journal of the Royal Society of Arts 57 (2949):568-576. Falco, Charles M. 2003. “The art and materials science of 190-mph superbikes.” MRS Bulletin 28 (7):512-516. Fay, CR. 1937. “John Loudon McAdam: Chapters in the History of Highways.” The Economic Journal 47 (Supplement_1):465-467. Feinstein, Charles H. 1998. “Pessimism perpetuated: real wages and the standard of living in Britain during and after the industrial revolution.” Journal of Economic History 58 (3):625-658. Feldenkirchen, Wilfried. 1994. Werner von Siemens: inventor and international entrepreneur. Columbus, Ohio: Ohio State University Press. Fernández, Lucía. 2021. “Global synthetic fiber production 1940-2019.” Statista, Last Modified Jul 19, 2021, accessed 25 February. https://www.statista.com/statistics/741368/world-synthetic-fiberproduction/. Fleming, JA. 1925. “William Sturgeon and the Centenary of the Electromagnet.” Journal of the Royal Society of Arts 73 (3787):689-710. Flint, Henry Martyn. 1868. The Railroads of the United States: Their History and Statistics Comprising the Progress and Present Condition of the Various Lines with Their Earnings and Expenses. To which are Added a Synopsis of the Railroad Laws of the United States, and an Article on the Comparative Merits of Iron and Steel Rails. Philadelphia: John E. Potter. Ford, Henry, and Samuel Crowther. 1922. My life and work. New York: Doubleday, Page & Co.

114

Chapter 2

Ford, Trevor D, and Bernard O’Connor. 2009. “A vanished industry: Coprolite mining.” Mercian Geologist 17 (2):93-100. Forward, EA. 1924. “The Early History of the Cylinder Boring Machine.” Transactions of the Newcomen Society 5 (1):24-38. Francis, Anthony John. 1977. The cement industry 1796-1914: A history. North Pomfret, Vermont: David & Charles. Frumkin, Maximilian. 1945. “The origin of patents.” Journal of Patent Office Society 27 (3):143-148. Galbi, Douglas A. 1997. “Child labor and the division of labor in the early English cotton mills.” Journal of Population Economics 10 (4):357-375. Gale, WKV. 1973. “The Bessemer steelmaking process.” Transactions of the Newcomen Society 46 (1):17-26. Gibb, Claude D. 1947. “Parsons—the man and his work.” Proceedings of the Institution of Mechanical Engineers 156 (1):213-218. Gibbs-Smith, Charles H. 1959. “Hops and Flights: A roll call of early powered take-offs.” Flight 75 (2619):469. Gilbert, KR. 1971. “Henry Maudslay 1771–1831: PRESIDENTIAL ADDRESS.” Transactions of the Newcomen Society 44 (1):49-62. Gillispie, Charles C. 1957. “The discovery of the Leblanc process.” Isis 48 (2):152-170. Gittins, Lawrence. 1979. “Innovations in textile bleaching in Britain in the eighteenth century.” The Business History Review:194-204. Gordon, Robert Boyd. 2001. American Iron, 1607-1900. Vol. 19. Baltimore, Maryland: John Hopkins University Press. Gorman, Michael E. 1995. “Confirmation, disconfirmation, and invention: The case of Alexander Graham Bell and the telephone.” Thinking & Reasoning 1 (1):31-53. Griffin, Emma. 2018. A short history of the British industrial revolution. London: Palgrave. Griffiths, Clare. 1997. “Remembering Tolpuddle: rural history and Commemoration in the inter-war Labour Movement.” History Workshop Journal (44):144-169. Grotheer, Morris, Richard Alkire, Richard Varjian, Venkat Srinivasan, and John Weidner. 2006. “Industrial electrolysis and electrochemical engineering.” Interface 15 (1):52-54. Guarnieri, Massimo. 2011. “When cars went electric, part one [historical].” IEEE Industrial Electronics Magazine 5 (1):61-62. Guarnieri, Massimo. 2013a. “The beginning of electric energy transmission: Part one [Historical].” IEEE Industrial Electronics Magazine 7 (1):5052.

The Machine Age

115

Guarnieri, Massimo. 2013b. “The beginning of electric energy transmission: Part two [historical].” IEEE Industrial Electronics Magazine 7 (2):5259. Guarnieri, Massimo. 2018. “Revolving and evolving-early dc machines [Historical].” IEEE Industrial Electronics Magazine 12 (3):38-43. Guarnieri, Massimo. 2019. “Messaging Before the Internet-Early Electrical Telegraphs [Historical].” IEEE Industrial Electronics Magazine 13 (1):38-53. Guarnieri, Massimo. 2020. “Electric Tramways of the 19th Century [Historical].” IEEE Industrial Electronics Magazine 14 (1):71-77. Guise-Richardson, Cai. 2010. “Redefining Vulcanization: Charles Goodyear, patents, and industrial control, 1834-1865.” Technology and Culture 51 (2):357-387. Habashi, Fathi. 2013. “Henry Bessemer 200 years since his birth.” Steel Times International 37 (1):36. Halacsy, Andrew A. 1971. “Anyos Jedlik, an inventor of the dynamoelectric principle.” Electronics and Power 17 (9):332-335. Harris, Frank R. 1984. “The Parsons centenary—a hundred years of steam turbines.” Proceedings of the Institution of Mechanical Engineers, Part A: Power and Process Engineering 198 (3):183-224. Harris, J. B. 1993. “Electric lamps, past and present.” Engineering Science & Education Journal 2 (4):161-170. Hassan, John A. 1978. “Relationships between Coal, Gas, and Oil Production: A Nineteenth-Century Scottish Case-Study.” Industrial Archaeology Review 2 (3):277-289. Headrick, Daniel R, and Pascal Griset. 2001.” Submarine telegraph cables: Business and politics, 1838–1939.” Business History Review 75 (3):543578. Heinze, G Wolfgang, and Heinrich H Kill. 2019. The development of the German railroad system. London and New York: Routledge. Hodgson, John Edmung. 1927. “James Sadler, of Oxford Aeronaut, Chemist, Engineer, and Inventor.” Transactions of the Newcomen Society 8 (1):66-82. Hounshell, David. 1985. From the American system to mass production, 1800-1932: The development of manufacturing technology in the United States. Baltimore, Maryland: John Hopkins University Press. Hughes, David, Simon Swann, and Alan Gardner. 2007. “Roman cement: part one: its origins and properties.” Journal of Architectural Conservation 13 (1):21-36. Hunt, LB. 1973. “The early history of gold plating.” Gold Bulletin 6 (1):1627.

116

Chapter 2

Husson, G. 1963. “Cent années d'aciérie Martin (1863-1963).” Revue de Métallurgie 60 (7-8):611-622. Huurdeman, Anton A. 2003. The worldwide history of telecommunications. Hoboken, New Jersey: John Wiley & Sons. Irlam, Geoffrey A. 1989. “Electricity Supply at Cragside.” Industrial Archaeology Review 11 (2):187-195. Ivell, David M. 2012. “Phosphate Fertilizer Production–From the 1830s to 2011 and Beyond.” Procedia Engineering 46:166-171. Jackson, Melvin H, and Carel De Beer. 1974. Eighteenth Century Gunfounding: The Verbruggens at the Royal Brass Foundry; a Chapter in the History of Technology. Washington, D.C.: Smithsonian Institution Press. Jarvis, C Mackechnie. 1969. “Nikola Tesla and the induction motor.” Electronics and Power 15 (12):436-440. Jaura, Arun. 2013. “Untiring Tyres.” Auto Tech Review 2 (10):64-64. Jenkins, Mick. 1980. The general strike of 1842. London: Lawrence and Wishart. Jenkins, Rhys. 1932. “A Cornish Engineer: Arthur Woolf, 1766-1837.” Transactions of the Newcomen Society 13 (1):55-73. Jevons, H Stanley. 1931. “The second industrial revolution.” The Economic Journal 41 (161):1-18. Jones, Peter. 2009. “Finding captain swing: Protest, parish relations, and the state of the public mind in 1830.” International Review of Social History 54 (3):429-458. Kazanskaya, Liliya, and Valentin Sobor. 2019. “Historical Railway Bridges of Russia as Objects of Architectural Heritage.” Urbanism Architecture Constructions/Urbanism Arhitectura Constructii 10 (2):123-132. Kiefer, David M. 2002. “Soda Ash, Solvay Style Introduced in the 1860s, the Solvay process for making soda ash made Leblanc's technology obsolete.” Todays Chemist at Work 11 (2):87-91. King, Peter. 2005. “The production and consumption of bar iron in early modern England and Wales.” The Economic History Review 58 (1):133. King, Peter. 2011. “The choice of fuel in the eighteenthဨcentury iron industry: the Coalbrookdale accounts reconsidered.” The Economic History Review 64 (1):132-156. Kitsikopoulos, Harry. 2013. “From Hero to Newcomen: The Critical Scientific and Technological Developments that Led to the Invention of the Steam Engine.” Proceedings of the American Philosophical Society 157 (3):304-344.

The Machine Age

117

Kreimeyer, Andreas, Peter Eckes, Christian Fischer, Harald Lauke, and Peter Schuhmacher. 2015. ““We Create Chemistry for a Sustainable Future”: Chemie schafft nachhaltige Lösungen für eine wachsende Weltbevölkerung.” Angewandte Chemie 127 (11):3220-3237. Lamoreaux, Naomi R, Margaret Levenstein, and Kenneth L Sokoloff. 2006. “Mobilizing venture capital during the second industrial revolution: Cleveland, Ohio, 1870-1920.” Capitalism and Society 1 (3). Landes, David S. 1969. The unbound Prometheus: Technological change and development in Western Europe from 1750 to the present. Cambridge: Cambridge University Press. Lewis, Michael Jonathan Taunton. 1970. Early Wooden Railways. London: Routledge and K. Paul. Lilienthal, Otto. 1896. “Practical experiments for the development of human flight.” The Aeronautical Annual:7-20. Limebeer, David JN, and Robin S Sharp. 2006. “Bicycles, motorcycles, and models.” IEEE Control Systems Magazine 26 (5):34-61. Lloyd, Terrence Henry. 2005. The English wool trade in the Middle Ages. Cambridge: Cambridge University Press. Loasby, G. 1951. “The Development of the Synthetic Fibres.” Journal of the Textile Institute Proceedings 42 (8): P411-P441. Long, Pamela O. 1991. “Invention, authorship, “intellectual property,” and the origin of patents: Notes toward a conceptual history.” Technology and Culture 32 (4):846-884. Lubar, Steven. 1984. “Managerial Structure and Technological Style: The Lowell Mills, 1821—1880.” Business and Economic History 13:20-30. MacLeod, Christine. 2002. Inventing the industrial revolution: The English patent system, 1660-1800. Cambridge: Cambridge University Press. Maddison, Angus. 1995. Monitoring the World Economy 1820–1992 Paris: OECD. Malizia, Fabio, and Bert Blocken. 2020. “Bicycle aerodynamics: history, state-of-the-art and future perspectives.” Journal of Wind Engineering and Industrial Aerodynamics 200:104134. Marsden, Richard. 1895. Cotton Weaving: Its Development, Principles, and Practice. Vol. 11. London: G. Bell & Sons. Marvel, Howard P. 1977. “Factory Regulation: A Reinterpretation of Early English Experience.” The Journal of Law and Economics 20 (2):379402. McCarter, William Matthew. 2011. “Where The Tin Lizzy Took Us.” Fast Capitalism 8 (2):99-109. McKenna, Joseph. 2019. Women in the struggle for Irish independence. Jefferson, North Carolina: McFarland.

118

Chapter 2

McKenzie, James. 2018. “Lighting the way.” Physics World 31 (3):21. Merdinger, Charles J. 1952. “Roads—through the Ages: I. Early Developments.” The Military Engineer 44 (300):268-273. Michalowicz, Joseph C. 1948. “Origin of the electric motor.” Electrical Engineering 67 (11):1035-1040. Miller, Naomi Churgin. 1968. “John Cartwright and radical parliamentary reform, 1808-1819.” The English Historical Review 83 (329):705-728. Modak, Jayant M. 2002. “Haber process for ammonia synthesis.” Resonance 7 (9):69-77. Mokyr, Joel. 1998. “The second industrial revolution, 1870-1914.” In Storia dell’economia Mondiale, edited by Valerio Castronono, 219-245. Rome Laterza. Mokyr, Joel. 2000. “The Industrial Revolution and the Netherlands: Why did it not happen?” De Economist 148 (4):503-520. Morgan, Paul W. 1981. “Brief history of fibers from synthetic polymers.” Journal of Macromolecular Science—Chemistry 15 (6):1113-1131. Morley, Neville. 2000. “Trajan's engines.” Greece & Rome 47 (2):197-210. Munro, John H. 2003. “Medieval Woollens: Textiles, Textile Technology, and Industrial Organisation, c. 800-1500.” In The Cambridge History of Western Textiles, edited by David Jenkins, 181-227. Cambridge: Cambridge University Press. Nardinelli, Clark. 1980. “Child labor and the factory acts.” The Journal of Economic History 40 (4):739-755. Nevell, Michael. 2005. “The social archaeology of industrialization: the example of Manchester during the 17th and 18th Centuries.” In Industrial Archaeology: Future Directions, edited by Eleanor Conlin Casella and James Symonds, 177-204. Boston, MA: Springer. Nevell, Michael. 2013. “Bridgewater: the archaeology of the first arterial industrial canal.” Industrial Archaeology Review 35 (1):1-21. Nübel, Otto. 1987. “The beginnings of the automobile in Germany.” In The Economic and Social Effects of the Spread of Motor Vehicles, 55-66. London: Palgrave Macmillan. Pabis-Braunstein, Mirosáawa. 1989. “Ignacy àukasiewicz (1822-1882).” Pharmacy in History 31 (4):174-176. Patil, Vinayak L. 2015. “Historical perspectives of development of antique analog telephone systems.” Journal of Telecommunications and Information Technology (3):70--98. Payen, Jacques. 1963. “Les brevets de Lenoir concernant le moteur à combustion interne.” Revue d'Histoire des Sciences et de Leurs Applications 16 (4):374-380.

The Machine Age

119

Phillips, John A, and Charles Wetherell. 1995. “The Great Reform Act of 1832 and the political modernization of England.” The American Historical Review 100 (2):411-436. Prakash, Om. 2014. The Dutch East India Company and the Economy of Bengal, 1630-1720. Princeton: Princeton University Press. Pumfrey, Stephen, and David Tilley. 2003. “William Gilbert: forgotten genius.” Physics World 16 (11):15-16. Rasmussen, Seth C. 2021. “From Parkesine to celluloid: The birth of organic plastics.” Angewandte Chemie 133 (15):8090-8094. Rautio, James C. 2014. “The long road to Maxwell's equations.” IEEE Spectrum 51 (12):36-56. Rees, Paul. 1978. “Chatsworth Street Cutting, Part of the Original Terminus of the Liverpool and Manchester Railway.” Industrial Archaeology Review 2 (3):38-51. Reid, John F. 2011. “The impact of mechanization on agriculture.” Bridge 41 (3):22-29. Repperger, Daniel W. 2003. “Celebrating the 100th anniversary of controlled, sustained, and powered air flight.” IEEE Control Systems Magazine 23 (6):12-16. Richards, John F. 1993. The Mughal Empire, The new Cambridge history of India. Cambridge: Cambridge University Press. Rifkin, Jeremy. 2016. “How the third industrial revolution will create a green economy.” New Perspectives Quarterly 33 (1):6-10. Robertson, J. 1971. “Robert Owen, Cotton Spinner: New Lanark, 1800– 1825.” In Robert Owen: Prophet of the Poor, edited by Sidney Pollard and John Salt, 145-165. Lewisburg, Pennsylvania: Bucknell University Press. Robinson, Sir Robert. 1957. “Sir William Henry Perkin: Pioneer of Chemical Industry.” Journal of Chemical Education 34 (2):54. Roe, Joseph Wickham. 1916. English and American tool builders. New Haven: Yale University Press. Rolt Lionel, TC, and JS Allen. 1997. The steam engine of Thomas Newcomen. Ashbourne, England: Landmark Publishing Rolt, Lionel Thomas Caswell. 1970. Victorian engineering. London: Pelican. Rosenberg, Nathan, and Rosenberg Nathan. 1982. Inside the black box: technology and economics. Cambridge: Cambridge University Press. Ryan, John F. 1929. “The story of Portland cement.” Journal of Chemical Education 6 (11):1854-1868. Sandler, Stanley. 1970. “The Emergence of the Modern Capital Ship.” Technology and Culture 11 (4):576-595.

120

Chapter 2

Scaife, W Garrett. 1985. “The parsons steam turbine.” Scientific American 252 (4):132-139. Schadewaldt, H. 1975. “Die Einführung des Atebrin in den Arzneischatz.” DMW-Deutsche Medizinische Wochenschrift 100 (50):2583-2585. Scharrer, Karl. 1949. “Justus von Liebig and today's agricultural chemistry.” Journal of Chemical Education 26 (10):515. Schmidt, Helmut. 1997. “Indigo–100 Jahre industrielle synthese.” Chemie in Unserer Zeit 31 (3):121-128. Seaton, Albert E. 2013. The Screw Propeller. Bremem: Maritime Press. Sellers Jr, Coleman. 1886. “Oliver Evans and his inventions.” Journal of the Franklin Institute 122 (1):1-16. Shearer, GR. 1977. “The Rolling Wheel—The Development of the Pneumatic Tyre.” Proceedings of the Institution of Mechanical Engineers 191 (1):75-87. Sheeran, George. 2005. The buildings of Bradford: an illustrated architectural history. Stroud, Gloucestershire: Tempus. Siemens, C William. 1862. “On a regenerative gas furnace, as applied to glasshouses, puddling, heating, etc.” Proceedings of the Institution of Mechanical Engineers 13 (1):21-45. Sievers, Immo. 2005. “100 Jahre NSU-Automobile.” ATZAutomobiltechnische Zeitschrift 107 (9):772-776. Smil, Vaclav. 2007. “The two prime movers of globalization: history and impact of diesel engines and gas turbines.” Journal of Global History 2 (3):373-394. Smiles, Samuel. 1884. Industrial biography: ironworkers and tool makers. London: John Murray. Smith, Edgar C. 1938. A short history of naval and marine engineering. Cambridge: Cambridge University Press. Sterling, Bruce. 1997. “The digital revolution in retrospect.” Communications of the ACM 40 (2):79-ff. Susskind, Charles. 1964. “Observations of electromagnetic-wave radiation before Hertz.” Isis 55 (1):32-42. Sutcliffe, Thomas. 1843. An exposition of the facts relating to the rise and progress of the woollen, linen and cotton manufacturers of Great Britain. Manchester: P. Grant. Sutton, Mark A, Albert Bleeker, CM Howard, JW Erisman, YP Abrol, M Bekunda, A Datta, E Davidson, W De Vries, and O Oenema. 2013. Our nutrient world. The challenge to produce more food & energy with less pollution. Centre for Ecology & Hydrology. Swallow, Peter, and David Carrington. 1995. “Limes and lime mortars— part one.” Journal of Architectural Conservation 1 (3):7-25.

The Machine Age

121

Swank, James M. 1904. “The Manufacture of Iron and Steel Rails in Western Pennsylvania.” The Pennsylvania Magazine of History and Biography 28 (1):1-11. Szreter, Simon, and Graham Mooney. 1998. “Urbanization, mortality, and the standard of living debate: new estimates of the expectation of life at birth in nineteenth-century British cities.” Economic History Review 51 (1):84-112. Tann, Jennifer. 1979. “Arkwright's Employment of Steam Power: A Note of Some New Evidence.” Business History 21 (2):247-250. Taylor, George R. 2015. The transportation revolution, 1815-60. London and New York: Routledge. Temple, Robert KG, Joseph Needham, and Religionsphilosoph Biochemiker. 2007. The genius of China: 3,000 years of science, discovery, and invention. Rochester, VT: Inner Traditions Thrupp, Sylvia L. 1989. The merchant class of medieval London, 1300-1500. Ann Arbor: University of Michigan Press. Toffler, Alvin. 1981. The third wave. New York: Bantam. Totten, George E. 2004. “A timeline of highlights from the histories of ASTM Committee D02 and the petroleum industry.” ASTM International, accessed July 6, 2021. https://www.astm.org/COMMIT/D02/print_version.html. Travis, Anthony S. 1991. “Heinrich Caro at Roberts, Dale & CO.” Ambix 38 (3):113-134. Tyerman, Christopher. 2019. The World of the Crusades. New Haven, Connecticut: Yale University Press. Vance Jr, James E. 1966. “Housing the worker: the employment linkage as a force in urban structure.” Economic Geography 42 (4):294-325. Vogler, Herbert. 2016. “Besser aromatisch als aliphatisch: 75 Jahre Polyesterfasern.” Chemie in Unserer Zeit 50 (4):275-277. Volta, Alessandro. 1800. “XVII. On the electricity excited by the mere contact of conducting substances of different kinds. In a letter from Mr. Alexander Volta, F. R. S. Professor of Natural Philosophy in the University of Pavia, to the Rt. Hon. Sir Joseph Banks, Bart. K.B. P. R. S.” Philosophical Transactions of the Royal Society of London 90:403431. von Viebahn, Georg. 1852. Der englische Gewerbfleiss, seine geschichtliche Entwickelung, gegenwärtigen Hauptsitze und volkswirtschaftlichen Ergebnisse. Braunschweig: Friedrich Vieweg. Voth, Hans-Joachim. 2003. “Living standards during the industrial revolution: An economist's guide.” American Economic Review 93 (2):221-226.

122

Chapter 2

Warington, R. 1900. “Sir John Bennet Lawes, Bart., FRS.” Nature 62 (1611):467-468. Warren, Kenneth. 1998. Steel, ships and men: Cammell Laird, 1824-1993. Oxford: Oxford University Press. Williams, Gwyndaf. 1996. “Manchester.” Cities 13 (3):203-212. Williams, J. B. 2018. “First You Have to Make It: The Spread of the Electricity Supply.” In The Electric Century, 31-43. Berlin: Springer. Williams, Kelvin. 1987. “Sebastian Ziani de Ferranti 1864–1930. Pioneer of electrical power systems.” Electronics & Power 33 (8):493-495. Wilson, J. F. 1990. “R. S. Fitton, The Arkwrights, Spinners of Fortune.” The Economic History Review 43 (1):143-145. Wilson, Robert E. 1928. “Fifteen Years of the Burton Process.” Industrial & Engineering Chemistry 20 (10):1099-1101. Wood, Henry Trueman, Robert Kay, John Kay, Nathan Fielding, and Wm Kay. 1911. “The Inventions of John Kay (1704—1770).” Journal of the Royal Society of Arts 60 (3081):73-86. Zhang, Ce, and Jianming Yang. 2020. “Second Industrial Revolution.” In A History of Mechanical Engineering, 137-195. Springer.

CHAPTER 3 THE DAWN OF THE ROBOTIC AGE: FROM CENTRIFUGAL GOVERNORS TO THE RISE OF ARTIFICIAL INTELLIGENCE AND ROBOTS

Progress in human civilization can be viewed from two aspects: the increased consumption of material goods, services, and intangible goods and the decreased involvement of human beings in producing one unit of output. The tools in the Paleolithic, Neolithic, Bronze, and Iron Ages were primarily enhancers of human physical strength, enabling people to perform tasks they would otherwise be unable to. The Industrial Revolution brought humanity into the Machine Age, in which human strength was no longer a critical factor in production. Society is currently on the brink of the Robotic Age, in which machine intelligence will replace human intelligence in the routine production process. When how many human workers are replaced by robots and artificial intelligence (AI), can we view society as in the Robotic Age? We may define the Robotic Age as when more than half of the working-age population has lost their jobs because robots and AI systems are doing their jobs, or more precisely, more than half of all working-age households (i.e., individuals or families with working-age parents or children). This is also when universal guaranteed basic income will be in place. Many people may call the new era the “AI Age” or the “Information Age.” This book will use the term the “Robotic Age” instead of the “AI Age” or the “Information Age” because many tasks and activities in human society involve handling physical objects, and AI or information per se does not imply physical handling. Moreover, information has existed throughout the entire history of humanity and has always been an essential factor in human activities. Since ancient times, humanity has been dreaming of automatons that could work on behalf of humans to relieve human beings from hard labor (Cholodenko 2007; Berryman 2003). Various mechanical automated

Chapter 3

124

control systems were invented during the Machine Age to improve the efficiencies of machine systems and workers. Electromechanical devices and logic circuitry were later incorporated into automation systems. The advent and growing application of digital computers led to computer numerical control (CNC) machine tools, which could make better products with even less human intervention. The development of intelligent robots and software that can mimic and outperform human brains in completing complex mental tasks foretells the coming of the Robotic Age, in which human physical and mental dexterities are no longer needed in all routine human jobs. This chapter will examine the progress in the automation of physical processes, automation of cognitive processes, and the combination of the two, i.e., intelligent robots.

1. Automation of Physical Processes The automation of physical processes is what people usually consider to be automation. In this narrow sense, automation, or automatic control, uses various control systems for operating equipment to replace or reduce human intervention. The main objective of automation is to save labor involved in the physical intervention of production processes. Automation also improves quality, accuracy, precision, and workplace safety. Many physical processes were fully automated before the digital revolution and the emergence of AI as a research discipline. Digitalization makes automation more convenient and more applicable.

1.1. Automation in Practice Automation emerged almost along with machines, using basic feedback control mechanisms. The earliest feedback control mechanism was the windmill fantail used to keep the main sails in the optimal perpendicular orientation and produce maximum power. Edmund Lee patented it in 1745 (Mayr 1970). Automation could be achieved by mechanical, hydraulic, pneumatic, electrical, electronic devices, and computers. These means are often combined, especially in complicated systems, such as automation systems in modern factories, airplanes, and ships. 1.1.1. Centrifugal Governors An early application of automation was the centrifugal governor, a governor that uses centrifugal force to measure and regulate the speed of a machine. “A governor is a part of a machine by means of which the velocity of the machine is kept nearly uniform, notwithstanding variations in the driving-

The Dawn of the Robotic Age

125

power or the resistance” (Maxwell 1868). The centrifugal governor was first used in the last quarter of the eighteenth century to adjust the gap between millstones through the spindle carried by the tentering gear. Oliver Evans used it in his automatic flour mill developed in 1785 (Kilgour 1964). James Watt adopted the centrifugal governor to stabilize the speed of steam engines in 1788 (White 1988). The governor was powered by the output of the steam engine and connected to a throttle valve that would regulate the steam flow into the cylinder. When the speed of the steam engine increased, the centrifugal force would reduce the aperture of the throttle valve via a lever. Thus, an equilibrium point would be reached at which the engine settled down to a more-or-less constant speed. 1.1.2. Automatic Steering Systems With the development of the shipbuilding industry, more big ships were built, and machines that helped steer ships became necessary. John McFarlane Gray patented a steam steering engine that incorporated feedback in 1866 (Bennett 1996). His steam steering engine had the characteristics of a modern servomechanism: an input, an output, an error signal, and a means for amplifying the error signal used for negative feedback to drive the error toward zero. The angle of the rudder was transmitted to a differential screw, which controlled a steam valve that supplied power to a motor that turned the rudder. The steam valve reduced power as the rudder approached the desired angle; it increased power when the rudder moved away from that angle and returned the rudder to its desired position. Gray’s steam steering engine was first successfully tried in the SS Great Eastern, the largest and most advanced ship of the day, in 1867 (Petree 1945). A critical development in the automatic steering system was the use of gyrocompasses. A gyrocompass is a type of non-magnetic compass based on a fast-spinning disc and the rotation of the earth to find geographical direction automatically (Gade 2016). The first functional gyrocompass was patented in 1904 by German inventor Hermann Anschütz-Kaempfe who founded a firm to manufacture it in 1905. It was widely used in the German Imperial Navy after successful tests in 1908 (Broelmann 2002). The American Elmer Ambrose Sperry patented a gyrocompass of his design in 1908 and founded the Sperry Gyroscope Company. The company launched its first gyrocompass in 1911, which the US Navy adopted (Roberts 2008; Trainer 2008). His automatic ship steering system compensated for water disturbances as sea conditions changed. It was the first successful proportional-integral-derivative (PID) controller. In 1922, Nicolas

126

Chapter 3

Minorsky designed another PID controller for the automated steering of ships and successfully tested it on board the battleship United States Ship (USS) New Mexico, but the Navy ultimately did not adopt the system (Bennett 1984). In 1912, Sperry designed the first autopilot gyroscopic stabilizer apparatus, which was used to improve the stability and control of aircraft (Chao, Cao, and Chen 2010). The autopilot connected gyroscopic heading and attitude indicators to hydraulically operated elevators and rudder. It would automatically adjust the control surfaces of an aircraft to maintain straight and level flight without a pilot's attention, significantly reducing the pilot's workload. His son Lawrence Sperry designed a smaller, lighter version and successfully demonstrated it in France in 1914. Lawrence Sperry flew the aircraft with his hands visibly away from the controls (Hallion 1980). Elmer Sperry Jr., the son of Lawrence Sperry, continued work on the same autopilot. In a test in 1930, a more compact and reliable autopilot kept a US Army Air Corps aircraft on a proper heading and altitude for three hours (Abbasi, Al-Saggaf, and Munawar 2014). 1.1.3. Automatic Telephone Switchboards and Control Systems in Industries The automatic telephone switchboard was patented by Almon Brown Strowger and put into service in 1892, along with dial telephones (Lipartito 1994). Automatic telephone switching facilitated the growth in call volume. Long-distance telephony made it necessary to enhance the signals and reduce their distortion. The invention of the Audion (triode) vacuum tube in 1906 by Lee De Forest led to the first amplifiers around 1912 (Brittain 2005). Harold Black came up with negative feedback on August 2, 1927, and published a classical paper on negative feedback amplifiers in 1934 (Kline 1993; Black 1934). Distortion levels were usually around 5% until 1934. His invention allowed distortion levels to be significantly reduced by negative feedback noise cancellation. This and other telephony applications contributed to control theory. The C. J. Tagliabue Company claimed to have installed the first pneumatic, automatic temperature controller on a milk pasteurization plant in New York City in 1907. The controller used mercury in a steel thermometer to operate a pilot valve which controlled the air pressure acting on the main valvewhich in turn influenced the flow of steam to the process (Bennett 2001). Edgar Bristol of the Foxboro Instrument Company invented the pneumatic “flapper nozzle amplifier” in 1914, which was highly nonlinear

The Dawn of the Robotic Age

127

as a 1% change of the full scale of the measurement caused a 100% change in the back pressure (Bennett 2001). These pneumatic controllers were then the industry standard. The development of the electrical industry and the electrification of other sectors from 1900 through the 1920s stimulated the practice of control systems. Relay logic was introduced as a method of controlling industrial electronic circuits using relays and contacts, and central control rooms became common in the 1920s. During the 1930s, the Foxboro flapper nozzle system incorporated integral and derivative terms, and industries rapidly adopted PID controllers (Bennett 2001). Military applications during World War II, such as fire-control systems and aircraft controls, contributed to and benefited from control theory. 1.1.4. Automatic Production Systems The automatic flour mill developed by Oliver Evans in 1785 was the first completely automated industrial process (Naude 1956). Milling began to be mechanized in the late eighteenth century, but manual labor was still required to move grain from one stage to the next. Evans used a bucket elevator to carry wheat from the bottom to the top of the mill to begin milling. He also developed the "hopper boy,” a device that gathered meal from a bucket elevator and spread it evenly over the drying floor (Hunter 2005). Michael Joseph Owens invented the bottle machine that could automate the production of glass bottles in 1903. It was financed by Edward D. Libbey and executed with the aid of engineers William Boch, C. William Schwenzfeier, and Richard LaFrance. The general-purpose bottle machine had an average production of 50,400 bottles a day. It introduced the safety, standardization, quality, and convenience of glass containers and cut labor costs by over 80%, so the Owens machines effectively ended child labor in glass-container plants (Cable 1999). The conveyor systems increased labor productivity and facilitated automatic production systems. Thomas Robins’ inventions in 1892 led to the development of a conveyor belt for carrying coal, ores, and other products. The Swedish company Sandvik invented and started the production of steel conveyor belts in 1901. Richard Sutcliffe invented the first conveyor belts in coal mines in 1905. Henry Ford introduced conveyor-belt assembly lines at Ford Motor Company's Highland Park, Michigan factory in 1913

128

Chapter 3

(Hoshimov, Rustamov, and Rozzokov 2018). Outputs from a production line often must be packaged and placed on pallets for transport. A pallet is a flat transport structure that stably supports goods while being lifted by a forklift. Manually placing boxes on pallets can be timeconsuming and tiring. Palletizers provide automatic means for stacking cases of products onto a pallet. The first mechanized palletizer was designed, built, and installed in 1948 by a company known as Lawson (now Lambert Material Handling). The row-forming palletizers, introduced in the early 1950s, arrange loads on a row-forming area first and then move them onto a layer-forming area (Palletizers.org No date). The in-line palletizer was developed in the 1970s for high-speed palletizing. It utilizes a continuous motion flow divider that guides the goods onto the desired area on the layerforming platform. Palletizing robots, introduced in 1963 by the Fuji Yusoki Kogyo Company (Erdo÷du 2021), have been widely used in palletizing systems since the early 1980s. The widespread use of instruments and the emerging use of controllers also led to the rise of continuous production of chemicals in the 1930s, many of which were previously made in batches. Texaco’s Port Arthur refinery became the first chemical plant to use digital control in 1959 (Stout and Williams 1995). With the fast development of the computer industry and the continuously falling prices of computer hardware, digital control began to spread rapidly since the 1970s. Robots also began to be widely used in industries. Today extensive automation is practiced in every manufacturing and assembly process.

1.2. Control Theory and Types of Control System Two types of control are often used in automation: sequential control and feedback control. An automatic sequential control system may trigger mechanical actuators in the correct sequence to perform a task. An automated feedback control system may regulate a variable at a reference value through a closed loop of sensors, control algorithms, and actuators. With the growing application of automation in industries, engineers and scientists started theoretical research into its mechanisms and conditions and developed the control theory, a science of control systems. Control systems can be classified according to their modes into logical control, onoff control, continuous control, and fuzzy logic.

The Dawn of the Robotic Age

129

1.2.1. Development of Control Theories The early governors could become unstable such that the engine’s speed increased when it should decrease to settle down to a steady speed, a phenomenon called “hunting.” In the nineteenth century, scientists attempted to analyze the governor’s mechanisms and determine the conditions for stable (non-hunting) operations. Jean Victor Poncelet in 1826 and 1836 and George Biddell Airy in 1840 showed differential equations that described the dynamic motion of the governor, but they did not arrive at the conditions for stable behavior. Airy’s paper in 1851 stated the conditions for stable operation but failed to show how he came to them (Bennett 1996). In 1868, James Clerk Maxwell published On Governors, widely considered to mark the beginning of the feedback control theory. By that time, mathematicians and physicists had understood that the stability of a dynamic system was determined by the real part of a complex root of the characteristic equation, and the system became unstable when the real part was positive. Maxwell analyzed several types of governors and provided linear differential equations for governor speed control. By examining the differential equations' coefficients, he showed that the system's stability could be determined for second-, third-, and fourth-order systems. He obtained the necessary and sufficient conditions for equations up to the fourth order (Maxwell 1868). Edward John Routh solved the problem formulated by Maxwell for the general class of linear systems in 1874. He published an extended treatise Treatise on the Stability of a Given State of Motion and expounded the stability criteria in 1877 (Routh 1877). Adolf Hurwitz independently obtained the stability criteria in 1895, called the Routh-Hurwitz theorem later. Nicolas Minorsky published a theoretical analysis of automatic ship steering in 1922 (Bennett 1984). He described the PID controller: 1) Proportional was the control required to steer the ship based on actual ship direction compared to the desired course set-point; 2) Integral was the amount of reset required to correct an amount of error; 3) Derivative was the attempt to see how far a process variable (ship course) had been from the set-point in the past and anticipating where the course correction would need to be in the future (Minorsky 1922). When the measured variable approached the set point rapidly, i.e., with a decreasing error, the power applied to correct the error would be smaller or backed off early. When the measured value began to move rapidly away from the set point, extra power was applied in proportion to that rapidity to move it back to the set point.

Chapter 3

130

1.2.2. Types of Control Systems We can classify the control systems into the following types: 1) Logic control: The system has components forming a logic circuit to control its output. The output is controlled by a combination of input or output conditions, and some logic system designs use Boolean logic. 2) Discrete (on-off) control: The system is the on-off control or bang-bang control, a feedback controller that switches abruptly between two states. The on-off controllers may be realized with any element that provides hysteresis. The thermostats used on household appliances are such on-off controllers. 3) Continuous control: The system continuously regulates the controlled process's activity. It usually involves taking measurements with a sensor, which is used as a feedback variable, and making calculated adjustments to keep the measured variable within a set range. Continuous control systems include a) Proportional control: its corrective action is based on the difference between the required set point (SP) and process value (PV) to reduce this difference, which is called the error; b) PID control: the integral term deals with the long-term steady-state errors and the derivative part is concerned with the rate-of-change of the error with time. 4) Fuzzy logic control: Fuzzy logic is a form of many-valued logic in which the truth values of variables may be any real number between 0 and 1, whereas the truth values of variables in Boolean logic may only be 0 or 1. The system attempts to apply the easy design of logic controllers to control complex, continuously varying systems.

1.3. Computers and Automation of Physical Processes Digital computers have greatly facilitated automation. The computer can perform sequential and feedback control in an industrial application. There are special-purpose microprocessors, such as programmable logic controllers (PLCs), and general-purpose computers. A PLC is a digital computer that typically automates industrial electromechanical processes, replacing many hardware components such as timers and drum sequencers in relay logic systems (Erickson 1996). General-purpose control process computers have increasingly replaced standalone controllers, and they can process data from a network of PLCs, instruments, and controllers to control many individual variables. For example, automated teller machines (ATMs) implement an interactive process in which a computer will perform a logic-derived

The Dawn of the Robotic Age

131

response to a user selection based on information retrieved from a networked database. Computers make it feasible to build complex industrial control systems (ICS), computer-based systems that monitor and control industrial processes in the physical world.

2. The Automation of Mental Processes The physical processes that can be readily automated tend to involve standardizable movements and minimum human intelligence. Human mental power distinguishes humans from lifeless objects, plants, and even higher animals. With the advancement of human knowledge, it is natural that people want machines that can help them perform tedious mental jobs. Computers have been the workhorse for the automation of human cognitive processes. Other information technologies and devices complement computers.

2.1. Computers A computer is a general-purpose device programmed to automatically carry out a set of arithmetic or logical operations. Computers used to be classified into two types: analog and digital. Analog computers use a direct mechanical or electrical model of the problem as a basis for computation. Currently, the term computer tends to refer to a digital one that consists of at least one processing element and some form of memory. The processing element, typically a central processing unit (CPU), contains a processing unit that carries out arithmetic and logic operations and a control unit that can change the order of operations in response to stored information. 2.1.1. Early Computers Charles Babbage originated the concept of a programmable computer in the early nineteenth century (Swade 1991). He conceptualized and invented the first mechanical computer, his difference engine, to aid navigational calculations in 1822. He developed an Analytical Engine that encapsulated most of the elements of modern computers in 1837 (Rojas 2020). Many early scientific computing needs were met by analog computers, among which slide rules and nomograms were the simplest. Sir William Thomson invented the first modern analog computer, the tide-predicting machine, in 1872 (Trainer 2004). James Thomson conceptualized the differential analyzer, a mechanical analog computer designed to solve differential equations in 1876 (Durand-Richard 2010). Harold Locke Hazen and Vannevar Bush constructed the first practical general-purpose differential

132

Chapter 3

analyzer from 1928 to 1931 (Brown 1981; Robinson 2005). The US Navy developed fire control systems using analog computers in the 1930s. It also developed electromechanical analog computers small enough for use on submarines. The Torpedo Data Computer (TDC) used trigonometry to solve the problem of firing a torpedo at a moving target. By World War II, most submarines in the US Navy had a TDC Mark 3 (Clymer 1993). The success of digital computers from the 1950s had spelled the end for most analog computing machines. Charles Eryl Wynn-Williams constructed a valve amplifier and thyratronbased automatic counting system with co-workers. He published the first recorded idea of using digital electronics for computing in his 1931 paper, “The Use of Thyratrons for High-Speed Automatic Counting of Physical Phenomena” (Wynn-Williams 1931). Alan Turing’s 1936 paper “On Computable Numbers, with an Application to the Entscheidungs Problem” defined an abstract machine that manipulated symbols on a strip of tape according to a set of rules (Turing 1936). This Turing machine was a mathematical model of computation devices. “A Symbolic Analysis of Relay and Switching Circuits,” written by Claude Shannon as his master’s thesis in 1937 and published in 1938, introduced the idea of using electronics for Boolean algebraic operations (Shannon 1938). Early digital computers were electromechanical, using electric switches to drive mechanical relays to perform calculations. German engineer Konrad Zuse built the Z1 during 1936–1938, the first freely programmable computer that used Boolean logic and binary floating-point numbers. He made an improved version in 1939, the Z2, replacing the arithmetic and control logic with electrical relay circuits (Weiss 1996). A system of data-manipulation rules is said to be Turing complete if it can be used to simulate any single-taped Turing machine. Zuse developed the Z3, the world's first working electromechanical programmable, fully automatic digital computer, in 1941. The Z3 was Turing complete. The Automatic Sequence Controlled Calculator (ASCC) was built by IBM and called Mark 1 by Harvard University’s staff. John von Neumann ran one of its first programs on 29 March 1944. Harvard Mark 1 was a general-purpose electromechanical computer. At the time, von Neumann worked on the Manhattan Project and needed to determine whether implosion was viable to detonate the atomic bomb (Guarnieri 2017). The electromechanical computer Z4, built between 1942 and 1945, was purchased by the Swiss Federal Institute of Technology Zurich and delivered in September 1950 (Weiss 1996); hence Z4 became the first commercial computer (O’Regan 2021).

The Dawn of the Robotic Age

133

John Vincent Atanasoff and Clifford E. Berry of Iowa State University developed the Atanasoff-Berry Computer (ABC) in 1942, the first “automatic electronic digital computer,” which was not Turing-complete and not programmable (Guarnieri 2017). Thomas Harold Flowers designed and built the world's first electronic digital programmable computer in 1943, the Colossus Mark 1, which was not Turing-complete (Tedre and Sutinen 2009). An improved Colossus Mark 2, Turing complete in theory, was first worked on 1 June 1944 (Good 1979). The ENIAC (Electronic Numerical Integrator and Computer), built under the direction of John Mauchly and J. Presper Eckert during 1943–1945, was the first electronic programmable computer Turing-complete (Tedre and Sutinen 2009). It was much faster and more flexible than the Colossus, adding or subtracting 5000 times a second, a thousand times faster than any other machine. Containing about 18,000 vacuum tubes, 1,500 relays, and hundreds of thousands of resistors, capacitors, and inductors, the machine weighed 30 tons and used 200 kilowatts of electric power (Weik 1955). 2.1.2. Modern Computers Modern computers are stored-program ones. A stored-program computer stores a set of instructions (a program) in memory that details the computation. Turing wrote the “Proposed Electronic Calculator” report, which specified such a device in 1945. John von Neumann also circulated his “First Draft of a Report on the EDVAC” (Electronic Discrete Variable Automatic Computer) in 1945, describing the logical design of a computer using the stored-program concept (Copeland 2000). The world's first storedprogram computer, the Manchester Small-Scale Experimental Machine (SSEM), nicknamed Baby and built by Frederic C. Williams, Tom Kilburn, and Geoff Tootill, ran its first program on 21 June 1948 (Burton and Cooper 2018). The first random-access digital storage device, the Williams tube, was implemented and tested for its reliability on SSEM. SSEM was the first working machine to contain all elements essential to a modern electronic computer. ENIAC was modified to run stored programs and first demonstrated as a stored-program computer on 16 September 1948, but it might have done so in March–May 1948 and become the first stored-program computer (Rope 2007). The SSEM was developed into a more usable computer, the Manchester Mark 1, whose first version was operational by April 1949. It was the Ferranti Mark 1 prototype, the world's first commercially available generalpurpose electronic computer, built by the British firm Ferranti and delivered

134

Chapter 3

to the University of Manchester in February 1951 (Campbell-Kelly 1990). The third stored-program computer was the electronic delay storage automatic calculator (EDSAC), constructed by Maurice Wilkes at the University of Cambridge Mathematical Laboratory in the UK. EDSAC ran its first programs on 6 May 1949 to calculate a table of squares and a list of prime numbers (Hartley 2011). The British firm J. Lyons & Co. produced the first commercially applied computer, LEO I, which ran its first business application in 1951 (Hendry 1987). The first stored-program computer in the US, the Binary Automatic Computer (BINAC), designed for Northrop Aircraft Company by the Eckert-Mauchly Computer Corporation, was delivered in September 1949. It was sometimes considered the world's first commercial digital computer (Stern 1979), but it was never fully functional. Whirlwind I, designed by Jay Forrester and Robert Everett in 1947 and developed by the Massachusetts Institute of Technology (MIT) Servomechanisms Laboratory for the US Navy, was sometimes called the first general-purpose digital computer. It was the first to use magnetic core memory and one of the first computers to calculate in parallel, operating on a complete 16-bit word every cycle in bitparallel mode. Whirlwind I first went online on April 20, 1951, and it was the most powerful computer from 1951 to 1954 (Forrester and Everett 1990). A computer program may consist of just a few to a multimillion instructions. A modern computer can execute billions of instructions per second (gigaflops). In most computers, individual instructions are stored as machine code, with each instruction given a unique number. Entire programs can be represented as lists of numbers and manipulated inside the computer similarly to numeric data. With early computers, programs were written as long lists of numbers (machine language). Each basic instruction could be given a short name indicative of its function and be easy to remember. These short names were collectively known as a computer's assembly language. A computer program called an assembler usually converts programs written in assembly language into machine language. The computers using vacuum tubes were considered the “first generation.” With the invention of the bipolar transistor in 1947, a team led by Tom Kilburn at the University of Manchester designed and built the first transistorized computer in the world, operational by 1953 (Anderson 2009). The computer was not the first fully transistorized machine because it still used a small number of vacuum tubes. The first fully transistorized machine Harwell CADET, built by the Atomic Energy Research Establishment at Harwell, ran a simple test program in February 1955 (Cooke-Yarborough

The Dawn of the Robotic Age

135

1998). IBM announced the IBM 608 transistor calculator in April 1955 as “the first completely transistorized computer available for commercial installation” (Apte and Scalise 2009). The MIT Lincoln Laboratory began to build the successful TX-0 (Transistorized Experimental computer zero) converted from the Whirlwind design in 1955, which became operational in 1956. The success of TX-0 led to a plan for the far more complex TX-1 and the completion of the scaled-back version TX-2 (McKenzie 1999). Transistorized computers could contain tens of thousands of binary logic circuits in a relatively compact space and were the “second generation.” The advent of the integrated circuit (IC) led to the “third generation” of computers. The idea of the IC was first conceived by Geoffrey W.A. Dummer, who presented its first public description on 7 May 1952 (Ganapati 2010). Jack Kilby at Texas Instruments and Robert Noyce at Fairchild Semiconductor invented the first practical ICs (Brock and Laws 2011). Kilby successfully demonstrated the first working integrated example on 12 September 1958 and submitted his patent application on 6 February 1959. Noyce devised his idea of an IC half a year later than Kilby (Kilby 1998; Phipps 2011). The development of ICs heralded an explosion in the commercial and personal use of computers and led to the invention of the microprocessor and the microcomputer including the personal computer (PC). The first commercially available single-chip microprocessor was the Intel 4004, marketed in 1971 (Aspray 1997). As Moore's law, named after Gordon Moore, the co-founder of Intel and Fairchild Semiconductor, described, the number of transistors in an IC doubles approximately every two years (initially every year) (Moore 1965). Computers using microprocessors with large-scale ICs (LSI) or very large-scale ICs (VLSI) were considered to be the “fourth generation.” 2.1.3. Supercomputers Supercomputers have a higher computational capacity than general-purpose computers. The concept appeared in the early 1960s, and the UNIVAC (Universal Automatic Computer) LARC (Livermore Advanced Research Computer), with a speed of 250 thousand instructions per second (KIPS), was considered one of the earliest supercomputers (Rosen 1969). It was built to a requirement published by Edward Teller to run hydrodynamic simulations for nuclear weapon design, and the first one was delivered to Livermore in June 1960. It had the title of the most powerful computer during 1960–1961. The IBM 7030 was IBM's first transistorized supercomputer, which held the most powerful computer title with a speed of 1.2 million instructions per second (MIPS) during 1961–1964. The Atlas

136

Chapter 3

Computer, with a rate of nearly one MIPS, jointly developed by the University of Manchester, Ferranti, and Plessey and officially commissioned in 1962, was also one of the world's first supercomputers (Rosen 1969; Tweedale 1992). The CDC 6600, with a speed of up to three million floating-point operations per second (megaFLOPS), designed by Seymour Cray and manufactured by Control Data Corporation (CDC), was generally considered the first successful supercomputer. It held the title of the most powerful computer from 1964 to 1968, and over 100 were sold over the machine's lifetime. It was succeeded by the CDC 7600 with a speed of 10 megaFLOPS (Rosen 1969), which held the title during 1969–1975. Seymour Cray and some colleagues founded Cray Research in 1972 and designed and manufactured Cray-1 with a speed of 136 megaFLOPS. It was the most powerful computer during 1976–1982 and was succeeded by the Cray X-MP/4 with a rate of 713 megaFLOPS, which held the title during 1983–1985 (Dongarra and Hinds 1986). The Cray-2, with four vector processors and a peak speed of 1.95 gigaFLOPS, had the title of the most powerful computer from 1985 to 1987 (Strawn and Strawn 2015). The fastest computer by 2022 was Fugaku, which achieved 1.42 exaFLOPS in the HPL-AI benchmark (Kudo et al. 2020). Its performance for the largest problem (Rmax) was 442.01 petaFLOPS. Fujitsu developed it for the Riken Center for Computational Science in Kobe, Japan. The current title holder is Hewlett Packard Enterprise Frontier with a Rmax of 1.102 exaFLOPS, hosted at the Oak Ridge Leadership Computing Facility (OLCF) in Tennessee, US, making it the world’s first exascale supercomputer. 2.1.4. Mobile Computing The continued miniaturization of computing resources and the advancements in portable battery life made truly portable computers a reality. The further improvement in miniaturizing computing resources and battery life allowed manufacturers to integrate computing resources into mobile phones. These so-called smartphones and tablets have become the dominant computing device on the market (Drill 2012). A smartphone is a mobile phone with an advanced operating system that combines features of a PC with other features useful for mobile or handheld use. IBM presented “Angler” on 23 November 1992, a prototype device combining a mobile phone and a personal digital assistant (PDA) into one device. The device allowed users to make and receive telephone calls,

The Dawn of the Robotic Age

137

facsimiles, emails, and cellular pages (Gold 2012). The speed of typical smartphone processors, evidencing the fast pace in computing power's progress, was about 2 to 4 gigaFLOPS in 2018, much faster than the most powerful supercomputers before 1987. In the twenty-first century, smartphones support other services besides telephony, such as text messaging, multimedia messaging service (MMS), email, Internet access, digital photography, and more general computing capabilities. A tablet is a mobile computer with a touchscreen display, circuitry, and battery in a single device. Some tablets also have sockets for subscriber identity module (SIM) cards and can be used as mobile phones. The boundary between smartphones and tablets with SIM cards is quite blurred. The development of transmission technologies, including computer networking, and the third, fourth, and fifth generations (3G, 4G, and 5G) of wireless mobile telecommunications technology has also played a massive role in providing ubiquitous entertainment, communications, and online connectivity.

2.2. Communications and Information Technology Sharing information and exchanging opinions, which the term communications describe, have been an essential aspect of human life from the beginning of humanity. Communication is the act of conveying intended meanings from one entity or group to another through the use of mutually understood signs and semiotic rules. When the exchange of information between participants includes the use of technology, telecommunication is considered to have occurred. According to the International Telecommunication Union (2019), telecommunication is the transmission of signs, signals, messages, writings, images, and sounds or intelligence of any nature by wire, radio, optical, or other electromagnetic systems. 2.2.1. Information Technology and the Digital Revolution The invention of the digital computer has greatly facilitated the development of information technology (IT) and communications. IT is the application of computers and computer networks (including the Internet) to store, retrieve, transmit, and manipulate data. It is a subset of information and communications technology (ICT). Claude Shannon, a Bell Labs mathematician, published his pioneering article, “A Mathematical Theory of Communication,” in 1948 and laid out the foundations of digitalization (Shannon 1948). With digitalization, it became possible to make copies identical to the original. In digital communications, the digital signal can be

138

Chapter 3

amplified and passed on without losing information. The widespread ownership of PCs by households has connected a large number of families as well as firms to the Internet, which has become the information highway. Digital data can be easily moved between media, accessed, or distributed remotely, and during the 1980s, digitally recorded music in the format of optical compact discs gradually replaced that in analog formats. The first proper digital camera Fujix DS-1P was created in 1988, and the Fujix DSX was marketed in December 1989 in Japan (Carter 2001). By the beginning of the twenty-first century, digital photography eclipsed traditional film in popularity. Motorola created the first mobile phone using analog communication, Motorola DynaTac, in 1983 (Murphy 2013). Digital cell phones have been sold commercially since 1991, when the 2G network opened in Finland, making mobile communication and information access available to most of the population (Liikanen, Stoneman, and Toivanen 2004). Text messaging became widely used around 2000. With 3G and 4G networks, video messages, and real-time video conversations have become commonplace. The 5G network started in 2019 (Letaief et al. 2019). It was meant to deliver multi-Gbps peak data speed, ultra-low latency, more reliability, and massive network capacity to connect virtually everyone and everything together, including machines, objects, and devices. 2.2.2. Computer Networks and Internet Computer networks are telecommunications networks developed to allow computers to exchange data. The earliest ideas for a computer network intended to allow general communications among computer users were formulated by computer scientist J. C. R. Licklider in April 1963 (Licklider 1963). Licklider worked in Bolt, Beranek, and Newman (BBN) from 1957 to October 1962, when he was appointed head of the Information Processing Techniques Office (IPTO) at the Advanced Research Projects Agency (ARPA), the United States Department of Defense. The best-known computer network is the Internet, the global system of interconnected computer networks that use the Transmission Control Protocol/Internet Protocol (TCP/IP) to link billions of devices worldwide. TCP/IP is an Internet communication protocol suite. The Internet evolved from the Advanced Research Projects Agency Network (ARPANET), an early packet-switching network and the first to implement the protocol suite TCP/IP. The term Internet was first used as a shorthand for internetworking by Vinton Cerf, Yogen Dalal, and Carl Sunshine in December 1974 in a Request for Comments, RFC 675 (Specification of Internet Transmission Control Program).

The Dawn of the Robotic Age

139

Initially funded by the ARPA, ARPANET started in 1969 (O'Neill 1995). It was a network composed of small computers called Interface Message Processors (IMPs), similar to the later routers. Its first successful message was sent by a student programmer Charley Kline, at 10:30 pm on 29 October 1969, from Boelter Hall 3420 at the University of California, Los Angeles (UCLA) (McDowall 2015). Other packet-switched networks, such as Mark I, CYCLADES, Merit Network, Tymnet, and Telenet, were also developed in the late 1960s and early 1970s. The packet-switching technology and the protocol suite TCP/IP used by the ARPANET became the technical foundation of the Internet. The initial ARPANET consisted of four IMPs, which grew to 40 by September 1973. Norway was the first country outside the US to be connected to the network when the Norwegian Seismic Array (NORSAR) was connected to the ARPANET by a transatlantic satellite link in June 1973. The Tanum Earth Station in Sweden also followed in 1973 with satellite links. At about the same time, a terrestrial circuit connected Peter T. Kirstein's research group at the Institute of Computer Science, the University of London, to the network (Livinginternet 2000). The Internet Protocol Suite (TCP/IP) standardization in 1982 permitted the worldwide proliferation of interconnected networks. Commercial Internet service providers (ISPs) emerged in the late 1980s and early 1990s, and the Internet rapidly expanded in Europe, Australia, and Asia. By the late 1980s, many businesses depended on computers and digital technology. 2.2.3. The World Wide Web The World Wide Web, invented by British scientist Tim Berners-Lee in 1989 while working at CERN (Conseil Européen pour la Recherche Nucléaire), is the primary tool billions of people use to interact on the Internet. The World Wide Web (WWW) is an information space where documents and other web resources are identified by Uniform Resource Locators (URLs), interlinked by hypertext links. It can be accessed via the Internet (Choudhury 2014). Berners-Lee built all the tools necessary for a working web: the first web browser/editor and the first web server in 1990. The first website describing the project was published on December 20, 1990 (Crew 2015). The WWW became publicly accessible on August 6, 1991 (Bryant 2011). On 30 April 1993, the European Organization for Nuclear Research, CERN, announced that the WWW would be free for everyone to use and develop (Brügger 2016). The introduction of the Mosaic web browser in 1993 by Marc Andreessen and Eric Bina allowed the web to become the most popular Internet protocol, eclipsing older

140

Chapter 3

protocols in use over the Internet, such as Gopher. The Mosaic was the first web browser to display inline images and the basis for later browsers such as Netscape Navigator and Internet Explorer (Strawn 2014). The Internet expanded quickly during the 1990s and became part of mass culture, and many businesses listed websites in their advertisements by 1996. The accessibility and speed of data transmission also improved fast. Most households used dial-up Internet access through a fixed landline during the 1990s because broadband service was too expensive. Internet access via broadband became the norm after 2000. The wireless application protocol (WAP), first launched in 1999, provided limited mobile access to the Internet. With 3G networks, mobile phone users could readily access the Internet via a cellular service provider. The wide availability of Wi-Fi (Wireless Fidelity) at public venues further increased Internet access by users with mobile devices such as smartphones and tablet computers. By 2016, tablet computers and smartphones exceeded PCs in Internet usage (Simpson 2016). Cloud computing, a type of Internet-based computing that provides shared computer processing resources and data to computers and other devices on demand, has entered the mainstream since the early 2010s (Sadiku, Musa, and Momoh 2014). Cloud computing and storage solutions provide users with various capabilities to store and process their data in third-party data centers. This allows companies to avoid upfront infrastructure costs and enables organizations to focus on their core businesses instead of spending time and money on computer infrastructure (Kim 2009). It has become a highly demanded service due to the advantages of high computing power, cheap cost of services, high performance, scalability, and accessibility.

2.3. AI at Basic Levels The primary function of computers is to perform calculations and make logical operations, which are essential components of human intelligence. Calculations and logic operations enable computers to play a vital role in control systems for automation, but more is needed to replace human beings in most jobs performed by human workers. Replacing human workers, in general, requires not just computing machines; it needs devices with AI comparable with human intelligence in one or more aspects. Many aspects of human intelligence can be mimicked with less complicated software and hardware to complete tasks that are more than straightforward computation and logic operations.

The Dawn of the Robotic Age

141

2.3.1. What is AI? Intelligence refers to a general mental capability that involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly, and learn from experience (Gottfredson 1997). A related term, intellect, is also used in studies of the human mind to refer to the ability of the mind to come to correct conclusions about what is true or real and how to solve problems. Therefore, it can be considered a branch of intelligence, reflecting mainly the logical and rational side without emotional and sensitive engagement (Roback 1922). In this sense, AI achieved so far is mainly artificial intellect. The term AI is used with two meanings: 1) the intelligence exhibited by machines or software and 2) the academic field, which is “the study and design of intelligent agents” (Bringsjord and Govindarajulu 2020). An intelligent agent is a system that perceives its environment and takes actions that maximize its chances of success. As such, computers and computer software actualize intelligent agents, and AI studies how to create computers and computer software capable of intelligent behavior. John McCarthy coined the term artificial intelligence in 1955 and defined it as “the science and engineering of making intelligent machines” (Rajaraman 2014; Andresen 2002). It is an interdisciplinary field in which computer science, mathematics, psychology, linguistics, philosophy, and neuroscience converge. The field was founded on the assumption that human intelligence could be precisely described such that a machine could simulate it. The utmost goal of the early AI researchers was to create intelligence equal to or exceeding human intelligence. The main objectives of the current AI research are to develop intelligent systems that simulate certain aspects of human intelligence, which include reasoning, knowledge, planning, learning, natural language processing, perception, and the ability to move and manipulate objects. The long-term goal is to simulate the critical aspects of human intelligence that currently lack useful tools, such as social intelligence, creativity, and general intelligence. 2.3.2. Historical Development of AI Artificial intelligent agents have been imagined since ancient times. Ancient legends and texts in China, Greece, the Near East, and the Middle East described different types of automatons or robots. AI in the public imagination is closely related to the idea of automatons with human-like

142

Chapter 3

intelligence, but AI can be possessed by immobile machines such as computers. 1) The birth of the AI field and the first AI winter The invention of the programmable digital electronic computer created an essential tool for AI research. Discoveries in neurology, information theory, and cybernetics also inspired AI research. The Dartmouth Summer Research Project on Artificial Intelligence (the Dartmouth Conference) during the summer of 1956 at Dartmouth College in Hanover, New Hampshire, was considered the seminal event for AI. The conference was organized by John McCarthy and formally proposed by McCarthy, Marvin Minsky, Nathaniel Rochester, and Claude Shannon in 1955. The attendees, including John McCarthy, Marvin Minsky, Allen Newell, Arthur Samuel, and Herbert Simon, became the leaders of AI research. The AI researchers made remarkable progress in the following years, producing programs enabling computers to play checkers, solve word problems, prove logical theorems, and speak English (Kuipers and Prasad 2022). AI research laboratories had been established worldwide by the middle of the 1960s. The Department of Defense heavily funded AI research in the United States. At the time, AI's founders were over-optimistic about the pace of achieving AI that equals general human intelligence, the strong AI. Herbert Simon predicted that “machines will be capable, within twenty years, of doing any work a man can do” (Makridakis 2017), and Marvin Minsky thought that “the problem of creating ‘artificial intelligence’ will substantially be solved” within a generation (Sands 2020). Sir Michael James Lighthill was asked by the Science Research Council (SRC) of the United Kingdom to compile a review of academic research in AI. The “Lighthill Report,” published in 1973, was highly critical of basic research in this area, stating that “in no part of the field have discoveries made so far produced the major impact that was then promised” (van Emden 2019; Agar 2020). Following the report’s publication, the British government ended support for AI research in all but three universities (Haenlein and Kaplan 2019). In the United States, the government also cut off all undirected exploratory research funding in AI in 1974. The next few years were later called an “AI winter” because of the difficulty in finding financing.

The Dawn of the Robotic Age

143

2) The revival and commercial success during the 1980s The rise of expert systems in the early 1980s revived AI research. An expert system was a computer system that emulated the decision-making ability of one or more human experts. Expert systems were introduced by the Stanford Heuristic Programming Project led by Edward Feigenbaum around 1965 (Feigenbaum 1994). The first commercial expert system was XCON, a production-rule-based system written by John P. McDermott of Carnegie Mellon University in 1978 for the Digital Equipment Corporation. It was an enormous success and was estimated to have saved the company 40 million dollars over just six years of operation (Sviokla 1990). In the 1980s, twothirds of Fortune 500 companies applied expert systems in daily business activities (Tuomi 2018). The Japanese government started a project to develop the fifth-generation computer in 1981, with objectives to write programs and build machines that could carry on conversations, translate languages, interpret pictures, and reason like human beings. In response to the Japanese initiative, several major computer and semiconductor manufacturers in the United States banded together and founded Microelectronics and Computer Consortium (MCC) in late 1982 (Frenkel 1985). The US Defense Advanced Research Projects Agency (DARPA, formerly ARPA) began to fund AI research again through the Strategic Computing Initiative (SCI) in 1983. Under the direction of the IPTO, the initiative was designed to support research on technologies required to develop machine intelligence in a ten-year time frame, from chip design and manufacture and computer architecture to AI software (Roland, Shiman, and Aspray 2002). In 1983, the British government sponsored the Alvey research program in IT, including Intelligent KnowledgeBased Systems (IKBS), AI, the Man-Machine Interface, and Natural Language Processing, among its focused areas (Oakley and Owen 1990). The thriving commercial expert systems brought in the industry to support them. Hardware companies, which built specialized computers called Lisp machines optimized to process the programming language Lisp preferred for AI, prospered in this AI boom. However, workstations by companies like Sun Microsystems offered a powerful alternative to Lisp machines by 1987, and later desktop computers built by Apple and IBM also provided a simpler and more popular architecture to run Lisp applications on. The market for specialized AI hardware collapsed, and many Lisp companies failed. Then it also became apparent in the United States that the SCI would fail to create machine intelligence at the levels that had been hoped for. The SCI canceled new spending on AI in 1988 (Roland, Shiman, and Aspray

144

Chapter 3

2002). A new AI winter began. Japan’s fifth-generation computer project had not met its objectives by 1991. 3) Challenging human intelligence since the 1990s A team led by Jonathan Schaeffer at the University of Alberta developed and improved Chinook, a computer program that could play checkers, during 1989–2007. Chinook was declared the Man-Machine World Champion in checkers in 1994 after six drawn games in a match against the world champion Marion Tinsley because Tinsley withdrew due to pancreatic cancer. Chinook defended its man-machine title in 1995 against Don Lafferty in a 32-game match, with one win and 31 draws over Lafferty (Schaeffer et al. 1996). Deep Blue, a chess-playing computer developed by IBM, won game one in a six-game match against the reigning world champion Garry Kasparov on 10 February 1996. However, the world champion defeated it by a score of four to two. On 11 May 1997, Deep Blue became the first computer chessplaying system to beat the reigning world chess champion in a six-game rematch by 3½ to 2½ (Newborn 2012). Modern chess programs like Houdini, Rybka, Deep Fritz, or Deep Junior are more efficient than Deep Blue. From November 25 to December 5, 2006, Deep Fritz, run on a PC containing two Intel Core 2 Duo CPUs, beat the then undisputed world chess champion Vladimir Kramnik by four to two (Newborn 2011). In 2005, IBM started a project to develop a question-answering system to play the Jeopardy quiz game. Watson, produced by this project, defeated the two greatest Jeopardy champions, Brad Rutter and Ken Jennings, by a significant margin in a Jeopardy quiz show exhibition match. Jennings commented in an article that IBM “sees a future in which fields like medical diagnosis, business analytics, and tech support are automated by questionanswering software like Watson. Just as new assembly line robots eliminated factory jobs in the twentieth century, Brad and I were the first knowledge-industry workers put out of work by the new generation of ‘thinking’ machines”(Gleason 2014). AlphaGo, a computer program developed by Google DeepMind in London to play the board game Go, became the first computer program to beat a professional human Go player on a full-sized 19×19 board in October 2015. Go had previously been regarded as a problem that was out of reach for the technology of the time. Its larger branching factor made using traditional AI methods such as Alpha-beta pruning, Tree traversal, and heuristic search

The Dawn of the Robotic Age

145

prohibitively difficult. In March 2016, AlphaGo became the first computer system to beat a nine-dan professional human Go player by winning four out of five games in a match with Go champion Lee Sedol (Chouard 2016). Contrary to the expectation of many people, AlphaGo showed an excellent understanding of the big picture. AlphaGo's algorithm used a Monte Carlo tree search to find its moves based on knowledge previously "learned" by machine learning (ML). It used artificial neural networks for extensive training with human and computer play data. AlphaGo Zero, which used no human play data for training, was more powerful than the original AlphaGo, which used human play data (Silver et al. 2017). AI achieved its greatest successes in the 1990s and the early twenty-first century and contributed to many different areas. AI is currently used for logistics, data mining, medical diagnosis, and many other areas throughout the technology industry. The Kinect, which provides a 3D body-motion interface for the Xbox 360 and Xbox One, and intelligent personal assistants in smartphones are examples of applications that use algorithms resulting from lengthy AI research. The success of AI research since the 1990s is due to many factors. The most important one is computers' rapidly increasing computational power. Other factors include a greater emphasis on solving specific sub-problems, a new commitment by researchers to solid mathematical methods and rigorous scientific standards, and the creation of new ties between AI and other fields working on similar problems. 2.3.3. Issues in Basic Level AI Most AI systems mentioned in the preceding section mimic human intelligence at some basic levels, but IBM’s Watson and Alphabet’s AlphaGo can be considered as AI that can perform a complete job system. For now, many basic-level AI systems are at work, and we take their convenience for granted. Sometimes we are unaware of their existence and may be unable to avoid their intervention. Reasoning and problem-solving, knowledge representation, planning for given conditions and objectives, learning, perception, and language processing are topics dealt with by AI at fundamental levels. 1) Reasoning and problem-solving Simulating human reasoning and problem-solving was an early endeavor of AI research. For easy problems, AI researchers developed algorithms that imitated humans’ step-by-step reasoning when solving puzzles or making logical deductions. Computers could readily generate and compare all the

146

Chapter 3

scenarios to arrive at the correct outcome. For complex problems, most of these algorithms require enormous computational resources, leading to “combinatorial explosions,” which means that the amount of memory or computer time required becomes astronomical. Human beings are flexible in their approach to solving problems, depending on the information available in terms of quantity and completeness, the time they have for a decision, their ability to understand or process the information, and the impact of their decision. For complicated problems with too much information or incomplete information, humans use fast, intuitive judgments rather than the conscious, step-by-step deduction, probably because they are either consciously or subconsciously aware that they do not have the capacity or time to process the information precisely. By the late 1980s and 1990s, AI research had developed successful methods for dealing with uncertain or incomplete information by employing concepts from probability and economics. Artificial neural network research is inspired by biological neural networks (the central nervous systems of animals). It attempts to simulate the structures inside the brain that give rise to this skill. Statistical approaches to AI mimic the probabilistic nature of the human ability to guess. AI also has made some progress at imitating human intuitive problem-solving. 2) Knowledge representation Most human reasoning and problem-solving depend on existing knowledge rather than working from scratch. Knowledge representation is dedicated to representing information about the world in a form that a computer system can utilize to solve complex tasks (Davis, Shrobe, and Szolovits 1993). Many problems machines are expected to solve will require extensive knowledge about the world. This knowledge includes objects, properties, categories, relations between objects; situations, events, states, time; and many others. Default reasoning, the breadth of common-sense knowledge, and the sub-symbolic form of knowledge have been among the most challenging problems in knowledge representation. The default reasoning issue concerns working assumptions that are not simply true or false in the way abstract logic requires. The breadth of common-sense knowledge issue is about the range and size of commonsense knowledge that enables computers to understand enough concepts to learn by reading from sources like the Internet. The sub-symbolic knowledge issue describes what people know as intuitions or tendencies

The Dawn of the Robotic Age

147

represented in the brain non-consciously and sub-symbolically. AI research has explored several solutions to these problems. 3) Planning Human beings can set goals and take steps to achieve them. They can visualize different states of the world and how their action can affect the probability of which state of the world occurs. AI agents are being developed with the ability to simulate such planning processes. When the agent is the only actor, it can be sure of its actions’ consequences. If there are other factors, the agent must periodically ascertain whether the world matches its predictions and change its plan accordingly, which requires the agent to reason under uncertainty. 4) Learning A key aspect of human intelligence is the ability to learn. Machine learning is the study of computer algorithms that improve automatically through experience (Jordan and Mitchell 2015). It has evolved from studying pattern recognition and computational learning theory in AI. Computational learning theory is concerned with the mathematical analysis of machine learning algorithms and their performance. Currently, machine learning mainly has three types: a) unsupervised learning, to infer a function to describe hidden structures from unlabeled data; 2) supervised learning, to infer a function from labeled training data; and 3) reinforcement learning, in which the agents take actions in an environment to maximize a cumulative reward. One of the most straightforward AI applications is classifying objects into different categories: classifiers. Classifiers are functions that use pattern matching to determine the closest match. In supervised learning, all the observations combined with their class labels are known as a data set. By inspecting the data set, machines can be trained to obtain the classifier capacity. When a new observation is received, that observation is classified based on previous experience. The performance of classifiers depends significantly on the characteristics of the data to be classified. 5) Natural language processing Natural language processing is intended to enable machines to read and understand human languages. Natural language user interfaces use linguistic phenomena such as verbs, phrases, and clauses that act as controls in computer-human interfaces for creating, selecting, and modifying data in

Chapter 3

148

software applications. A standard method of processing and extracting meaning from natural language is latent semantic indexing, which uses singular value decomposition (SVD) to identify patterns in the relationships between the terms and concepts in an unstructured text collection (Papadimitriou et al. 2000). A sufficiently powerful natural language processing system would enable acquiring knowledge directly from humanwritten sources. Natural language processing can be applied in information retrieval (or text mining), question answering, and machine translation. 6) Perception Machine perception is the capability of a system to interpret data, like how humans use their senses to relate to the world around them. It involves using input from sensors to deduce aspects of the world. Computer vision is the ability to extract high-dimensional data from the real world to produce numerical or symbolic information in decisions involving methods for acquiring, processing, analyzing, and understanding digital images. Speech, facial, and object recognition also belong to machine perception.

3. The Automation of Complete Job Systems The automation of a complete job system is the automation of a typical human worker’s professional activities, which are deemed an essential part of that job. We know that all the professional activities of a professional Go player are playing Go. AlphaGo can play with them and defeat any human players. Hence AlphaGo has automated the complete job system of Go players (Chouard 2016). IBM’s Watson beat the best players in the Jeopardy Quiz show, which automated the complete job system of Jeopardy players (Gleason 2014). Both jobs deal with non-physical operations. Automation of physical jobs depends on robots. A robot is a mechanical artificial agent, usually an electromechanical machine guided by a computer program or electronic circuitry. They can be autonomous or semiautonomous, and their appearances vary widely. The branch of technology that deals with the design, construction, operation, and application of robots is robotics.

3.1. Approaches for Automation of Complete Non-Physical Job Systems Although there has been heated discussion on social intelligence, creativity, and general intelligence for future AI, automation of complete non-physical

The Dawn of the Robotic Age

149

job systems does not require social intelligence, creativity, or general intelligence. The capability to perform the full responsibility of a human worker for a particular job is all that is needed to automate that complete job system. AI systems for non-physical jobs can share many common underlying approaches. 3.1.1. Logic and Symbolic Methods The development of formal logic played a vital role in automated reasoning. Many aspects of AI, for example, knowledge representation, problemsolving, planning, and learning, depend on logical methods. Inductive logic programming is a subfield of machine learning which uses logic programming as a uniform representation of background knowledge, examples, and hypotheses (Muggleton 1991). Fuzzy systems can be used for uncertain reasoning. The non-monotonic logic can deal with situations where the consequence relation is not monotonic. Default logic is a nonmonotonic logic proposed by Raymond Reiter to formalize reasoning with default assumptions (Reiter 1980). Circumscription is a non-monotonic logic created by John McCarthy to formalize the common-sense assumption. These forms of logic are designed to help with default reasoning and the qualification problem (Thielscher 2001). Allen Newell and Herbert Simon formulated the physical symbol system hypothesis (PSSH) and stated that a physical symbol system had the necessary and sufficient means for general intelligent action (Newell 1980; Simon 1981; Simon and Newell 1976). Symbolic AI was the collective name for all AI research methods based on high-level “symbolic” (humanreadable) representations of problems, logic, and search. It was the dominant paradigm of AI research from the mid-1950s until the late 1980s. AI researchers in the 1960s and the 1970s were convinced that symbolic approaches would eventually succeed in creating a machine with artificial general intelligence. There were different views and thoughts within the symbolic approach. Using the results of psychological experiments, Herbert Simon and Allen Newell attempted to simulate people’s problem-solving techniques (Simon and Newell 1971). John McCarthy felt that machines did not need to simulate human thought. Instead, they should be logic-based and able to catch the essence of abstract reasoning and problem-solving (McCarthy and Hayes 1969; McCarthy 1981). Researchers at MIT, such as Marvin Minsky and Seymour Papert, argued that no simple and general principle would capture all aspects of intelligent behaviors (Minsky and Papert 1988). From

150

Chapter 3

the 1970s, AI researchers began to appreciate that even simple AI applications required enormous knowledge. They built knowledge into AI applications, leading to the development and deployment of expert systems. 3.1.2. Search and Optimization Many problems in the real world are to find the best solutions among all feasible solutions. A problem's resolution often involves finding a sequence of actions that leads to a desirable goal. Logical proof can also be viewed as searching for a path that leads from premises to conclusions. Newell and Simon (1961) introduced means-ends analysis as a problem-solving strategy in 1961. Many search algorithms are based on optimization. For many problems, “heuristics” or “rules of thumb” are used to eliminate choices that are unlikely to lead to the goal (called “pruning the search tree”), limiting the search for solutions to smaller sample sizes. Search algorithms based on the mathematical optimization theory emerged in the 1990s. These algorithms include hill climbing, simulated annealing (SA), beam search, and random optimization (RO). Hill climbing is an iterative algorithm that starts with an arbitrary solution to a problem, incrementally changes a single element to find a better solution, and repeats this process until no further improvements can be found (Hinson and Staddon 1983). Simulated annealing is a metaheuristic to approximate global optimization in an ample search space (Rutenbar 1989). Beam search is a heuristic search algorithm that explores a graph by expanding the most promising node in a limited set (Rubin and Reddy 1977). Random optimization is a family of numerical optimization methods that do not require the gradient of the problem to be optimized (Matyas 1965). Evolutionary computation (EC) provides another optimization search based on Darwinian principles (Fogel 2000). They belong to the family of trialand-error problem solvers and can be considered global optimization methods with a metaheuristic or stochastic optimization character. The algorithms may begin with a population of candidate solutions and allow them to mutate and recombine, selecting only the fittest to survive each generation. Swarm intelligence (Hinchey, Sterritt, and Rouff 2007) and evolutionary algorithms (Whitley et al. 1996) are two essential subsets of evolutionary computation. 3.1.3. Decision with Uncertainty and Statistical Methods Many problems in AI require the agent to operate with incomplete or uncertain information, which might be addressed with tools from

The Dawn of the Robotic Age

151

probability theory and economics. Bayesian networks are a general tool used for reasoning, learning, planning, and perception (Stuart and Peter 2003; Jensen 2009). Maximization of expected utility explains how an agent can make choices. These methods have been responsible for many of AI's successes. The shared mathematical language has also permitted a high level of collaboration with more established fields such as mathematics, economics, or operations research. 3.1.4. Soft Computing/Computer Intelligence Soft computing, or computational intelligence, uses inexact solutions to computationally challenging tasks (Chaturvedi 2008). Its principal constituents are fuzzy logic (FL), evolutionary computation, machine learning, and probabilistic reasoning (PR). Neural networks are an example of soft computing as a part of machine learning, and AlphaGo uses neural networks for extensive training to learn the skills for playing the game Go. An artificial neural network is a model or system inspired by biological neural networks and used to estimate or approximate functions that can depend on many generally unknown inputs (Abdi 1994). Neural networks can derive solutions to problems that cannot be solved with complete logical certainty and where an approximate solution is often enough. They can be divided into two categories: acyclic or feedforward neural networks (where the signal passes in only one direction) and recurrent neural networks (which allow feedback). Neural networks can be applied to the problem of intelligent control or learning. A deep neural network is an artificial neural network with multiple hidden layers of units between the input and output layers, one of the structures supporting deep learning (Liu et al. 2017).

3.2. Automation of Complete Physical Job Systems Because of the highly specialized division of labor, many industrial jobs have already been replaced by less sophisticated robots. Many physical jobs had been automated before AI emerged. ATMs, vending machines, and industrial robots at assembly lines have replaced many human workers. However, in all these applications, humans must do some procedures, and automation of these remaining jobs poses the most severe challenge to robotics. Building a robot able to perform a complete physical job system is more challenging than developing an AI system able to perform an entire non-

152

Chapter 3

physical job system because in performing physical jobs, the robot needs to perceive correctly the properties of the physical objects to be handled as well as their immediate environment and handle the things with dexterity comparable with human workers. Both can be more challenging than developing an AI system to search for and optimize solutions to problems. 3.2.1. Automatons, Remote Control, and Electronic Robots Automatons exist in the myths and legends of many cultures around the world. In Greek myths, Hephaestus made Talos, a bronze robot, to protect Europa; Pygmalion of Cyprus carved an ivory statue (Galatea), which came to life (Mayor 2018). Humanoid automatons were thought to be built by Yan Shi in ancient China during the reign of King Mu of Zhou (976–922 BCE); the Mohist philosopher Mozi and his contemporary Lu Ban were said to have invented artificial wooden birds (ma yuan) that could successfully fly (Yates, Vaessen, and Roupret 2011). The clay golems of Jewish legend (Contrada 1995) and clay giants of Norse mythology (Fiorini 2018) were also legendary automatons. In the fourth century BCE, the Greek mathematician Archytas of Tarentum was reputed to have built the first self-propelled bird-like flying device (Reay 2014). The Greek engineer Ctesibius (c. 270 BCE) was said to have built water clocks with moving figures (Guarnieri 2010). Hero of Alexandria (10–70 CE) was believed to have created user-configurable automated devices and machines powered by air pressure, steam, and water (Tybjerg 2003). In 1066, the Chinese inventor Su Song built a water clock with mechanical figurines which chimed the hours (Yan and Lin 2002). The percussion instruments were operated with levers controlled by pegs (cams) in a programmable drum machine, which could be adjusted to change the rhythms. In Renaissance Italy, Leonardo da Vinci sketched plans for a humanoid robot, now known as Leonardo's robot, around 1495 (Moran 2006), and in Japan, the eighteenth-century Karakuri zui [Illustrated Machinery, 1796] described complex animal and human automata built by then (Kitamori et al. 1984). One such automaton was the Karakuri ningyǀ, a mechanized puppet. In France, between 1738 and 1739, Jacques de Vaucanson exhibited several life-sized automatons: a flute player, a pipe player, and a duck. The duck had over 400 moving parts in each wing (Fryer and Marshall 1979; Moran 2007b)

The Dawn of the Robotic Age

153

Torpedoes controlled remotely via wire were developed in the 1870s by John Ericsson (pneumatic), John Louis Lay (electric wire-guided), Victor von Scheliha (electric wire-guided), and Louis Brennan (electric wireguided). The Brennan torpedo, invented in 1877, was the world's first practical guided missile. The twin contra-rotating propellers were spun by steam-powered winding engines that rapidly pulled out wires from two drums wound inside the torpedo. The relative speeds of the twin winding engines that an operator could control electrically allowed the torpedo to be guided to its target (Gray 2004). The wireless remote control began to be tested in the 1890s. The British inventor Ernest Wilson demonstrated a remotely-controlled vessel on the Thames River in 1897, and he was granted a patent for a torpedo remotely controlled by “Hertzian” (radio) waves in the same year (Hammond and Purington 1957). Nikola Tesla publicly demonstrated his invention of a wireless-controlled boat in 1898 (Pérez Yuste 2008; Salazar-Palma et al. 2011). However, the radio remote control technology was not mature enough for practical use then. Archibald Low demonstrated a remotecontrolled aircraft to the Royal Flying Corps in 1917 and built the first wireguided rocket in the same year. He was known as the “father of radio guidance systems” for his pioneering work (Bartsch, Coyne, and Gray 2016). Alban J. Roberts built the first humanoid automaton, “Kaiser,” which was electrically powered and remotely controlled by light, in 1920. It could walk and skate without an external prop. He built a second humanoid automaton that rolled around on a singular round base (Hoggett 2009). Roy James Wensley filed a patent for Televox in 1927, which could be controlled by sounds of different pitches (Sharkey and Sharkey 2006). In 1928, William H. Richards invented a humanoid robot that could move its hands and head and named it Eric. It could be controlled through remote or voice control (Zielinska 2016). Also, in 1928, Japanese biologist Makoto Nishimura designed and constructed Japan's first robot, Gakutensoku. It could change its facial expression, move its head and hands via an air pressure mechanism, and write Chinese characters with a pen (Frumer 2020). The Westinghouse Electric Corporation built the seven-foot-tall (2.1 m) humanoid robot Elektro between 1937 and 1938. It could walk by voice command, speak about 700 words (using a 78-rpm record player), smoke cigarettes, blow up balloons, and move its head and arms (Bell 2018). The first electronic autonomous robots were created in 1948 by William Grey Walter of the Burden Neurological Institute in Bristol, UK. His robots were described as tortoises due to their shape and slow rate of movement.

154

Chapter 3

When they ran low on battery power, the three-wheeled tortoise robots could find their way to a recharging station. They used purely analog electronics to simulate brain processes (Bladin 2006). BEAM (from biology, electronics, aesthetics, and mechanics), a style of robotics that primarily uses simple analog circuits instead of a microprocessor to produce an unusually simple design, is the modern descendant of Walter's tortoises. Some researchers, such as Mark Tilden, are enthusiastic advocators of robotics using simple analog circuits instead of microprocessors (Tilden 2001). 3.2.2. Digital Robots George Devol invented the first digital and programmable robot in 1954 and filed for a patent in the same year, which was granted in 1961. Devol and Joseph Engelberger started the world's first robot manufacturing company, Unimation, in 1956. The first successful industrial robot was named Unimate and sold to General Motors in 1961 (Ballard et al. 2012). Unimate's instructions were stored on a magnetic drum. It was used to lift pieces of hot metal from die-casting machines. Since then, more and more robots have replaced humans in performing repetitive and dangerous tasks. From 1959, staff at the Rancho Los Amigos Hospital in Downey, California, started to work on orthotic manipulator arms with pneumatic power sources. They succeeded in developing a powered orthosis, the Rancho Electric Arm (REA), which had six joints that gave it flexibility similar to that of a human arm (Rahman et al. 2000). The Atomic Energy Commission (AEC) and the National Aeronautics and Space Administration (NASA) of the US were interested in robotic arms because they needed to have jobs done in dangerous and uninhabitable environments, so they contracted a project to develop a self-propelled anthropomorphic manipulator (SAM) for Rancho Los Amigos Hospital. Acquired by Stanford University in 1963, the Rancho Arm became one of the first computer-controlled robotic arms. Marvin Minsky at MIT created the Tentacle Arm with twelve joints in 1968; the arm was controlled by a PDP-6 computer from Digital Equipment Corporation (DEC) and powered by hydraulics (Moran 2007a). It moved like an octopus and could lift the weight of a person. Mechanical Engineering student Victor Scheinman working in the Stanford Artificial Intelligence Lab (SAIL), created the Stanford Arm when he worked on an arm solution in closed form in 1969 (Roth, Rastegar, and Scheinman 1974). In 1970, the Stanford Research Institute (SRI) built the first mobile robot capable of reasoning about its surroundings, Shakey, which combined

The Dawn of the Robotic Age

155

multiple sensor inputs, including TV cameras, laser rangefinders, and "bump sensors" to navigate (Kuipers et al. 2017). In 1973, Scheinman started Vicarm Inc. to manufacture his robot arms. He invented the programmable universal manipulation arm in 1976 and sold the design to Unimation (Roth 2016). A three-degrees-of-freedom manipulator, Freddy (1969–1971), and a five-degrees-of-freedom one, Freddy II (1973–1976), were built in the Department of Machine Intelligence and Perception at the University of Edinburgh (Srinivasa et al. 2012; Mowforth and Bratko 1987). In 1974, David Silver designed the Silver Arm, capable of fine movements replicating human hands with touch and pressure sensors providing feedback analyzed by a computer (Roshidul et al. 2010). The Stanford Cart, which relied primarily on stereo vision to navigate and determine distances, crossed a room full of chairs in 1979 (Moravec 1983). In 1973, KUKA Robotics in Germany created its first industrial robot FAMULUS, which had six electromechanically driven axes (Singh, Sellappan, and Kumaradhas 2013). KUKA developed the welding robot IR 6/60 in 1976. The first palletizing robot was introduced in 1963 by the Fuji Yusoki Kogyo Company (Erdo÷du 2021), but it began to be more widely used in the early 1980s. Palletizing robots or robotic palletizers have an endof-arm tool (end effector) to grab the product from a conveyor or layer table and position it onto a pallet. The Selective Compliance Assembly Robot Arm (SCARA) was created in 1978 as an efficient, four-axis robotic arm in the laboratory of Professor Hiroshi Makino at Yamanashi University in Japan. It could move vertically and horizontally but had limited wrist motion, making it more efficient and faster in assembly operations. SCARA robots were introduced to commercial assembly lines in 1981 (Milutinoviü and Potkonjak 1990). The Canadarm, the Shuttle Remote Manipulator System (SRMS), was developed by Spar Aerospace and its subcontractors according to a Memorandum of Understanding signed in 1975 between NASA and the Canadian National Research Council (NRC) and first tested in orbit on the space shuttle Columbia's STS-2 mission. Five were built and used on the space shuttle orbiters to deploy, maneuver and capture payloads. Those robotic arms were 15.2 m (50 ft) long and 38 cm (15 in) in diameter with six degrees of freedom (Aikenhead, Daniell, and Davis 1983). Takeo Kanade created the first Ādirect drive armā in 1981, with the arm's motors within the robot (Asada, Kanade, and Reddy 1981; Asada, Kanade, and Takeyama 1983). In 1984 Wabot-2, a musician humanoid robot able to

156

Chapter 3

communicate with a person, read a standard musical score with its eyes, and play tunes of average difficulty on an electronic organ, was created at Waseda University in Shinjuku, Tokyo, Japan (Kato et al. 1987). The Mobile Robot Laboratory at MIT produced a hexapedal insect robot named Genghis in 1989. Genghis used four microprocessors, 22 sensors, including infrared sensors and pressure-sensitive metal whiskers, and 12 servo motors. It followed rudimentary logic rules to chase any moving object it detected in its vicinity (Tedeschi and Carbone 2014). David Barrett, a doctoral student at MIT, built the biomimetic robot RoboTuna, designed for swimming and resembling a bluefin tuna, in 1996 to study how fish swim in water (Barrett, Grosenbaugh, and Triantafyllou 1996). The Cyberknife, a stereotactic radiosurgery performing robot invented by Dr. John Adler in 1994, could deliver tumor treatment with a comparable accuracy to surgery performed by human doctors (Gerszten et al. 2004). Honda’s humanoid research and development program started in 1986, and its P2 (Prototype Model 2) humanoid robot was first shown in 1996 and appeared more human-like in its motions (Hirose and Ogawa 2007).

3.3. Toward Robots That Perform Complete Job Systems Since the 1990s, robots have been used and tested in more fields with better functions. The Sojourner rover, weighing 23 lbs (10.43 kg), performed semi-autonomous operations on the surface of Mars for 83 days as part of the Mars Pathfinder mission in 1997, although it had been expected to operate for only seven days. It was equipped with an obstacle avoidance program and capable of planning and navigating routes to study the planet’s surface (Bajracharya, Maimone, and Helmick 2008). Honda presented the P3 humanoid robot in 1998 as a part of its continuing humanoid project (Hirose and Ogawa 2007). Sony introduced the AIBO, a robotic dog capable of interacting with humans, in 1999 (Fujita 2001). Honda revealed a new humanoid robot in 2000, named ASIMO, capable of running, walking, communicating with humans, facial and environmental recognition, voice and posture recognition, and interacting with its environment (Kaneko et al. 2009). In 2000, Sony revealed its Sony Dream Robot SDR-3X, a small humanoid robot (height 50cm) in development for entertainment, capable of bipedal motion. SDR-4X was announced in 2002; it incorporated “Real-time Integrated Adaptive Motion Control,” which activated 38 joints around the robot's body and advanced figuration and sound recognition functions. Sony

The Dawn of the Robotic Age

157

presented SDR-4XII renamed QRIO, in 2003, designed for use within the home and featured enhanced safety and durability and increased capacity for communication (Kuroki et al. 2003). In April 2001, the Canadarm2, a larger, more capable version of the Canadarm used by the space shuttle, was launched into orbit and attached to the International Space Station. The Canadarm2 was 17.6 m when fully extended and capable of handling large payloads of up to 116,000 kg. It had seven motorized joints and could move end-over-end to reach many parts of the space station in an inchworm-like movement (Gibbs and Sachdev 2002). In the same month, the Unmanned Aerial Vehicle Global Hawk made the first autonomous non-stop flight over the Pacific Ocean in 22 hours, from Edwards Air Force Base in California to Royal Australian Air Force (RAAF) Base Edinburgh in Southern Australia (Wilson 2002). The company iRobot released its first generation of Roomba, a robotic vacuum cleaner that could change direction upon encountering obstacles and detect dirty spots on the floor, in 2002 (Jones 2006). Nine generations had been introduced by the end of 2019. In 2012, iRobot acquired Evolution Robotics, which launched the mopping robot Mint in 2010 and rebranded Mint as Braava. The company had sold nearly 50 million robots worldwide by the end of 2022 (iRobot 2023). On 3 January 2004, the Mars rovers Spirit (Mars Exploration Rover-A), launched in 2003, landed on the surface of Mars. It completed its planned 90-sol (the solar day on Mars) mission and went on to function effectively over twenty times longer than NASA planners expected. It also logged 7.73 km (4.8 mi) of driving instead of the planned 600 m (0.4 mi). Spirit became stuck in soft soil on May 1, 2009, and its last communication with Earth was sent on 22 March 2010 (Sanderson 2010). On 25 January 2004, the Opportunity (Mars Exploration Rover – B) landed in Meridiani Planum on the other side of the planet. It remained active until 10 June 2018, having logged 28.06 miles (45.16 kilometers) of driving (Callas, Golombek, and Fraeman 2019). In 2005, a team of researchers at Cornell University implemented a selfassembling machine. The robots comprised modular cubes called “molecubes,” each containing identical machinery and the complete computer program for replication. Since molecubes were not natural resources and robots could not make molecubes, they were not genuinely self-replicating robots (Weiss 2005). Also, in 2005, Honda presented a new version of its ASIMO robot with new behaviors and capabilities (Honda

158

Chapter 3

2005). In 2006, the team at Cornell University revealed its “Starfish,” a four-legged robot capable of self-modeling and learning to walk after damage (Bongard, Zykov, and Lipson 2006). In 2007, TOMY launched the humanoid entertainment robot i-sobot, which was 6.5 inches in height, fully articulating and bipedal. It could walk like a human, performing kicks, punches, entertaining tricks, and special actions (Watanabe and Yoneda 2009). In October 2010, Google announced that its completely autonomous cars had been driven successfully in traffic on American roads and highways. New models sold by Tesla Motors since October 9, 2014, have the option to have autopilot, which allows hands-free driving in many situations. As of March 2016, Google had test-driven their fleet of driverless cars in autonomous mode, a total of 1,500,000 mi (2,400,000 km). Google’s selfdriving car project separated from Google on December 13, 2016, and became Waymo, a subsidiary company of Alphabet, Google’s parent company. In October 2018, Waymo announced that its test vehicles had traveled in an automated manner for over 10,000,000 miles (16,000,000 km), increasing by about 1,000,000 miles (1,600,000 kilometers) per month. On December 5, 2018, Waymo was the first to commercialize a fully autonomous taxi service, “Waymo One,” in Phoenix, Arizona, United States. The cars were being monitored in real time by a team of remote engineers, and there were cases where the remote engineers needed to intervene. The Google/Waymo autonomous cars' success shows the progress of self-driving technology in just a few years, as all 15 teams competing in the 2004 DARPA Grand Challenge failed to complete the course. In that competition, none of the robots successfully navigated more than five percent of the 150-mile road course, leaving the $1 million prize unclaimed (Ozguner, Stiller, and Redmill 2007). On February 24, 2011, Robonaut 2, the latest generation of astronaut helpers, was launched aboard Space Shuttle Discovery on the STS-133 mission and delivered to the International Space Station (ISS). It was the first humanoid robot in space, and it was hoped that one day upgraded versions could venture outside the station to help spacewalkers make repairs or additions to the station or perform scientific work (Diftler et al. 2011). With the fast progress in AI and robotics, industrial robots have become increasingly widely used in different sectors, performing jobs more cheaply, with greater accuracy and reliability than humans, or under conditions dangerous or unpleasant for humans. More and more of today's robots are inspired by nature, especially by biological materials and made of soft and

The Dawn of the Robotic Age

159

deformable materials. These robots lead to the emergence of bio-inspired robotics and soft robotics. Soft robotics is a branch of robotics that deals with non-rigid robots constructed with soft and deformable materials (AlbuSchaffer et al. 2008). These materials include silicone, plastic, fabric, rubber, or compliant mechanical parts like springs.

4. Automation, Robots, and AI in the Modern Economy Robots have been widely used in many different sectors for various jobs. They can be classified into different types according to specific criteria. Robots can be classified into 1) industrial; 2) service; 3) agricultural; 4) educational; 5) healthcare and personal helpers; 6) robots for entertainment and toys; 7) Research robots; and 8) military robots, etc., according to their usage. Some have been widely used for a while, and others are still in their early stages.

4.1. Industrial Robots and Automation The International Organization for Standardization ISO 8373 defines a manipulating industrial robot as “an automatically controlled, reprogrammable, multipurpose, manipulator programmable in three or more axes, which may be either fixed in place or mobile for use in industrial automation applications.” Industrial robots are extensively used in automobile production, mass production of printed circuit boards (PCBs), and palletizing and packaging manufactured goods. They outperform human workers in speed, accuracy, and reliability. Manipulating robots have also been used in academic and industrial research and development. The high-throughput screening of drug candidates and characterization of a compound’s biochemical and pharmaceutical properties in the pharmaceutical industry rely heavily on manipulating robots. The Robotic Age is a time when robots and AI systems become the main actors on the economic stage. By now, robots have been used widely in some industrial sectors and professions, while there is no noticeable progress in some others. Industrial robots were first widely used in manufacturing. Their typical applications include welding, painting, assembly, picking and placing for PCBs, packaging and labeling, palletizing, product inspection, and testing. Manufacturers such as FANUC, Motoman, KUKA, and ABB are important suppliers of industrial robots.

160

Chapter 3

4.1.1. The Automotive Industry The automotive industry was the first sector to use industrial robots. In 1960, Devol sold the first Unimate robot to General Motors. It was shipped in 1961 and installed at General Motors' die-casting plant in New Jersey (Ballard et al. 2012). It was used to lift hot pieces of metal from a die-casting machine and stack them. General Motors also used the Unimate robot for spot welding of car bodies. Soon other companies such as Chrysler and the Ford Motor Company saw the necessity for large Unimate purchases and followed General Motors' lead to install Unimates in their manufacturing facilities (Gasparetto and Scalera 2019). The introduction of robotics to the manufacturing process effectively transformed the automotive industry. The first Unimate was sold at a $35,000 loss, but the rapid adoption of the technology enabled the company to build the robotic arms for significantly less and thus begin to turn a substantial profit (Fifth Wave Manufacturing 2022). Automobile factories have now become dominated by robots. A typical factory contains hundreds of industrial robots working on fully automated production lines. A vehicle chassis on a conveyor of an automated production line is welded, glued, painted, and finally assembled at a sequence of robot stations. In Japan, for every 10,000 workers in the automotive industry, there are 1,414 industrial robots. In Germany, the United States, and the Republic of Korea, there are more than 1,000 industrial robots for every 10,000 workers in the automotive industry (Saba et al. 2021). Welding, material handling, and painting robots are critical in modern automotive manufacturing. The most widespread application of robots in the automotive industry is welding. Robot welding uses mechanized programmable tools to completely automate a welding process by performing the welding and handling the part. Robot welding took off in the 1980s when the automotive industry used robots extensively for spot welding. Robot arc welding has proliferated recently and commanded about 20% of industrial robot applications. Robotic welding has proven to be a technology that helps many original equipment manufacturers increase accuracy, repeatability, and throughput. Signature image processing (SIP) is a valuable weld quality assurance technology. It has been developed since the late 1990s for analyzing electrical data in real-time collected from automated, robotic welding, thus enabling the optimization of welds. Statistical analyses of the signatures following image processing operations provide a quantitative assessment of the welding process, revealing its stability and reproducibility and providing fault detection and process diagnostics.

The Dawn of the Robotic Age

161

Painting a car body with industrial robotic arms has been an essential application of robots in the automotive industry. Since painting is a very complex process, which is hard work and toxic, and highly qualified painters are hard to find, it is much easier for a company to use robots for this application. Painting requires consistent results throughout the whole production, which is more readily achieved by robots, given the sizes of cars. Using painting robots can also reduce waste material since they are equipped with a flowmeter so that the same amount of paint is distributed on each part. Another type of robot widely used in the automotive industry is the materialhandling robot. Material handling encompasses various product movements on the shop floor, including part selection, transferring, picking and packing, palletizing, loading and unloading, machine feeding, or disengaging. Accordingly, there are picking and packing robots, part transferring robots, material removal robots, machine tending robots, palletizing robots, and assembly robots, for which end-of-arm tooling can be customized to cater to the manufacturing requirements used for all applications. In the automotive industry, picking, part transferring, loading, and unloading are widely used as components of the production lines or their logistic support. Mobile robots can be used to transfer materials or perform other jobs that need to be done at different locations. They follow markers or wires on the floor or use vision or lasers to reach the correct places. Automated guided vehicles (AGVs) transport goods around extensive facilities, such as warehouses, container ports, or hospitals. An intelligent AGV drops off goods without needing lines or beacons in the workspace. Robots that help human workers perform their tasks have also been developed. Human workers are still contributing to the completion of the cars despite the large number of robots used in production lines since the final assembly tasks are generally handmade. For example, wiring and many other operations, such as wheel installation, remain human tasks. Doing hard or dirty jobs all day can be exhausting for a human worker and cause injuries. BMW has introduced collaborative robots to its assembly line to help human workers in their daily tasks. These human-friendly robots on the BMW assembly line perform the final assembly of the car doors with a door sealant that keeps sound and water out of the car. Besides human-robot collaboration, robot-robot collaboration has also been used in automotive production. On the welding line in the Chinese automotive plant, Great Wall Motors (GWM), handling robots and welding robots collaborate to perform more than 4,000 welding operations on the car body in an 86-second cycle time, including the transferring operations. The ABB IRB 7600 precisely

162

Chapter 3

places the panel at the right location, while an ABB IRB 6640 performs the welding operations (Salimbeni 2016). 4.1.2. Electrical/Electronics The electrical/electronics industry is the second largest user of industrial robots. Mass-produced PCBs are almost exclusively manufactured by surface mount technology (SMT) component placement systems, commonly called pick-and-place machines or P&Ps. The pick-and-place robots are robotic machines that place surface-mount devices (SMDs) onto a PCB (Prasad 2013). They are used for the high-speed, high-precision placing of a broad range of electronic components onto the PCBs, like capacitors, resistors, and integrated circuits. Small electronic components are removed from strips or trays and placed onto PCBs with great accuracy. They can put in hundreds of thousands of parts per hour, far outperforming a human in speed, accuracy, and reliability. Those robots are typically equipped with SCARA manipulators. The PCBs are used in computers, consumer electronics, and industrial, medical, automotive, military, and telecommunications equipment. The contract manufacturer, Foxconn Technology, replaced 60,000 human workers with robots in a single factory in 2011 (Brown 2016). Foxconn had been criticized for years for inadequate working conditions, and a string of suicides at its manufacturing facilities a few years ago had been attributed to those conditions. Foxconn employees had also complained of being overworked, sleeping in less-than-ideal dormitory conditions, and not being paid fairly. The increased use of robots reduced input from human workers and attenuated the problems caused by inadequate working conditions. 4.1.3. The Chemical, Pharmaceutical, and Life Sciences Industry Production processes in the chemical, pharmaceutical, and life science industries are usually highly automated, and robots can perform various tasks, such as dispensing, material handling, and packaging (Williams 2009). Robots in these sectors can perform other roles. A critical application of industrial robots is employing them in dangerous environments and having them handle hazardous materials. Robots are ideal for use in hazardous environments because they remove people from direct exposure to unfriendly conditions such as radioactive or highly explosive materials. They are not bothered by fumes that might be irritating or toxic to human workers, particularly under close or prolonged exposure. Robots also handle non-hazardous materials that could produce potentially explosive dust.

The Dawn of the Robotic Age

163

They can also be used for sterile handling and assembling the small, delicate parts used in prosthetics, implants, and medical devices. 4.1.4. The Food Industry The vast quantity and variety of food and beverage products pose a serious challenge to distribution and supply chain managers, which drives them to embrace automation and robotic technologies at an unprecedented rate. At the same time, food manufacturing has also grown into rather complex processes which involve preparing, cooking, sorting, packaging, palletizing, etc. Industrial food robots are being integrated into food manufacturing to save time and space and improve cleanliness and safety. Food manufacturing robots are commonly used in dispensing, packaging, pick-and-placing products into containers, and palletizing for transfer and shipment. Robots can also form food, grind, and cut food products. Many incorporate sensors to check for thickness, color, and any inconsistencies in food products. In Japan’s Kura sushi restaurant chain, robots help make the sushi, and conveyor belts replace waiting staff while managers remotely monitor restaurant operations. The robots developed by San Francisco startup company Momentum Machines can make gourmet-quality hamburgers from freshly ground meat and grill them to order (Ford 2015). 4.1.5. Mining Mining robots are designed to solve several problems due to the hazardous nature of mining, particularly underground mining. Autonomous, semiautonomous, and teleoperated robots used in mining have significantly increased recently. Several vehicle manufacturers provide autonomous trains, trucks, and loaders that load material, transport it from the mine site to its destination, and unload it without human intervention (Rogers et al. 2019; Nanda et al. 2010). Currently, the mining industry still uses many human workers, particularly in developing countries with low labor costs, so there is less incentive to increase efficiency through automation (Paredes and Fleming-Muñoz 2021). Mine equipment automation comes in four different forms: 1) remote control, usually referring to control of mining vehicles such as excavators or bulldozers with a handheld remote device; 2) teleoperation, referring to control of mining vehicles by an operator at a remote location with the use of cameras, sensors, and possibly additional positioning software; 3) driver assist, referring to partly automated control of mining machines where only some of the functions are automated, and operator intervention is needed;

164

Chapter 3

and 4) full automation, referring to the autonomous control of one or more mining vehicles where robotic components manage all critical vehicle functions, and implement control without the need for operator intervention.

4.2. Service Robots and Automated Ordering and Checking-out Systems Service robots perform jobs in the economy besides manufacturing and producing raw materials. They include professional service robots and personal and domestic service robots. Domestic service robots are usually simple robots dedicated to single-task jobs in the home, for example, the vacuum cleaner robot Roomba. The International Organization for Standardization defined a service robot as a “robot in personal use or professional use that performs useful tasks for humans or equipment” (ISO 8373) in 2012. In its 2021 edition of ISO 8373:2021, medical robots intended to be used as medical electrical equipment or electrical systems are classified as a third category next to industrial and service robots. 4.2.1. Automated Retail, Self-Checkout, and Online Shopping Vending machines have provided some rudimentary forms of automated service for customers. They can be viewed as a simple combination of automated ordering, serving, and checkout. Amazon has launched a grocery store without lines or checkout counters, Amazon Go, an 1800-square-foot retail space in the company’s hometown of Seattle. Shoppers there can grab the items they want and leave, and the order will be charged to their Amazon account afterward. It uses computer vision and sensors to detect what items shoppers take from the store. You start by scanning an app as you enter Amazon Go and shop as in a regular supermarket. The sensors throughout the store identify the items in your cart and charge them to your account when you walk out the door, so there is no waiting in line or fussing around with self-checking machines. It is easy to imagine that robots will replace many employees in service sectors in the future. The retail food industry has started to apply automation to the ordering process. Many supermarkets and even smaller stores are rapidly introducing self-checkout systems. In some countries, fast food chains like Kentucky Fried Chicken (KFC) and McDonald’s have online ordering and delivery services. The online ordering systems are fully automated. Some cafes and restaurants, for example, KFC and McDonald’s, have utilized mobile and tablet “apps” to make ordering more efficient by customers ordering and paying on their devices. Ordering food from KFC and McDonald’s on

The Dawn of the Robotic Age

165

mobile phones is popular in China. McDonald's has also introduced touchscreen ordering and payment systems in many restaurants. Customers order using touch panel screens in Japan’s Kura sushi restaurant chain. The wide use of the Meituan and ElHme apps in ordering takeaways has changed the dining habits of many Chinese, especially the young, and increased their job choices. Many people now make a living as delivery couriers riding Ebikes, delivering all types of goods, and collecting parcels to be posted from customers. Online shopping could be considered a form of automated retail for ordering and payment through automated online transaction processing systems. Online shopping has made Amazon, Alibaba, and eBay among the most valuable worldwide. Alibaba and eBay are online shop platforms, while Amazon is an online shop and web service provider and a platform for other retail firms. Tencent provides a WeChat shop platform. The popularity of online shopping leads to huge volumes of deliveries by courier services, which in turn has stimulated the development of courier services and reduced their costs due to economies of scale. The increase in labor productivity of the courier service and transport, in general, helps the development of e-commerce. Large online firms like Amazon and Jingdong are developing automated delivery systems using autonomous surface vehicles or drones. The overall effects of online retail platforms are to drive down the profit margin of all retail shops and make physical shops differentiate into either high-end boutique shops, where in-person shopping is much more preferred to online shopping, or low-end convenience shops in residential communities. High-end in-person shopping is likely to be challenged by the development of virtual reality that mimics the effects of trying clothes on in person. Vending machines have been competing with convenience shops in various places. 4.2.2. Automated Catering Service Vending machines for hot drinks can make drinks of various flavors, and they are the rudimentary form of combining orders, preparing drinks, and paying. Customers make payments to the machines and obtain goods from them. At the University of Texas at Austin, the coffee kiosk installed by a start-up called Briggo LLC is run by a robotic barista. It is segmented into brewed coffee and espresso drinks, which can also be prepared according to customers’ milk, syrup, and sweetener preferences (Hill 2012). In many Japanese sushi restaurants, customers pick up the dishes from conveyor belts. Some other restaurants have also automated food delivery to customers’ tables using a conveyor belt system, for example, the catering

166

Chapter 3

service for the Beijing 2022 Winter Olympics (Lin et al. 2021; Gao et al. 2022). Robots are used to send meals to restaurant dining tables, replacing some waiting staff functions. In Japan’s Kura sushi restaurant chain and other chains, robots are helping people make sushi (Sakamotosupa and Allensupb 2011). 4.2.3. Transport and Communications Current technology is sufficient to operate automated vehicles on tracks and roads if there is no error from human road users. Automated vehicles' problems are more about avoiding other users, especially when they make mistakes. Underground track transport systems are now largely automated, and the drivers play a less significant role. The US Armed Force’s application of drones (crewless or unmanned aerial vehicles, UAVs) in surveillance and air strikes shows that automated air transport is feasible in the future, and so is automated shipping. There are two aspects to make automated transport possible: the traffic control system for roads, tracks, waterways, and airspace; and the automated vehicle. In London’s underground system, the ticket office has disappeared in many stations, and passengers can buy tickets from machines and recharge their travel cards. They also travel using cash or credit cards with contactless payment functions. In the British railway system, especially the London Underground, ticket inspectors at the gates have long been replaced by automated ticket barriers. For the time being, refilling supplies are still done by human workers. When maintenance robots can refill supplies and fix breakdowns, human workers will no longer be needed. An automated highway system (AHS) or Smart Road is a proposed intelligent transportation system (ITS) technology designed to relieve traffic congestion by drastically reducing following distances and headway, thus allowing a given stretch of road to carry more cars. To implement such a system, the vehicles should have power steering and automatic speed controls coordinated by a computer. As demands for safety and mobility have grown and technological possibilities have multiplied, interest in automation has grown. In 1991, the United States Congress authorized more than $650 million over six years for intelligent transportation systems and demonstration projects in the Intermodal Surface Transportation Efficiency Act (ISTEA). The ISTEA stipulated that “the Secretary of Transportation shall develop an automated highway and vehicle prototype from which future fully automated intelligent vehicle-highway systems can be developed. Such development shall include research into human factors to

The Dawn of the Robotic Age

167

ensure the success of the man-machine relationship. This program aims to have the first fully automated roadway or an automated test track in operation by 1997. This system shall accommodate equipment installation in new and existing motor vehicles” [ISTEA 1991, part B, Section 6054(b)](Lyons 1995). A prototype AHS, the National Automated Highway System Consortium (NAHSC) project, was tested in San Diego County, California, in 1997 along Interstate 15. The AHS system placed sensory technology in cars that could read passive road markings, communicate with each other, and organize themselves without drivers’ intervention (Horowitz and Varaiya 2000). Car manufacturers such as BMW, Mercedes-Benz, Tesla, Toyota, and Volkswagen have been developing autonomous cruise control systems (Ashley 1998; Bimbraw 2015). The European Safe Road Trains for the Environment (SARTRE) Project was started in September 2009 to investigate the implementation of platooning on unmodified European motorways. Vehicle platooning is a convoy of vehicles in which a professional driver in a lead vehicle heads a line of closely following vehicles. Each following vehicle autonomously measures the distance, speed, and direction and adjusts to the vehicle in front. Therefore, drivers in the following vehicles can do other things while the platoon proceeds toward its long-haul destination. All vehicles are detached and can leave the procession at any time. This vehicle platooning technology was successfully demonstrated at the Volvo Proving Ground near Gothenburg, Sweden, with a lead truck followed by a single car. It was also carried out in Barcelona, Spain, with a lead truck followed by three cars driven entirely autonomously at speeds of up to 90 km/h (56 mph). The gap between the two adjacent vehicles was no more than 6 m (20 ft). Volvo Trucks and Volvo Car Corporation participated in SARTRE (Coelingh and Solyom 2012). Automation is accomplished by combining sensors, computers, and communications systems in vehicles and along the roadway. Fully autonomous cars will allow closer vehicle spacing and higher speeds, enhance road safety by reducing the opportunity for driver error, and increase fuel economy. An autonomous vehicle (driverless, self-driving, and robotic) can sense its environment and navigate without human input. In an October 2010 blog post, Google announced that its completely autonomous vehicles had been driving successfully in traffic on American roads and highways. In August 2012, it announced that its self-driving car had completed over 300,000 autonomous-driving miles (500,000 km)

168

Chapter 3

accident-free. In May 2014, it revealed a new prototype of its driverless car, which had no steering wheel, gas pedal, or brake pedal, and was fully autonomous. Google’s fleet of driverless vehicles had covered 1,500,000 mi (2,400,000 km) in autonomous mode by March 2016. In December 2016, Alphabet (Google's parent company) announced that the self-driving car technology unit would be spun off as a company called Waymo. The idea of autonomous cars is not new, as experiments have been conducted on automating cars since at least the 1920s. Some promising trials took place in the 1950s, but the first (semi-)autonomous cars appeared in the 1980s. Carnegie Mellon University started its Navigation Laboratory (Navlab) and Autonomous Land Vehicle (ALV) projects in 1984 and built its first vehicle Navlab 1, in 1986 using a Chevrolet panel van and Navlab 2 in 1990 using a US Army HMMWV (High Mobility Multipurpose Wheeled Vehicle). Mercedes-Benz and Bundeswehr University Munich's EUREKA Prometheus Project was begun in 1987. In 1994, the twin robot vehicles VaMP and VITA-2 from the Prometheus Project drove more than one thousand kilometers on a Paris multi-lane highway in standard heavy traffic at speeds up to 130 km/h. Since then, many automotive makers have started their autonomous car projects (Bimbraw 2015). In the United States, the National Highway Traffic Safety Administration (NHTSA) released a formal classification system for automated vehicles in 2013. SAE International (formerly Society of Automotive Engineers, SAE), an automotive standardization body, published its classification system in 2014. The NHTSA adopted the SAE standard in September 2016 and abandoned its own system (Chan 2017). In mid-October 2015, Tesla Motors rolled out version 7 of their software, including Tesla Autopilot capability in the United States. On 9 January 2016, Tesla released version 7.1 as an over-the-air update, adding a new “summon” feature that allows cars to self-park at parking locations without the driver in the car. Tesla's autonomous driving features could be classified as somewhere between level two and level three under the five levels of vehicle automation. The system would not read traffic signals or obey stop signs in urban driving. Starting on 19 October 2016, all Tesla cars have been built with hardware to allow full self-driving capability at the highest safety level (SAE Level 5) (Talpes et al. 2020).

The Dawn of the Robotic Age

169

4.2.4. Professional Services Besides retail and commercial services, there is another group of services called professional services. The term “professional” conveys the meaning of rigorous training and expertise. These professions include doctors, lawyers, teachers, accountants, tax advisers, financial advisers, management consultants, architects, journalists, etc. There are some common features in those professions: 1) they need a relatively high level of formal training to do the job; 2) they have special knowledge which is not readily accessible to ordinary people due to technical jargon, the requirement of foundational knowledge, cost of the information in traditional media, or physical inconvenience; 3) they provide solutions to problems met by ordinary people or advise people; 4) their solutions and way of thinking are generally logic-based. These features make professionals more vulnerable to advanced levels of AI. Due to the high information costs in the past, it was more expensive financially and timewise to search for relevant information alone than to go to a professional for advice. The widespread use of the Internet, especially the WWW and search engines, dramatically reduces the cost of acquiring relevant information. Most professional knowledge can be obtained on the Internet at negligible monetary costs. Only time and pre-requisite knowledge stand in the way for ordinary people to utilize professional expertise on the Internet fully. Correct and accurate information is often mixed with inaccurate and sometimes wrong information online. Internet users need to spend time finding the correct and precise information, even if they have the foundational knowledge to select the right thing. As most people may need to gain the foundational knowledge to select the correct and accurate information on the Internet, or they do not want to spend the time and energy to work out the correct information, they will still turn to professionals for advice and solutions to their problems. However, with the advance of AI, machines and software programs will replace the experts and provide the same or better advice than average experts. Both Watson and AlphaGo use deep learning, which can be used as more generalpurpose AI to play the role of experts in different professions after learning from human knowledge and self-training. Narrative Science, Inc. developed an AI engine, Quill, which could produce automated articles in various areas, including sports, business, and politics. Top media outlets now use Narrative Science technology. Quill’s predecessor was a software called “StatsMonkey,” created by students and researchers at Northwestern University’s Intelligent Information Laboratory. It was designed to automate sports reporting by transforming

170

Chapter 3

objective data about a particular game into a compelling narrative. The researchers set up Narrative Science to commercialize the technology. The company produced the more powerful Quill, designed to be a generalpurpose analytical and narrative-writing engine. Quill could collect data from various sources, perform analysis and then weave all the information into a coherent narrative. Narrative Science claimed that the report produced by Quill matched that made by the best human analysts (Wright 2015; Carlson 2015). The AI language model ChatGPT introduced on November 30, 2022, by OpenAI (https://openai.com/) appears more powerful and can create essays, movie scenes, poems, jokes, and advertisements on demand (Kirmani 2022; Gordijn and Have 2023; Stokel-Walker 2022). ChatGPT was first finetuned from a model in the GPT-3.5 series to interact conversationally. “The dialogue format makes it possible for ChatGPT to answer follow-up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests” (Open AI 2022b). GPT stands for Generative Pretrained Transformer, a class of autoregressive language models. It is a machine learning model that uses autoregressive techniques to predict a word in a sequence of words based on the words coming before it. Tasks such as natural language processing and machine translation can use autoregressive language models. On 14 March 2023, Open AI released GPT-4, which “is a large multimodal model (accepting image and text inputs, emitting text outputs) that… exhibits human-level performance on various professional and academic benchmarks” (Open AI 2023). Several ChatGPT clones or extensions for particular usages have appeared. Google unveiled its chatbot Bard on February 6, 2023, by its CEO Sundar Pichai (Pichai 2023). Bard was initially powered by Google’s Language Model for Dialogue Applications (LaMDA), whose database is updated in real-time. On March 21, 2023, Google granted access to limited users in the US and UK on a rolling basis, and now it is more widely available. Bard is currently powered by Google’s most advanced large language model (LLM), PaLM 2, released in April 2022 and will allow Bard to be much more efficient. On May 10, Google added image capabilities, coding features, and app integration to Bard (Hsiao 2023). People comparing Bard with ChatGPT tend to think Bard performs better. From the cases of Watson, AlphaGo, Quill, ChatGPT, and Bard, we can now see that AI has been able to perform deep learning, retrieve data and analyze data to provide a superior or comparable solution to those produced by human experts. With further developments in AI, the costs of such

The Dawn of the Robotic Age

171

systems will drop enormously, and their function will be much more potent than Watson, AlphaGo, Quill, ChatGPT, and Bard. By then, there will be little room left for human experts. Watson has been developed to provide medical advice to physicians and other health professionals. As physicians’ physical examinations of patients have largely been replaced by machine and laboratory tests, they are more replaceable by AI than surgeons who perform operations manually. In the future, robots might and probably will be more dexterous than human surgeons in performing operations and replace human surgeons entirely. 4.2.5. Professional Documentation In some professions, such as lawyers, the preparation and compilation of documents are essential. Document automation is the computer-assisted creation of electronic documents and the assembly of relevant pre-existing text, data, and documents. Document automation software is used primarily in the legal, financial services, and risk management industries. Automation software applications can automatically fill in the correct document variables based on the transaction data. They can also automate the preparation of documents and organize documents into logical groups. Automation technology's role in producing legal documents has been widely recognized. In large law firms, document assembly systems are increasingly used to systemize work, such as complex term sheets and the first drafts of credit agreements (Ribstein 2010). Following the Legal Services Act 2007 in the UK, large institutions have broadened their services to include legal assistance for their customers. Legal services move toward commoditization, which results in high-volume, low-margin services being ‘packaged’ and provided to a mass-market audience. Most of these companies use some element of document automation technology to provide legal document services over the web (Dale 2019). The insurance sector also tends to have a large number of documents to be processed. Policy documents can also be hundreds of pages long and include specific information on the insured. Document automation can assist in the preparation and assembly of insurance documents. 4.2.6. Home Automation and Domestic Robots Home automation or the smart home refers to the automation of household appliances and features in residential dwellings, the residential extension of building automation. It involves controlling and automating lighting, heating (smart thermostats), ventilation, air conditioning, security, and

172

Chapter 3

home appliances such as washers, dryers, ovens, refrigerators, or freezers using Wi-Fi for remote monitoring. Switches and sensors are connected to a central hub, sometimes called a “gateway,” to form the home automation system. The system is controlled via the central hub with a user interface that interacts with a wall-mounted terminal, mobile phone software, tablet computer, or web interface. Popular communications protocols include X10, Ethernet, RS-485, 6LoWPAN, Bluetooth LE (BLE), ZigBee, and Z-Wave, but only a few worldwide accepted industry standards; therefore, the smart home market is heavily fragmented. The home automation market was worth US$5.77 billion in 2015, US$ 84.33 billion in 2021, and is expected to reach 210.15 billion by 2030 (Reportlinker 2022). Home automation suffers from platform fragmentation and lack of technical standards, a situation where the variety of home automation devices, in terms of both hardware variations and differences in the software running on them, makes the task of developing applications that work consistently between different inconsistent technology ecosystems challenging (Hosek et al. 2017). Customers may be hesitant to bet their Internet of Things (IoT) future on proprietary software or hardware devices that use proprietary protocols that may fade or become difficult to customize and interconnect. Home automation is also called domotics. It includes domestic robots connected to the domotic network to perform or help in household chores. It may also have robots dedicated to some particular purposes, for example, to help administer medications and alert a remote caregiver if the patient is about to miss their medicine dose. The care-providing robot FRIEND, developed at the Institute of Automation Technology (IAT) of the University of Bremen, is a semi-autonomous robot designed to support disabled and older adults’ daily activities. Robotic vacuum cleaners and floor-washing robots have been widely used in cleaning floors with sweeping and wet-mopping functions.

4.3. Agricultural Robots and Automation Agricultural robots or agrobots are robots deployed for agricultural purposes. The application of robots in agriculture today is mainly at the harvesting stage. Robots are developed to automate picking fruit in orchards, which may cost less than hiring human pickers. Driverless tractor/sprayer and sheep shearing robots are designed to replace human labor. Before the Industrial Revolution, almost all the labor force worked on the land. The mechanization of agriculture has dramatically reduced the number of farmers and farm workers. The application of robots in agriculture will

The Dawn of the Robotic Age

173

further reduce farming jobs or the hours worked by farmers. A possible emerging application of robots or drones is for weed control. In the near future, sowing and planting, as well as field management, might be carried out by robots under minimum human supervision or support (Talaviya et al. 2020; Bogue 2016). 4.3.1. Harvesting Robots Harvesting crops like wheat, barley, corn, or soybeans has long been mechanized and very efficient, although making autonomous harvesters and combines is still worthwhile. The new progress focuses on harvesting fruits, milking, and sheep shearing. Fruit-picking robots are usually grasping devices for harvesting the target crop. The design usually consists of two mechanical fingers that can move synchronously when performing their task. Another solution is the use of suction grippers (vision-systems.com). These robots must be carefully designed not to bruise the fruit while picking. Citrus fruit robot pickers have thus far been the focus of research and development. A strawberry harvester developed by Shibuya Seiki in Japan in 2013 could pick a strawberry every eight seconds. The robot identifies which strawberries are ready to harvest using three cameras, then picks the ready one and gently places it in a basket. Agrobot has also developed a strawberry-picking robot, SW6010, with two workstations qualified to monitor and pack the fruit (Bogue 2020). Global Positioning System (GPS)- and vision-based self-guided tractors and harvesters have already been available commercially. The manipulator (robot) allows the gripper and other end-effectors to navigate their environment. The manipulator consists of four-bar parallel links that maintain the gripper's position and height. Robots can also be used in livestock applications such as automatic milking, washing, and castrating. Robots for milking (the milk bot) are widespread among British dairy farms because of their efficiency (Butler, Holloway, and Bear 2012). The milk bot need not move and can complete the milking task to a consistent and particular standard. 4.3.2. Horticulture and Field Management Robots can be used for horticultural tasks such as thinning, pruning, weeding, spraying, organizing potted plants, and monitoring. The RV 100 developed by Harvest Automation Inc. is designed to transport potted plants in a greenhouse or outdoor setting. The functions of RV100 in handling and organizing potted plants include spacing capabilities, collection, and

174

Chapter 3

consolidation. The benefits of using horticultural robots such as RV 100 include high placement accuracy, autonomous outdoor and indoor function, and reduced production costs (Paul et al. 2020). The IBEX autonomous weed spraying robot can work in extreme terrains. A rice-planting robot has been developed by the Japanese National Agricultural Research Centre (Torii 2000).

4.4. Educational Robots and AI Systems Robots have been used mainly as educational assistants to teachers so far. As the success of computers in the Jeopardy game, chess, and Go has shown that AI can hold or access more knowledge more accurately, humanoid robot lecturers may teach university students in the future. The education system is among the longest-existing institutions in human society. It has become even more critical in modern times. With the new challenges created by technological progress, people look to education for solutions. Many economists do not think that the unemployment caused by new technologies is an issue of workers being replaced by machines and computer software. Instead, they consider it a skill mismatch problem between workers and employers. For those economists, education and retraining hold the key to solving the unemployment issue. However, teachers may be in danger of being replaced by robots and computer software. 4.4.1. ICT, AI, and Robots in Education ICT has influenced education greatly, especially in delivering lectures in higher education. Although many university teachers still prefer the chalkand-blackboard approach for delivering courses, most teachers have shifted to prepared PowerPoint files and projectors. The chalk-and-blackboard method requires lecturers to be very familiar with the taught materials and able to write legibly and perhaps elegantly. However, writing on a blackboard or whiteboard with colored pens takes time and leaves less time for explaining the content. With prepared files, lecturers have more time to introduce concepts and theories and need less time to memorize lecture contents. As students often have electronic files of their teacher’s lecture notes before the class, they need not take notes from the blackboard. This tends to have two different effects on students’ learning outcomes: 1) not taking notes gives students more time and attention to listen to their teacher and think actively; 2) students lose the benefits of learning and focusing through note-taking.

The Dawn of the Robotic Age

175

AIs such as Watson, ChatGPT, and Bard may well take over the role of teachers when robots possess a course’s relevant materials/knowledge better than human teachers and can continue to learn. Lecturing might not be more complex than competing in the Jeopardy game at the highest level. Lectures can be replaced by video recordings of the best teachers in the discipline, while personal tutoring can be performed by Watson-, ChatGPT-, and Bardtype AI. Android AI robots may also deliver real-time lecturing in the future. It is imaginable that with cloud computing, robots will know more than human teachers about the subject they teach. Robots can retrieve and process data much faster, and AI can be replicated at almost zero marginal costs. With proper training on performance, it can model the best practice of human teachers and may eventually replace nearly all human teachers. 4.4.2. Massive Open Online Course (MOOC) A MOOC is an online course aimed at unlimited participation and open access via the web (McAuley et al. 2010). It is a recent and widely researched development in distance education. MOOCs emerged from the open educational resources (OER) movement in 2008 and became popular in 2012. The term MOOC was coined in 2008 by Dave Cormier of the University of Prince Edward Island in response to a course called Connectivism and Connective Knowledge (CCK08). MOOCs from private, non-profit institutions emphasize prominent faculty members and expand existing distance learning offerings (e.g., podcasts) into free, open online courses. Some other E-learning platforms such as Khan Academy, Peer-toPeer University (P2PU), Udemy, and ALISON can also be viewed as similar to MOOCs and work outside the university system or emphasize individual self-paced lessons.

4.5. Healthcare Robots and Personal Helpers Robots in healthcare may provide two primary services: a) personal care for patients, disabled and elderly individuals, and b) support to pharmacies and hospitals (Bogue 2011; Robinson, MacDonald, and Broadbent 2014). The aging population means increasing numbers of older adults to be cared for and fewer young people to care for them. Basic robotic assistants, such as the Handy 1, were developed in the 1980s. Semi-autonomous robots, such as FRIEND (Functional Robot arm with a user-frIENdly interface for Disabled people), can assist older and disabled people with everyday tasks. FRIEND allows paraplegic patients to perform tasks without help from other people like therapists or nursing staff.

176

Chapter 3

4.6. Robots for Entertainment and Toys The early humanoid robots had been mainly used for entertainment, and those currently in development are also used occasionally to entertain people. Robotic toys may also be viewed as robots for entertainment. AIBO, the first robotic pet, grew out of Sony's Computer Science Laboratory (CSL). People in the culture, science, and technology sectors are considered creative knowledge workers. AI is at its best when dealing with logic-based knowledge. Comparatively, it is less capable of dealing with human emotions and mimicking manual dexterity. In science and technology, logic-based knowledge is more dominant than emotions and manual dexterity, so robots might play a significant role in the future. Culture is more emotionally and manually involved, but even here, there are many disciplines in which knowledge and logic are more important than emotions and manual dexterity.

4.7. Culture, Sports, and Entertainment There are two aspects through which IT, computers, and robotics influence culture. One is that computers, game consoles, and the Internet provide a medium or platform both for electronic games and traditional forms of entertainment and culture. The other is that robots or computer programs can deal with and generate cultural products. With the advance of ICT, the traditional music industry has been eclipsed by online shops, downloads, and file exchanges. The music of performers has become more accessible to ordinary people. Films, especially classical ones, are freely accessible to Internet users. Consumers with mobile Internet access can be entertained by various cultural products available on the Internet. Technology increases the reach of superstars and probably their incomes while leaving little room for local talent to develop and support themselves by providing services to local consumers. AI has begun to acquire the ability to produce cultural products. The Quill mentioned earlier can write various news reports and analyses. In July 2012, the London Symphony Orchestra performed a composition, Transits – Into an Abyss, composed entirely by an AI algorithm called Iamus running on a cluster of computers. Researchers at the University of Malaga in Spain have designed an artificial composer Iamus, which uses Melomics, a compositional computer algorithm (Ball 2012). It has already produced millions of unique compositions in the classical modernist style. Another artificial composer

The Dawn of the Robotic Age

177

Melomics 109, which is geared toward contemporary popular music and has been active since November 2013, also uses the Melomics algorithm (Farrell 2015). Simon Colton, a professor of creative computing at the University of London, has built an AI program called “The Painting Fool.” The software can identify emotions in photographs of people and then paint an abstract portrait that conveys their emotional state. It can also generate imaginary objects using techniques based on genetic programming. Open AI revealed Dall-E, a deep learning model that uses a version of GPT-3 modified to create images, in January 2021. Its successor, Dall-E2, announced by Open AI in April 2022, can generate more realistic images at higher resolutions from natural language descriptions (Open AI 2022a). Midjourney, an AI program and service created and hosted by the independent research lab Midjourney (https://www.midjourney.org/), can create artwork according to requests in texts (Roose 2022).

4.8. Research Robots and Automation Apart from manipulating robots used in laboratories to perform repetitive or dangerous jobs, robots in the future can achieve many other jobs in scientific research that humans have to do. Several robots have been developed to explore volcanos. Most space probes, such as the early lunar probes, the Voyager probes, and the Galileo probes, can be considered robots. Mars rovers are robots. The impact of AI and automation on science and technology can also be examined from two perspectives. One is laboratory or experiment automation, which increases the productivity of scientists and technicians, implying that some jobs could be lost due to increased productivity. Before the emergence of automated measurements, experimental scientists and technicians had to spend much time on manual tasks. Laboratory automation addresses the manual part of the research. The other is the participation of AI in forming research ideas, designing experiments, analyzing research results, and conceptualizing new theories, which is the mental part of the research. Automation has been extensively employed in laboratories, especially for industrial research and development. At least since 1875, there have been reports of automated devices for scientific investigation (Olsen 2012). Dr. Masahide Sasaki opened the first fully automated laboratory in the early 1980s (Sasaki et al. 1998). Syracuse University’s analytical chemistry

178

Chapter 3

professor, Daniel J. Macero, and a graduate student, Brian J. McGrattan (Olsen 2012), introduced one of the first multifunctioning, programmable, computer-controlled robots for laboratory use. In 1993, Dr. Rod Markin at the University of Nebraska Medical Center created one of the world's first clinical automated laboratory management systems (Markin 1993). Automation has yet to become widespread in academic research laboratories due to its high cost and the usual small sample size in academic research. Integrating low-cost devices with standard laboratory equipment will increase the adoption of automation. Autosamplers, coupled with an analytical instrument and periodically providing samples for analysis, are commonly used in laboratory automation. Laboratory automation can increase productivity, elevate experimental data quality and reduce lab process cycle times. It might enable experimentation that otherwise would be impossible. Robots are widely used in the pharmaceutical, biotechnological, and chemical industries for high-throughput screening, combinatorial chemistry, automated clinical and analytical testing, diagnostics, and largescale biorepositories. AI might play a big part in the mental aspect of discovery and invention processes. As shown by AlphaGo and AlphaZero, computer algorithms with deep learning capacity do not just imitate the ways human players do; they can have moves and strategies that human players have never thought of (Silver et al. 2017). In 2009, Hod Lipson, the director of the Creative Machines Lab at Cornell University, and Ph.D. student Michael Schmidt built a system called Eureqa based on genetic programming. They gave Eureqa the control of a double pendulum, a device with one pendulum attached to and dangling below another. When both pendulums are swinging, the motion is highly complex. Eureqa took only a few hours to develop several physical laws describing the pendulum's movement, including Newton’s second law (Schmidt and Lipson 2009). ChatGPT can review research literature, identify data, and generate research ideas (Dowling and Lucey 2023; Pavlik 2023). Google Bard can also perform these tasks. In response to the popularity of ChatGPT in essay writing, the Science journal has banned the use of ChatGPT or similar tools in writing papers for publication by it or listing ChatGPT as an author. The journal will view articles written with ChatGPT as plagiarizing from the AI bot (Thorp 2023). Another example of AI’s impact on research is DeepMind’s AlphaFold, a computer algorithm that can work out the 3-D spatial structure of proteins (Jumper et al. 2021). Before AlphaFold entered this field, with efforts from thousands of structural biologists for many years, there were about 190K

The Dawn of the Robotic Age

179

proteins whose 3-D spatial structure had been elucidated by experiments. AlphaFold increased that number to 200 million in one year. Most structural biologists in the 2020s need to look for other tasks after the appearance of AlphaFold, in the same way as DNA sequencing researchers in the 1990s after the arrival of autonomated DNA sequencers. It seems inevitable that computer algorithms will replace many scientists in the future. The question is to what extent. Perhaps all scientists who view research as a means for wealth and fame will be replaced by AI and intelligent robots, only those who view research as a tool to solve their curiosity will stay.

4.9. Military Robots and AI Systems The US military used hundreds of robots in Iraq and Afghanistan, such as iRobot's Packbot and the Foster-Miller TALON, to defuse roadside bombs (Voth 2004). The Special Weapons Observation Reconnaissance Detection System (SWORD) comprises a weapons system mounted on the standard TALON chassis, and three were deployed to Iraq in 2007. Automated or remotely controlled aerial, water surface or underwater vehicles are the most widely used military robots. There are concerns over using military robots that make some of their decisions autonomously (Singer 2009; Pfaff 2019). A UAV or unmanned aircraft system (UAS) is an aircraft without a human pilot aboard, commonly known as a drone. UAVs originated in the early twentieth century mainly in military applications to provide practice targets for training military personnel. World War I saw its continued development. Nikola Tesla considered building a fleet of crewless aerial combat vehicles in 1915. Archibald M. Low began to build a powered UAV as an “Aerial Target” in 1916 and had its first trial on 21 March 1917 at Upavon Central Flying School near Salisbury Plain. The Dayton-Wright Airplane Company invented a Kettering aerial torpedo that would explode at a pre-set time in 1918. Peter Cooper Hewitt and Elmer Ambrose Sperry tried to develop the Hewitt-Sperry pilotless Automatic Airplane in 1917-1918 (Prisacariu 2017). The film star and model airplane enthusiast Reginald Denny developed the first scaled remote-piloted vehicle in 1935. World War II witnessed more UAV developments for applications to train antiaircraft gunners and to fly attack missions. Nazi Germany produced the V-1 guided missile and used various UAV aircraft during the war. UAVs also expanded in commercial, scientific, recreational, agricultural, and other applications. After World War II, the Australian Government Aircraft Factories developed the Jindivik, a radio-controlled target drone, from 1948 to 1952. Teledyne Ryan

180

Chapter 3

began to produce Firebee I in 1951 as a target drone for US Air Force, Navy, and Army. Beechcraft developed Model 1001 as an out-of-sight target for the US Navy in 1955 (Prisacariu 2017). The progress in computing power and miniaturization manufacturing has made UAVs more versatile and available to more institutions and individuals. Flight control for UAVs is similar to crewed aviation. Key technologies, like plane flight dynamics, helicopter flight dynamics, controls, automation, and multirotor flight dynamics, had been researched for improving crewed aviation long before the rise of UAVs. Most UAVs use a radio frequency front-end to connect the antenna to the analog-todigital converter and a flight computer that controls avionics. The UAV avionics may be capable of autonomous or semi-autonomous operation. Non-military applications or intended ones of UAVs include policing and surveillance, product deliveries, aerial photography, agriculture, and drone racing, among many others. The US military has used remotely controlled crewless or unmanned combat air vehicles (UCAVs) for reconnaissance and attacking ground targets. They use UAV Predators and Reapers for counterterrorism operations and military missions in war zones where the enemy lacks sufficient firepower to shoot them down. Predators and Reapers are not designed to withstand antiaircraft defenses or air-to-air combat. In the current Russia-Ukraine war, Ukrainian troops have used UAVs to reconnoiter Russian military positions, guide artilleries, and attack military targets within Russian borders. Russian troops use UAVs to attack Ukrainian infrastructure extensively, cutting electricity and water supply in many Ukrainian cities. Crewless or unmanned surface vehicles (USVs) are vehicles that operate on the surface of the water without a crew, while crewless or unmanned underwater vehicles (UUV), sometimes known as underwater drones, can operate underwater without a human occupant (Gafurov and Klochkov 2015). Both USVs and UUVs may include two types: remotely operated vehicles, which a remote human operator controls, and autonomous vehicles, which operate independently of direct human input. Autonomous surface vehicles (ASVs) are autonomous USVs. Autonomous UUVs and ASVs would constitute a kind of robot. Those vehicles are valuable research tools in oceanography because they are far cheaper than the equivalent weather ships and research vessels but more capable than moored or drifting weather buoys. Liquid Robotics, an American company headquartered in Sunnyvale, California, has developed the USV Wave Glider, which harnesses wave energy for primary propulsion and uses solar cells to power its electronics.

The Dawn of the Robotic Age

181

It has months of marine persistence for both academic and naval applications (Zhang et al. 2015). The Israeli Rafael Advanced Defense Systems developed the Protector USV with offensive capability, the first operational combat USV in service. Powered seaborne targets are also one of the military applications for USVs (Yan et al. 2010). USVs and UUVs are currently used mainly for military and research purposes. They have not been proposed for transporting cargo and passengers or for productive purposes, probably because their commercial use would not be cost-effective yet. However, USVs and UUVs might be used for production and transporting cargo and passengers when technological development has made their commercial applications costeffective.

5. What is the Future of AI and Robots? We have discussed so far mainly AI systems and robots for performing existing human jobs or jobs that are too hazardous for humans, but AI researchers have been working on more challenging components of human intelligence: creativity, emotion, and social intelligence. Many researchers aim for the ultimate goal of general intelligence and expect that robots will soon be more intelligent than humans in terms of general intelligence. Many celebrities have started to warn about the dangers AI and robots will pose to humanity. However, we must point out that artificial general intelligence is not a precondition for entering the Robotic Age. To replace most human workers and enter the Robotic Age, AI that mimics or surpasses one or a few aspects of human intelligence necessary for doing those jobs is sufficient. AI that mimics or surpasses one or a few aspects of human intelligence is sometimes called the weak form of AI or weak AI. The Robotic Age discussed in this book is still robotics with a weak form of AI.

5.1. Creativity Creativity is a crucial element of human intelligence. With the recent success of ChatGPT in performing many tasks requested by people, many still argue that ChatGPT compiles ideas and knowledge found on the Internet without creativity. Therefore, the creation of creativity would represent a significant achievement. Research on creativity has been carried out from philosophical, psychological, and practical perspectives. The practical perspective is to create specific implementations which generate outputs that can be considered creative. Computational creativity aims to

182

Chapter 3

model, simulate or replicate creativity using computers. It is a sub-field of AI that addresses creativity both theoretically and practically. Artificial intuition and artificial thinking are also AI research branches closely related to creativity research (Jordanous 2012).

5.2. Social Intelligence The ability of human beings to interact with others is considered social intelligence; affective computing studies systems and devices that recognize, interpret, process, and simulate human affect and emotion. It also aims to develop such systems and devices. Rosalind Picard's pioneering paper on affective computing in 1995 started this branch of computer science; one motivation is to simulate empathy (Picard 1995). Affective computing is an interdisciplinary field where computer sciences, psychology, and cognitive science are the key players. A machine with social intelligence should be able to interpret the emotional state of humans correctly and adapt its behavior to them accordingly. Thus, social intelligence is to interpret human emotional states and give an appropriate response to those emotions.

5.3. General Intelligence The long-term goal of many AI researchers is artificial general intelligence, also known as strong AI, which combines all elements of human intelligence afore-mentioned and exceeds human abilities at most or all of them. Current AI research projects usually focus on simulating one or a few aspects of human intelligence, but the perfect simulation of these aspects may require general intelligence. For example, although machine translation software such as Google Translation and ChatGPT can produce fairly readable text, good translation by a machine requires that the machine understand the content (knowledge), follow the author's argument (reason), and reproduce the author’s intention (social intelligence). The currently successful machine translation usually depends on autoregressive language models based on statistical probability rather than understanding the two languages. The most challenging problems in AI research, such as understanding natural languages, are informally known as AI-complete or AI-hard. This implies that solving these computational problems is equivalent to producing a machine with general intelligence or strong AI. Some researchers believe that strong AI may also need artificial consciousness and other anthropomorphic features, and a few think that strong AI requires an artificial brain (Searle 1990, 1980).

The Dawn of the Robotic Age

183

6. Summary It has been a dream for humanity since ancient times to have machines work like humans. Automation grew from simple mechanical devices to complex electro-mechanical devices that automated many production processes before the advent of digital computers. Digital computers opened up new avenues for AI and robotics. The development of AI and robotics make it possible for machines to replace human mental power in production. The assembly line workers used little of their intelligence in performing their jobs, so they were the first to be replaced by industrial robots. AI systems that mimic only human mental power have been more successful, with Deep Blue beating World Chess Champion Garry Kasparov, IBM Watson defeating Brad Rutter and Ken Jennings in a Jeopardy quiz show, Alpha-Go defeating Go world champions Lee Sedol and Ke Jie, and ChatGPT and Google Bard performing more and more tasks for humans. Service and maintenance workers are more difficult to replace with robots because they face multiple non-standard challenges and must perform manual and mental tasks. In the coming Robotic Age, the development of AI and robotics will lead to multifunctional service and maintenance robots that can replace service and maintenance workers.

References Abbasi, Abid, Ubaid M Al-Saggaf, and Khalid Munawar. 2014. “Design of Unmanned Aerial Vehicle Autopilot based on Freescale Qorivva 32-bit Microcontroller.” Life Science Journal 11 (11):874-878. Abdi, Herve. 1994. “A neural network primer.” Journal of Biological Systems 2 (03):247-281. Agar, Jon. 2020. “What is science for? The Lighthill report on artificial intelligence reinterpreted.” The British Journal for the History of Science 53 (3):289-310. Aikenhead, Bruce A, Robert G Daniell, and Frederick M Davis. 1983. “Canadarm and the space shuttle.” Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films 1 (2):126-132. Albu-Schaffer, Alin, Oliver Eiberger, Markus Grebenstein, Sami Haddadin, Christian Ott, Thomas Wimbock, Sebastian Wolf, and Gerd Hirzinger. 2008. “Soft robotics.” IEEE Robotics & Automation Magazine 15 (3):20-30. Anderson, David P. 2009. “Biographies: Tom Kilburn: a pioneer of computer design.” IEEE Annals of the History of Computing 31 (2):8286.

184

Chapter 3

Andresen, Scott L. 2002. “John McCarthy: father of AI.” IEEE Intelligent Systems 17 (5):84-85. Apte, Pushkar, and George Scalise. 2009. “The recession's silver lining.” IEEE Spectrum 46 (10):46-51. Asada, Haruhiko, Takeo Kanade, and R Reddy. 1981. “Design Concept of Direct-Drive Manipulators Using Rare-Earth DC Torque Motors.” IJCAI. Asada, Haruhiko, Takeo Kanade, and Ichiro Takeyama. 1983. “Control of a direct-drive arm.” Journal of Dynamic Systems, Measurement, and Control 105 (3):136-142. Ashley, Steven. 1998. “Smart cars and automated highways.” Mechanical Engineering 120 (05):58-62. Aspray, William. 1997. “The Intel 4004 microprocessor: What constituted invention?” IEEE Annals of the History of Computing 19 (3):4-15. Bajracharya, Max, Mark W Maimone, and Daniel Helmick. 2008. “Autonomy for Mars rovers: Past, present, and future.” Computer 41 (12):44-50. Ball, Philip. 2012. “Computer science: Algorithmic rapture.” Nature 488 (7412):458-458. Ballard, Leslie Anne, Selma Sabanovic, Jasleen Kaur, and Stasa Milojevic. 2012. “George Charles Devol, jr.[history].” IEEE Robotics & Automation Magazine 19 (3):114-119. Barrett, David, Mark Grosenbaugh, and Michael Triantafyllou. 1996. “The optimal control of a flexible hull robotic undersea vehicle propelled by an oscillating foil.” In Proceedings of Symposium on Autonomous Underwater Vehicle Technology, 1-9. IEEE. Bartsch, Ron, James Coyne, and Katherine Gray. 2016. Drones in Society: Exploring the strange new world of unmanned aircraft. London and New York: Routledge. Bell, Genevieve. 2018. “Making life: a brief history of human-robot interaction.” Consumption Markets & Culture 21 (1):22-41. Bennett, Stuart. 1984. “Nicholas Minorsky and the automatic steering of ships.” IEEE Control Systems Magazine 4 (4):10-15. Bennett, Stuart. 1996. “A brief history of automatic control.” IEEE Control Systems Magazine 16 (3):17-25. Bennett, Stuart. 2001. “The past of PID controllers.” Annual Reviews in Control 25:43-53. Berryman, Sylvia. 2003. “Ancient automata and mechanical explanation.” Phronesis 48 (4):344-369. Bimbraw, Keshav. 2015. “Autonomous cars: Past, present and future a review of the developments in the last century, the present scenario and

The Dawn of the Robotic Age

185

the expected future of autonomous vehicle technology.” In 12th international conference on informatics in control, automation and robotics (ICINCO), 191-198. IEEE. Black, Harold S. 1934. “Stabilized feedback amplifiers.” Bell system technical journal 13 (1):1-18. Bladin, Peter F. 2006. “W. Grey Walter, pioneer in the electroencephalogram, robotics, cybernetics, artificial intelligence.” Journal of Clinical Neuroscience 13 (2):170-177. Bogue, Robert. 2011. “Robots in healthcare.” Industrial Robot: An International Journal 38 (3):218-223. Bogue, Robert. 2016. “Robots poised to revolutionise agriculture.” Industrial Robot: An International Journal 43 (5):450-456. Bogue, Robert. 2020. “Fruit picking robots: has their time come?” Industrial Robot: The International Journal of Robotics Research and Application. Bongard, Josh, Victor Zykov, and Hod Lipson. 2006. “Resilient machines through continuous self-modeling.” Science 314 (5802):1118-1121. Bringsjord, Selmer, and Naveen Sundar Govindarajulu. 2020. Artificial Intelligence. In The Stanford Encyclopedia of Philosophy edited by Edward N. Zalta. Stanford, CA: Metaphysics Research Lab. Brittain, James E. 2005. “Electrical engineering hall of fame: Lee de Forest.” Proceedings of the IEEE 93 (1):198-202. Brock, David, and David Laws. 2011. “The early history of microcircuitry: An overview.” IEEE Annals of the History of Computing 34 (1):7-19. Broelmann, Jobst. 2002. “Hermann Anschütz-Kaempfe-Richtungsweiser ohne Spuren.” Deutsches Schiffahrtsarchiv 25:41-55. Brown, Gordon S. 1981. “Harold Locke Hazen, 1901-1980.” Annals of the History of Computing 3 (1):4-12. Brown, Ronald C. 2016. “Made in China 2025: Implications of Robotization and Digitalization on MNC Labor Supply Chains and Workers' Labor Rights in China.” Tsinghua China Law Review 9:186210. Brügger, Niels. 2016. “Introduction: The Web’s first 25 years.” New Media & Society 18 (7):1059-1065. Bryant, Martin. 2011. “20 years ago today, the World Wide Web opened to the public.” The Next Web.com, accessed April 8, 2022. https://thenextweb.com/news/20-years-ago-today-the-world-wide-webopened-to-the-public. Burton, Chris, and Martin Cooper. 2018. “Manchester Baby’s 70th Anniversary.” ITNOW 60 (2):30-31.

186

Chapter 3

Butler, Deborah, Lewis Holloway, and Christopher Bear. 2012. “The impact of technological change in dairy farming: robotic milking systems and the changing role of the stockperson.” Journal of the Royal Agricultural Society of England 173 (622):1-6. Cable, Michael. 1999. “Mechanization of glass manufacture.” Journal of the American Ceramic Society 82 (5):1093-1112. Callas, John L, Matthew P Golombek, and Abigail A Fraeman. 2019. Mars Exploration Rover Opportunity end of mission report. Pasadena, CA: Jet Propulsion Laboratory, National Aeronautics and Space Carlson, Matt. 2015. “The robotic reporter: Automated journalism and the redefinition of labor, compositional forms, and journalistic authority.” Digital Journalism 3 (3):416-431. Carter, Rodger L. 2001. “Digicam history.” Digicamhistory.com, accessed April 6, 2022. http://www.digicamhistory.com/1988.html. Chan, Ching-Yao. 2017. “Advancements, prospects, and impacts of automated driving systems.” International Journal of Transportation Science and Technology 6 (3):208-216. Chao, HaiYang, YongCan Cao, and YangQuan Chen. 2010. “Autopilots for small unmanned aerial vehicles: a survey.” International Journal of Control, Automation and Systems 8 (1):36-44. Chaturvedi, Devendra K. 2008. “Soft computing.” Studies in Computational Intelligence 103:509-612. Cholodenko, Alan. 2007. “Speculations on the Animatic Automaton.” The Illusion of Life 2:486-528. Chouard, Tanguy. 2016. The Go Files: AI computer clinches victory against Go champion. Nature News. Choudhury, Nupur. 2014. “World wide web and its journey from web 1.0 to web 4.0.” International Journal of Computer Science and Information Technologies 5 (6):8096-8100. Clymer, A Ben. 1993. “The mechanical analog computers of Hannibal Ford and William Newell.” IEEE Annals of the History of Computing 15 (2):19-34. Coelingh, Erik, and Stefan Solyom. 2012. “All aboard the robotic road train.” IEEE Spectrum 49 (11):34-39. Contrada, Norma. 1995. “Golem and robot: a search for connections.” Journal of the Fantastic in the Arts 7 (2/3 (26/27):244-254. Cooke-Yarborough, EH. 1998. “Some early transistor applications in the UK.” Engineering Science & Education Journal 7 (3):100-106. Copeland, B Jack. 2000. “The modern history of computing.” In Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. Stanford, CA:

The Dawn of the Robotic Age

187

The Metaphysics Research Lab, Philosophy Department, Stanford University. Crew, Bec. 2015. “The World's First Website Just Turned 25 Years Old.” Sciencealert.com, accessed April 8, 2022. https://www.sciencealert.com/the-world-s-first-website-just-turned-25years-old. Dale, Robert. 2019. “Law and word order: NLP in legal tech.” Natural Language Engineering 25 (1):211-217. Davis, Randall, Howard Shrobe, and Peter Szolovits. 1993. “What is a knowledge representation?” AI Magazine 14 (1):17-17. Diftler, Myron A, JS Mehling, Muhammad E Abdallah, Nicolaus A Radford, Lyndon B Bridgwater, Adam M Sanders, Roger Scott Askew, D Marty Linn, John D Yamokoski, and FA Permenter. 2011. “Robonaut 2-the first humanoid robot in space.” In 2011 IEEE International Conference on Robotics and Automation, 2178-2183. IEEE. Dongarra, Jack J, and Alan Hinds. 1986. “Comparison of the CRAY X-MP4, Fujitsu VP-200, and Hitachi S-810/20.” Simulation 47 (3):93-107. Dowling, Michael, and Brian Lucey. 2023. “ChatGPT for (finance) research: The Bananarama conjecture.” Finance Research Letters:103662. Drill, Sabrina L. 2012. “Mobile applications for extension.” Journal of Extension 50 (5):5TOT1. Durand-Richard, Marie-José. 2010. “Planimeters and integraphs in the 19th Century. Before the differential analyser.” Nuncius: Journal of the History of Science 25 (1):101-124. Erdo÷du, Mehmet. 2021. “Development of Logistics Management and Relationship with Industry.” International Journal on Engineering, Science and Technology 3 (2):91-112. Erickson, Kelvin T. 1996. “Programmable logic controllers.” IEEE Potentials 15 (1):14-17. Farrell, Jonathan P. 2015. Artificial Composers: Tools of the Modern Musician or Affront to Human Creativity? Inquiries Journal 7 (03). Feigenbaum, Edward A. 1994. “roots of knowledge systems and expert system applications.” Artificial Intelligence in Perspective 59:233-240. Fifth Wave Manufacturing. 2022. “Robots and automation come to their senses.” Fifth Wave Manufacturing, Last Modified April 6, 2022, accessed November 2, 2022. https://fifthwavemfg.com/robots-andautomation-come-to-their-senses/. Fiorini, Paolo. 2018. “History of Robots and Robotic Surgery.” In The SAGES Atlas of Robotic Surgery, edited by Yuman Fong, Yanghee Woo, Woo Jin Hyung, Clayton Lau and Vivian E. Strong, 3-14. Cham, Switzerland: Springer.

188

Chapter 3

Fogel, David B. 2000. “What is evolutionary computation?” IEEE Spectrum 37 (2):26-32. Ford, Martin. 2015. Rise of the Robots: Technology and the Threat of a Jobless Future. New York: Basic Books. Forrester, Jay W, and Robert R Everett. 1990. “The Whirlwind computer project.” IEEE Transactions on Aerospace and Electronic Systems 26 (5):903-910. Frenkel, Karen A. 1985. “Report on the microelectronics and computer technology corporation conference.” Communications of the ACM 28 (8):808-813. Frumer, Yulia. 2020. “The short, strange life of the first friendly robot: Japan's Gakutensoku was a giant pneumatic automaton that toured through Asia—until it mysteriously disappeared.” IEEE Spectrum 57 (6):42-48. Fryer, David M, and John C Marshall. 1979. “The motives of Jacques de Vaucanson.” Technology and Culture 20 (2):257-269. Fujita, Masahiro. 2001. “AIBO: Toward the era of digital creatures.” The International Journal of Robotics Research 20 (10):781-794. Gade, Kenneth. 2016. “The seven ways to find heading.” The Journal of Navigation 69 (5):955-970. Gafurov, Salimzhan A, and Evgeniy V Klochkov. 2015. “Autonomous unmanned underwater vehicles development tendencies.” Procedia Engineering 106:141-148. Ganapati, Priya. 2010. “May 7, 1952: The Integrated Circuit... What a Concept!”. Wired, Last Modified May 7, 2010, accessed March 16, 2022. https://www.wired.com/2010/05/0507integrated-circuit-conceptpublished/. Gao, Feng, Shuo Li, Yue Gao, Chenkun Qi, Qiyan Tian, and Guang-Zhong Yang. 2022. “Robots at the Beijing 2022 Winter Olympics.” Science Robotics 7 (65):eabq0785. Gasparetto, A, and L Scalera. 2019. “A brief history of industrial robotics in the 20th century.” Advances in Historical Studies 8 (1):24-35. Gerszten, Peter C, Cihat Ozhasoglu, Steven A Burton, William J Vogel, Barbara A Atkins, Shalom Kalnicki, and William C Welch. 2004. “CyberKnife frameless stereotactic radiosurgery for spinal lesions: clinical experience in 125 cases.” Neurosurgery 55 (1):89-99. Gibbs, Graham, and Savi Sachdev. 2002. “Canada and the international space station program: overview and status.” Acta Astronautica 51 (19):591-600. Gleason, Shannon C. 2014. “Don’t fear the cyborg: Toward embracing posthuman and feminist cyborg discourses in teacher education and

The Dawn of the Robotic Age

189

educational technology research.” Canadian Journal of Science, Mathematics and Technology Education 14 (2):120-134. Gold, Jon. 2012. “The smartphone is 20 years old, believe it or not.” Network World, accessed April 6, 2022. https://www.networkworld.com/article/2161682/the-smartphone-is-20years-old--believe-it-or-not.html. Good, Irving John. 1979. “Early work on computers at Bletchley.” Annals of the History of Computing 1 (1):38-48. Gordijn, Bert, and Henk ten Have. 2023. “ChatGPT: evolution or revolution?” Medicine, Health Care and Philosophy:1-2. Gottfredson, Linda S. 1997. “Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography.” 24 (1):13-23. Gray, Edwyn. 2004. Nineteenth-century Torpedoes and their Inventors. Annapolis, Maryland: Naval Institute Press. Guarnieri, Massimo. 2010. “The roots of automation before mechatronics [historical].” IEEE Industrial Electronics Magazine 4 (2):42-43. Guarnieri, Massimo. 2017. “Trailblazers in Electromechanical Computing [Historical].” IEEE Industrial Electronics Magazine 11 (2):58-62. Haenlein, Michael, and Andreas Kaplan. 2019. “A brief history of artificial intelligence: On the past, present, and future of artificial intelligence.” California Management Review 61 (4):5-14. Hallion, Richard P. 1980. “William Wyatt Davenport, Gyro! The Life and Times of Lawrence Sperry (Book Review).” Technology and Culture 21 (1):132. Hammond, John H, and ES Purington. 1957. “A history of some foundations of modern radio-electronic technology.” Proceedings of the IRE 45 (9):1191-1208. Hartley, David. 2011. “EDSAC at 60—A Celebration of 60 Years Since the First Program Ran on the EDSAC at Cambridge.” The Computer Journal 54 (1):136-138. Hendry, John. 1987. “The teashop computer manufacturer: J. Lyons, LEO and the potential and limits of high-tech diversification.” Business History 29 (1):73-102. Hill, David J. 2012. “Automation Comes To The Coffeehouse With Robotic Baristas.” Singularity Hub, Last Modified May 9, 2012, accessed January 14, 2017. https://singularityhub.com/2012/05/09/automationcomes-to-the-coffeehouse-with-robotic-baristas/. Hinchey, Michael G, Roy Sterritt, and Chris Rouff. 2007. “Swarms and swarm intelligence.” Computer 40 (4):111-113.

190

Chapter 3

Hinson, John M, and JER Staddon. 1983. “Matching, maximizing, and hillဨ climbing.” Journal of the Experimental Analysis of Behavior 40 (3):321331. Hirose, Masato, and Kenichi Ogawa. 2007. “Honda humanoid robots development.” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 365 (1850):11-19. Hoggett, Reuben. 2009. “1920 – “The Kaiser” Mechanical Man – Capt. Alban J. Roberts.” Cyberneticzoo, accessed May 5, 2022. http://cyberneticzoo.com/robots/1920-the-kaiser-capt-alban-j-robertsnew-zealandaustraliau-k/. Honda. 2005. “Honda Debuts New ASIMO.” Honda Global, accessed May 6, 2022. https://global.honda/newsroom/news/2005/c051213-asimo-eng.html. Horowitz, Roberto, and Pravin Varaiya. 2000. “Control design of an automated highway system.” Proceedings of the IEEE 88 (7):913-925. Hosek, Jiri, Pavel Masek, Sergey Andreev, Olga Galinina, Aleksandr Ometov, Franz Kropfl, Werner Wiedermann, and Yevgeni Koucheryavy. 2017. “A SyMPHOnY of integrated IoT businesses: Closing the gap between availability and adoption.” IEEE Communications Magazine 55 (12):156-164. Hoshimov, A, A Rustamov, and J Rozzokov. 2018. “Industrial Conveyors’ Taxonomy and Its Applications.” Acta of Turin Polytechnic University in Tashkent 8 (3):60-62. Hsiao, Sissie. 2023. “What’s ahead for Bard: More global, more visual, more integrated.” blog.google, May 26, 2023. https://blog.google/technology/ai/google-bard-updates-io-2023/. Hunter, Brooke. 2005. “Wheat, War, and the American Economy during the Age of Revolution.” The William and Mary Quarterly 62 (3):505-526. International Telecommunication Union. 2019. “Constitution of the International Communication Union.” In. Geneva: International Telecommunication Union. https://www.itu.int/en/council/Documents/basic-texts/ConstitutionE.pdf. iRobot. 2023. “Corporate Profile.” PR Newswire, accessed March 30, 2023. Jensen, Finn V. 2009. “Bayesian networks.” Wiley Interdisciplinary Reviews: Computational Statistics 1 (3):307-315. Jones, Joseph L. 2006. “Robots at the tipping point: the road to iRobot Roomba.” IEEE Robotics & Automation Magazine 13 (1):76-78. Jordan, Michael I, and Tom M Mitchell. 2015. “Machine learning: Trends, perspectives, and prospects.” Science 349 (6245):255-260.

The Dawn of the Robotic Age

191

Jordanous, Anna. 2012. “A standardised procedure for evaluating creative systems: Computational creativity evaluation based on what it is to be creative.” Cognitive Computation 4 (3):246-279. Jumper, John, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, and Anna Potapenko. 2021. “Highly accurate protein structure prediction with AlphaFold.” Nature 596 (7873):583-589. Kaneko, Kenji, Fumio Kanehiro, Mitsuharu Morisawa, Kanako Miura, Shin'ichiro Nakaoka, and Shuuji Kajita. 2009. “Cybernetic human HRP4C.” In 2009 9th IEEE-RAS International Conference on Humanoid Robots, 7-14. IEEE. Kato, Ichiro, Sadamu Ohteru, Katsuhiko Shirai, Toshiaki Matsushima, Seinosuke Narita, Shigeki Sugano, Tetsunori Kobayashi, and Eizo Fujisawa. 1987. “The robot musician ‘wabot-2’(waseda robot-2).” Robotics 3 (2):143-155. Kilby, Jack S. 1998. “6600 LBJ Freeway, Dallas.” Silicon Materials Science and Technology: Proceedings of the Eighth International Symposium on Silicon Materials Science and Technology 98 (1):342-349. Kilgour, Frederick G. 1964. “Technological Innovation in the United States.” Cahiers d'Histoire Mondiale. Journal of World History. Cuadernos de Historia Mundial 8 (1):742. Kim, Won. 2009. “Cloud computing: Today and tomorrow.” Journal of Object Technology 8 (1):65-72. Kirmani, Ahmad R. 2022. “Artificial Intelligence-Enabled Science Poetry.” ACS Energy Letters 8:574-576. Kitamori, Toshiyuki, N Suda, E Shimemura, M Masubuchi, H Kimura, and Y Yoshitani. 1984. “Control engineering in Japan: Past and present.” IEEE Control Systems Magazine 4 (4):4-10. Kline, Ronald. 1993. “Harold Black and the negative-feedback amplifier.” IEEE Control Systems Magazine 13 (4):82-85. Kudo, Shuhei, Keigo Nitadori, Takuya Ina, and Toshiyuki Imamura. 2020. “Implementation and numerical techniques for one EFlop/s HPL-AI benchmark on Fugaku.” In 2020 IEEE/ACM 11th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems (ScalA), 6976. IEEE. Kuipers, Benjamin, Edward A Feigenbaum, Peter E Hart, and Nils J Nilsson. 2017. “Shakey: from conception to history.” AI Magazine 38 (1):88-103. Kuipers, Martijn, and Ramjee Prasad. 2022. “Journey of Artificial Intelligence.” Wireless Personal Communications 123:3275-3290. Kuroki, Yoshihiro, Bill Blank, Tatsuo Mikami, Patrick Mayeux, Atsushi Miyamoto, Robert Playter, Ken’ichiro Nagasaka, Marc Raibert,

192

Chapter 3

Masakuni Nagano, and Jin’ichi Yamaguchi. 2003. “Motion creating system for a small biped entertainment robot.” In Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003)(Cat. No. 03CH37453), 1394-1399. IEEE. Letaief, Khaled B, Wei Chen, Yuanming Shi, Jun Zhang, and Ying-Jun Angela Zhang. 2019. “The roadmap to 6G: AI empowered wireless networks.” IEEE Communications Magazine 57 (8):84-90. Licklider, JCR. 1963. Memorandum for Members and Affiliates of the Intergalactic Computer Network. Washington DC: Advanced Research Projects Agency. Liikanen, Jukka, Paul Stoneman, and Otto Toivanen. 2004. “Intergenerational effects in the diffusion of new technology: the case of mobile phones.” International Journal of Industrial Organization 22 (8-9):1137-1154. Lin, Ting-Yu, Kun-Ru Wu, You-Shuo Chen, Wei-Hau Huang, and Yi-Tuo Chen. 2021. “Takeout service automation with trained robots in the pandemic-transformed catering business.” IEEE Robotics and Automation Letters 6 (2):903-910. Lipartito, Kenneth. 1994. “When women were switches: Technology, work, and gender in the telephone industry, 1890-1920.” American Historical Review 99 (4):1075-1111. Liu, Weibo, Zidong Wang, Xiaohui Liu, Nianyin Zeng, Yurong Liu, and Fuad E Alsaadi. 2017. “A survey of deep neural network architectures and their applications.” Neurocomputing 234:11-26. Livinginternet. 2000. “ARPANET – The First Internet.” Livinginternet.com, accessed April 8, 2022. https://www.livinginternet.com/i/ii_arpanet.htm. Lyons, William M. 1995. “Policy innovations of the US intermodal surface transportation efficiency act and clean air act amendments.” Transportation 22 (3):217-240. Makridakis, Spyros. 2017. “The forthcoming Artificial Intelligence (AI) revolution: Its impact on society and firms.” Futures 90:46-60. Markin, Rodney S. 1993. “Implementing automation in a modern clinical laboratory.” Chemometrics and Intelligent Laboratory Systems 21 (23):169-179. Matyas, J. 1965. “Random optimization.” Automation and Remote control 26 (2):246-253. Maxwell, James Clerk. 1868. “I. On governors.” Proceedings of the Royal Society of London (16):270-283. Mayor, Adrienne. 2018. “Gods and robots.” In Gods and Robots. Princeton: Princeton University Press.

The Dawn of the Robotic Age

193

Mayr, Otto. 1970. “The origins of feedback control.” Scientific American 223 (4):110-119. McAuley, Alexander, Bonnie Stewart, George Siemens, and Dave Cormier. 2010. The MOOC model for digital practice. SSHRC Knowledge Synthesis Grant on the Digital Economy. McCarthy, John. 1981. “Epistemological problems of artificial intelligence.” In Readings in artificial intelligence, edited by Bonnie Lynn Webber and Nils J. Nilsson, 459-465. Los Altos, CA: Elsevier. McCarthy, John, and P.J. Hayes. 1969. “Some Philosophical Problems from the Standpoint of Artificial Intelligence.” In Machine Intelligence 4, edited by D. Michie, 463-502. New York, NY: Elsevier. McDowall, Mike. 2015. “How a simple 'hello' became the first message sent via the Internet.” PBS NewsHour, accessed April 7, 2022. https://www.pbs.org/newshour/science/internet-got-started-simplehello McKenzie, John A. 1999. TX-0 computer history. In RLE Technical Report. Cambridge, Massachusetts: Massachusetts Institute of Technology. Milutinoviü, Dragan, and Veljko Potkonjak. 1990. “A new concept of the SCARA robot.” Robotics and Computer-Integrated Manufacturing 7 (34):337-343. Minorsky, Nicolas. 1922. “Directional stability of automatically steered bodies.” Journal of the American Society for Naval Engineers 34 (2):280-309. Minsky, Marvin L, and Seymour A Papert. 1988. Perceptrons: expanded edition. Cambridge, MA: MIT Press. Moore, Gordon E. 1965. “Cramming more components onto integrated circuits.” Electronics 38 (8):114-117. Moran, Michael E. 2006. “The da Vinci robot.” Journal of Endourology 20 (12):986-990. Moran, Michael E. 2007a. “Evolution of robotic arms.” Journal of Robotic Surgery 1 (2):103-111. Moran, Michael E. 2007b. “Jacques de Vaucanson: The father of simulation.” Journal of Endourology 21 (7):679-683. Moravec, Hans P. 1983. “The Stanford cart and the CMU rover.” Proceedings of the IEEE 71 (7):872-884. Mowforth, Peter, and Ivan Bratko. 1987. “Ai and robotics; flexibility and integration.” Robotica 5 (2):93-98. Muggleton, Stephen. 1991. “Inductive logic programming.” New Generation Computing 8 (4):295-318.

194

Chapter 3

Murphy, Tom. 2013. “40 Years After the First Cell Phone Call: Who Is Inventing Tomorrow? s Future?” IEEE Consumer Electronics Magazine 2 (4):44-46. Nanda, Santosh Kumar, Ashok Kumar Dash, Sandigdha Acharya, and Abikshyana Moharana. 2010. “Application of robotics in mining industry: A critical review.” Indian Mining and Engineering Journal 10:108-112. Naude, SM. 1956. “New horizons for science and industry.” South African Journal of Science 53 (1):2-9. Newborn, Monty. 2011. “2006: Deep Fritz Clobbers Kramnik, 4–2.” In Beyond Deep Blue, 141-148. Berlin and Heidelberg: Springer. Newborn, Monty. 2012. Kasparov versus Deep Blue: Computer chess comes of age. Berlin and Heidelberg: Springer Science & Business Media. Newell, Allen. 1980. “Physical symbol systems.” Cognitive Science 4 (2):135-183. Newell, Allen, and Herbert A Simon. 1961. “Computer Simulation of Human Thinking: A theory of problem solving expressed as a computer program permits simulation of thinking processes.” Science 134 (3495):2011-2017. O'Neill, Judy E. 1995. “The Role of ARPA in the Development of the ARPANET, 1961-1972.” IEEE Annals of the History of Computing 17 (4):76-81. O’Regan, Gerard. 2021. A Brief History of Computing. Cham: Springer. Oakley, Brian, and Kenneth Owen. 1990. Alvey: Britain's strategic computing initiative. Cambridge, MA: MIT Press. Olsen, Kevin. 2012. “The first 110 years of laboratory automation: technologies, applications, and the creative scientist.” Journal of Laboratory Automation 17 (6):469-480. Open AI. 2022a. “Dall-E2.” Open AI, accessed May 28, 2023. https://openai.com/product/dall-e-2. Open AI. 2022b. “Introducing ChatGPT.” January 30, 2023. openai.com/blog/chatgpt. Open AI. 2023. “GPT-4.” Open AI, accessed April 2, 2023. openai.com/research/gpt-4. Ozguner, Umit, Christoph Stiller, and Keith Redmill. 2007. “Systems for safety and autonomous behavior in cars: The DARPA Grand Challenge experience.” Proceedings of the IEEE 95 (2):397-412. Palletizers.org. No date. “Palletizers.” Palletizers.org accessed March 8. https://www.palletizers.org/.

The Dawn of the Robotic Age

195

Papadimitriou, Christos H, Prabhakar Raghavan, Hisao Tamaki, and Santosh Vempala. 2000. “Latent semantic indexing: A probabilistic analysis.” Journal of Computer and System Sciences 61 (2):217-235. Paredes, Dusan, and David Fleming-Muñoz. 2021. “Automation and robotics in mining: Jobs, income and inequality implications.” The Extractive Industries and Society 8 (1):189-193. Paul, Prantosh, Ripu Ranjan Sinha Ripu Ranjan Sinha, PS Aithal, Ricardo Saavedra M, Prof Sir Bashiru Aremu, and Shivlal Mewada. 2020. “Agricultural Robots: The Applications of Robotics in Smart Agriculture—Towards More Advanced Agro Informatics Practice.” Asian Review of Mechanical Engineering 9 (1):38-44. Pavlik, John V. 2023. “Collaborating With ChatGPT: Considering the Implications of Generative Artificial Intelligence for Journalism and Media Education.” Journalism & Mass Communication Educator: 10776958221149577. Pérez Yuste, Antonio. 2008. “Early Developments of Wireless Remote Control: the Telekino of Torres Quevedo.” Proceedings of the IEEE 96 (1):186-190. Petree, J Foster. 1945. “The Seventeenth Thomas Lowe Gray Lecture: Mechanical Engineering in the Shipyard.” Proceedings of the Institution of Mechanical Engineers 152 (1):362-369. Pfaff, C Anthony. 2019. “The Ethics of Acquiring Disruptive Technologies.” Prism 8 (3):128-145. Phipps, Charles. 2011. “The early history of ICs at Texas Instruments: A personal view.” IEEE Annals of the History of Computing 34 (1):37-47. Picard, Rosalind W. 1995. “Computer learning of subjectivity.” ACM Computing Surveys (CSUR) 27 (4):621-623. Pichai, Sundar. 2023. “An important next step on our AI journey.” blog.google, May 26, 2023. https://blog.google/technology/ai/bardgoogle-ai-search-updates/. Prasad, Ray. 2013. Surface mount technology: principles and practice. Dordrecht: Springer Science & Business Media. Prisacariu, Vasile. 2017. “The history and the evolution of UAVs from the beginning till the 70s.” Journal of Defense Resources Management (JoDRM) 8 (1):181-189. Rahman, Tariq, Whitney Sample, Rahamim Seliktar, Michael Alexander, and Mena Scavina. 2000. “A body-powered functional upper limb orthosis.” Journal of Rehabilitation Research and Development 37 (6):675-680. Rajaraman, Vaidyeswaran. 2014. “JohnMcCarthy—Father of artificial intelligence.” Resonance 19 (3):198-207.

196

Chapter 3

Reay, David Anthony. 2014. The history of man-powered flight: Elsevier. Reiter, Raymond. 1980. “A logic for default reasoning.” Artificial Intelligence 13 (1-2):81-132. Reportlinker. 2022. “Home Automation Market by Connection Type, by Application, by Management, and by System Type—Global Opportunity Analysis and Industry Forecast 2022-2030.” Reportlinker, accessed November 7, 2022. https://www.reportlinker.com/p06272303/Home-Automation-Marketby-Connection-Type-by-Application-by-Management-and-by-SystemType-Global-Opportunity-Analysis-and-IndustryForecast.html?utm_source=GNW. Ribstein, Larry E. 2010. “The death of big law.” Wisconsin Law Review (3):749. Roback, Abraham Aaron. 1922. “Intelligence and intellect.” The Journal of Philosophy 19 (12):325-330. Roberts, GN. 2008. “Trends in marine control systems.” Annual Reviews in Control 32 (2):263-269. Robinson, Hayley, Bruce MacDonald, and Elizabeth Broadbent. 2014. “The role of healthcare robots for older people at home: A review.” International Journal of Social Robotics 6 (4):575-591. Robinson, Tim. 2005. “The Meccano set computers: a history of differential analyzers made from children's toys.” IEEE Control Systems Magazine 25 (3):74-83. Rogers, W Pratt, M Mustafa Kahraman, Frank A Drews, Kody Powell, Joel M Haight, Yaxue Wang, Kritika Baxla, and Mohit Sobalkar. 2019. “Automation in the mining industry: Review of technology, systems, human factors, and political risk.” Mining, Metallurgy & Exploration 36 (4):607-631. Rojas, Raúl. 2020. “The computer programs of Charles Babbage.” IEEE Annals of the History of Computing 43 (1):6-18. Roland, Alex, Philip Shiman, and William Aspray. 2002. Strategic computing: DARPA and the quest for machine intelligence, 1983-1993. Cambridge, MA: MIT Press. Roose, Kevin. 2022. “An AI-Generated Picture Won an Art Prize. Artists Aren’t Happy.” The New York Times, September 2, 2022. https://www.nytimes.com/2022/09/02/technology/ai-artificialintelligence-. Rope, Crispin. 2007. “ENIAC as a stored-program computer: a new look at the old records.” IEEE Annals of the History of Computing 29 (4):82-87. Rosen, Saul. 1969. “Electronic computers: A historical survey.” ACM Computing Surveys (CSUR) 1 (1):7-36.

The Dawn of the Robotic Age

197

Roshidul, H, ARBM Shariff, BS Blackmore, IB Aris, ARB Ramli, and J Hossen. 2010. “High Adoption of Behavior Based Robotics IN THE AUTONOMOUS MACHINES.” Journal of Information Systems Technology & Planning 3 (6):30-41. Roth, Bernard, Jahangir Rastegar, and Victor Scheinman. 1974. “On the design of computer controlled manipulators.” In On Theory and Practice of Robots and Manipulators, edited by Paolo Serafini, Elisabeth Guazzelli, Bernhard Schrefler and Friedrich Pfeiffer Franz G. Rammerstorfer, 93-113. Vienna: Springer. Roth, Bernie. 2016. “The Unexpected Passing of a Pioneering Robot Designer [In Memoriam].” IEEE Robotics & Automation Magazine 23 (4):200-201. Routh, Edward John. 1877. A Treatise on the Stability of a Given State of Motion: Particularly Steady Motion. Being the Essay to which the Adams Prize was Adjudged in 1877, at the University of Cambridge. New York: Macmillan and Company. Rubin, Steven M, and Raj Reddy. 1977. “The LOCUS Model of Search and its Use in Image Interpretation.” International Joint Conference on Artificial Intelligence 2:590-595. Rutenbar, Rob A. 1989. “Simulated annealing algorithms: An overview.” IEEE Circuits and Devices Magazine 5 (1):19-26. Saba, Djamel, Youcef Sahli, Rachid Maouedj, Abdelkader Hadidi, and Miloud Ben Medjahed. 2021. “Towards artificial intelligence: concepts, applications, and innovations.” In Enabling AI Applications in Data Science, edited by Aboul-Ella Hassanien, Mohamed Hamed N. Taha and Nour Eldeen M. Khalifa, 103-146. Cham: Springer. Sadiku, Matthew NO, Sarhan M Musa, and Omonowo D Momoh. 2014. “Cloud computing: opportunities and challenges.” IEEE Potentials 33 (1):34-36. Sakamotosupa, Rumi, and Matthew Allensupb. 2011. “There's something fishy about that sushi: how Japan interprets the global sushi boom.” Japan Forum 23 (1):99-121. Salazar-Palma, Magdalena, Alejandro Garcia-Lamperez, Alejandro GarciaLampérez, Tapan K Sarkar, and Dipak L Sengupta. 2011. “The father of radio: A brief chronology of the origin and development of wireless communications.” IEEE Antennas and Propagation Magazine 53 (6):83-114. Salimbeni, Renzo. 2016. Skillman. Sector Skills Alliance for Advanced Manufacturing in the Transport Sector. Rome: IFAC-CNR. Sanderson, Katharine. 2010. “Mars rover spirit (2003-10).” Nature 463 (7281):600.

198

Chapter 3

Sands, Timothy. 2020. “Development of deterministic artificial intelligence for unmanned underwater vehicles (UUV).” Journal of Marine Science and Engineering 8 (8):578. Sasaki, Masahide, Takeshi Kageoka, Katsumi Ogura, Hiromi Kataoka, Tadashi Ueta, and Shigeyoshi Sugihara. 1998. “Total laboratory automation in Japan: Past, present and the future.” Clinica Chimica Acta 278 (2):217-227. Schaeffer, Jonathan, Robert Lake, Paul Lu, and Martin Bryant. 1996. “Chinook the world man-machine checkers champion.” AI Magazine 17 (1):21-21. Schmidt, Michael, and Hod Lipson. 2009. “Distilling free-form natural laws from experimental data.” Science 324 (5923):81-85. Searle, John R. 1980. “Minds, brains, and programs.” Behavioral and Brain Sciences 3 (3):417-424. Searle, John R. 1990. “Is the brain's mind a computer program?” Scientific American 262 (1):25-31. Shannon, Claude E. 1938. “A symbolic analysis of relay and switching circuits.” Electrical Engineering 57 (12):713-723. Shannon, Claude Elwood. 1948. “A mathematical theory of communication.” The Bell System Technical Journal 27 (3):379-423. Sharkey, Noel, and Amanda Sharkey. 2006. “Artificial intelligence and natural magic.” Artificial Intelligence Review 25 (1):9-19. Silver, David, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai, and Adrian Bolton. 2017. “Mastering the game of go without human knowledge.” Nature 550 (7676):354-359. Simon, Herbert A. 1981. “Studying Human Intelligence by Creating Artificial Intelligence: When considered as a physical symbol system, the human brain can be fruitfully studied by computer simulation of its processes.” American Scientist 69 (3):300-309. Simon, Herbert A, and Allen Newell. 1971. “Human problem solving: The state of the theory in 1970.” American Psychologist 26 (2):145. Simon, Herbert A, and Allen Newell. 1976. “Computer science as empirical inquiry: symbols and search.” Communications of the ACM 19 (3):11126. Simpson, Ronnie. 2016. “Mobile and tablet internet usage exceeds desktop for first time worldwide.” StatCounter, accessed April 8, 2022. https://gs.statcounter.com/press/mobile-and-tablet-internet-usageexceeds-desktop-for-first-time-worldwide. Singer, Peter W. 2009. “Military robots and the laws of war.” The New Atlantis (23):25-45.

The Dawn of the Robotic Age

199

Singh, Balkeshwar, N Sellappan, and P Kumaradhas. 2013. “Evolution of industrial robots and their applications.” International Journal of Emerging Technology and Advanced Engineering 3 (5):763-768. Srinivasa, Siddhartha S, Dmitry Berenson, Maya Cakmak, Alvaro Collet, Mehmet R Dogar, Anca D Dragan, Ross A Knepper, Tim Niemueller, Kyle Strabala, and Mike Vande Weghe. 2012. “Herb 2.0: Lessons learned from developing a mobile manipulator for the home.” Proceedings of the IEEE 100 (8):2410-2428. Stern, Nancy. 1979. “The BINAC: a case study in the history of technology.” Annals of the History of Computing 1 (1):9-20. Stokel-Walker, Chris. 2022. AI bot ChatGPT writes smart essays-should academics worry? Nature. https://doi.org/10.1038/d41586-022-04397-7. Stout, Thomas M, and Theodore J Williams. 1995. “Pioneering work in the field of computer process control.” IEEE Annals of the History of Computing 17 (1):6-18. Strawn, George. 2014. “Masterminds of the world wide web.” IT Professional 16 (04):58-59. Strawn, George, and Candace Strawn. 2015. “The Father of Supercomputing: Seymour Cray.” IT Professional 17 (2):58-60. Stuart, Russell, and Norvig Peter. 2003. Artificial intelligence: a modern approach. Upper Saddle River, NJ: Prentice Hall Sviokla, John J. 1990. “An examination of the impact of expert systems on the firm: the case of XCON.” MIS Quarterly:127-140. Swade, Doron. 1991. “'It will not slice a pineapple'-Charles Babbage and the first computer.” IEE Review 37 (6):217-220. Talaviya, Tanha, Dhara Shah, Nivedita Patel, Hiteshri Yagnik, and Manan Shah. 2020. “Implementation of artificial intelligence in agriculture for optimisation of irrigation and application of pesticides and herbicides.” Artificial Intelligence in Agriculture 4:58-73. Talpes, Emil, Debjit Das Sarma, Ganesh Venkataramanan, Peter Bannon, Bill McGee, Benjamin Floering, Ankit Jalote, Christopher Hsiong, Sahil Arora, and Atchyuth Gorti. 2020. “Compute solution for Tesla's full self-driving computer.” IEEE Micro 40 (2):25-35. Tedeschi, Franco, and Giuseppe Carbone. 2014. “Design issues for hexapod walking robots.” Robotics 3 (2):181-206. Tedre, Matti, and Erkki Sutinen. 2009. “Crossing the Newton-Maxwell gap: Convergences and contingencies.” Spontaneous Generations: A Journal for the History and Philosophy of Science 3 (1):195-212. Thielscher, Michael. 2001. “The qualification problem: A solution to the problem of anomalous models.” Artificial Intelligence 131 (1-2):1-37.

200

Chapter 3

Thorp, H Holden. 2023. ChatGPT is fun, but not an author. American Association for the Advancement of Science. Tilden, Mark W. 2001. “The simply-does-it robotic approach.” Astronomy Now 15 (6):55-57. Torii, Toru. 2000. “Research in autonomous agriculture vehicles in Japan.” Computers and Electronics in Agriculture 25 (1-2):133-153. Trainer, Matthew. 2004. “The patents of William Thomson (Lord Kelvin).” World Patent Information 26 (4):311-317. Trainer, Matthew. 2008. “Albert Einstein’s expert opinions on the Sperry vs. Anschütz gyrocompass patent dispute.” World Patent Information 30 (4):320-325. Tuomi, Ilkka. 2018. “The impact of artificial intelligence on learning, teaching, and education.” Luxembourg: Publications Office of the European Union. Turing, Alan Mathison. 1936. “On computable numbers, with an application to the Entscheidungsproblem.” J. of Math 58 (345-363):5. Tweedale, Geoffrey. 1992. “Marketing in the second industrial revolution: A case study of the Ferranti Computer Group, 1949–63.” Business History 34 (1):96-127. Tybjerg, Karin. 2003. “Wonder-making and philosophical wonder in Hero of Alexandria.” Studies in History and Philosophy of Science Part A 34 (3):443-466. van Emden, Maarten. 2019. “Reflecting Back on the Lighthill Affair.” IEEE Annals of the History of Computing 41 (4):119-123. Voth, Danna. 2004. “A new generation of military robots.” IEEE Intelligent Systems 19 (4):2-3. Watanabe, Kimitaka, and Yosuke Yoneda. 2009. “The world's smallest biped humanoid robot “i-Sobot”.” In 2009 IEEE Workshop on Advanced Robotics and its Social Impacts, 51-53. IEEE. Weik, Martin H. 1955. A survey of domestic electronic digital computing systems. In Ballistic research laboratories report Aberdeen Proving Ground, Maryland: Ballistic research laboratories. Weiss, Eric. 1996. “Konrad Zuse, 1910-1995.” IEEE Annals of the History of Computing 18 (02):3-5. Weiss, Peter. 2005. “In its own image: Simple robot replicates itself block by block.” Science News 167 (20):310-310. White, G. 1988. “Epicyclic gears applied to early steam engines.” Mechanism and Machine Theory 23 (1):25-37. Whitley, Darrell, Soraya Rana, John Dzubera, and Keith E Mathias. 1996. “Evaluating evolutionary algorithms.” Artificial Intelligence 85 (12):245-276.

The Dawn of the Robotic Age

201

Williams, Theodore J. 2009. “Advances in industrial automation: Historical perspectives.” In Springer Handbook of Automation, edited by Shimon Y. Nof, 5-11. Heidelberg: Springer. Wilson, JR. 2002. “Mini Global Hawk long-range UAV starts test program. (Design Watch).” Military & Aerospace Electronics 13 (8):6-7. Wright, Alex. 2015. “Algorithmic authors.” Communications of the ACM 58 (11):12-14. Wynn-Williams, Charles Eryl. 1931. “The use of thyratrons for high speed automatic counting physical phenomena.” Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character 132 (819):295-310. Yan, Hong-Sen, and Tsung-Yi Lin. 2002. “A study on ancient Chinese time laws and the time-telling system of Su Song's clock tower.” Mechanism and Machine Theory 37 (1):15-33. Yan, Ru-jian, Shuo Pang, Han-bing Sun, and Yong-jie Pang. 2010. “Development and missions of unmanned surface vehicle.” Journal of Marine Science and Application 9 (4):451-457. Yates, David R, Christophe Vaessen, and Morgan Roupret. 2011. “From Leonardo to da Vinci: the history of robotဨassisted surgery in urology.” BJU International 108 (11):1708-1713. Zhang, Fumin, Giacomo Marani, Ryan N Smith, and Hyun Taek Choi. 2015. “Future trends in marine robotics [tc spotlight].” IEEE Robotics & Automation Magazine 22 (1):14-122. Zielinska, Teresa. 2016. “Professional and personal service robots.” International Journal of Robotics Applications and Technologies (IJRAT) 4 (1):63-82.

PART II CONSUMPTION AND PRODUCTION IN THE ROBOTIC AGE

CHAPTER 4 CONSUMPTION, SATISFACTION, AND LIFE’S PURPOSE: A NEW FRAMEWORK OF CONSUMPTION THEORY

Recently, human history has been a history of want rather than affluence. Like all other animals, humans struggled to find food and other resources to make a subsistence living. The productive capacity of humans had been relatively limited, and resources were scarce. Therefore, for almost everyone, the survival of self and their genes became the purpose of life, and the priority in life was to obtain enough materials to live on. However, this may differ in an affluent society where automation and artificial intelligence (AI) have significantly increased productivity, as we have examined in Part One of this book. Since the rational demand for material goods and services by individuals and society will be almost fully satisfied, humans will no longer take obtaining enough food and clothes as their purpose. Fulfilling life’s purpose will become more diverse, proportionally less material, more affective, and more spiritual. This chapter aims first to examine the various features of consumption and their time, space, and physiological implications; and second, to present a theoretical framework for incorporating temporal, spatial, and physiological constraints into consumer or personal choices. By doing so, we may develop a new theory of individual choice that applies to situations when goods and services are scarce and when they are abundant. The rest of this chapter is structured as follows: Section 1 examines human desires, the objects, relations, and activities that cause satisfaction; Section 2 looks into ways that satisfy human desires; Section 3 investigates the process and determinants of consumption; Section 4 models constraints of consumption; Section 5 classifies different human needs that drive consumption; Section 6 explores the issue of utility maximization with segmented needs as well as temporal, spatial, and physiological constraints; Section 7 discusses the implications of the new framework, and Section 8 summarizes this chapter.

Consumption, Satisfaction, and Life’s Purpose

205

1. Objects, Relations, and Activities That Cause Satisfaction There are many objects modern humans desire to possess or consume, many relationships they desire to have, and many activities they desire to participate in because these objects, relationships, and actions would make them happy, satisfied, and excited. The desires for some objects, relationships, and activities are shared with other animals, such as the desire for food, especially delicious food. They come from human animality and are essential for the survival of individuals and humankind. More human desires, like fame and esteem, are unique to human beings and arise from human sociality or humanity rather than animality. The first type of desire is, of course, for food, which provides the energy for all other human activities if we include water and other drinks as part of food and ignore the need for breathing air/oxygen. In most of human history, the average person, like most animals, struggled to get sufficient food for survival. This type of desire or need is fundamental to life. Without life, other desires usually become meaningless. With the advance of production technology, humans have more choices in their food menu, and gradually eating acquires more psychological and sociological significance. Eating some food becomes a symbol of social status and wealth. The second type of desire is for shelters to protect humans against natural or non-natural adverse factors. Animals and the early ape-man used caves to defend against bad weather and predator attacks. Humans built and owned houses initially as more convenient and more comfortable replacements for natural caves. Soon buildings also became a symbol of social status and power. Rich and powerful people live in magnificent palaces, while the poor live in slums. Clothes are largely unique to humans and can be viewed as some mobile shelter or externalization of furs, feathers, and shells in other animals. Furs, down, and feathers in animals might already have psychological or social significance, but human clothes have gained more symbolic power regarding social belonging and status. This desire for shelters develops into that for abstractive safety and security of themselves, their family members, and their belongings. The third type of desire is for acceptance by peers. Many animals are social ones who seem afraid of living alone. Generally speaking, individuals want to be accepted by their peers and value such acceptance. Being isolated and rejected by others often causes an individual to become distressed or depressed. In an age of revolution, many revolutionaries would rather

206

Chapter 4

sacrifice their lives than betray their cause, partly because they could not bear being viewed and rejected as traitors by their fellow revolutionaries. In human society, individuals try to be accepted by their peers by behaving according to social conventions, even by being submissive to the herd leader. These behaviors may include wearing clothes of similar styles and engaging with or refraining from certain activities. Unconventional individuals are often isolated by their peers. Many social conventions are formed to subdue human animality. For example, in many societies in the past, extra-marital and pre-marital sexual relationships were frowned upon or explicitly prohibited, especially for women. Most resisted such relationships to be viewed as decent people. People who violate such conventions are often labeled indecent and isolated in the community. Following the sexual revolution (also called sexual liberation) from the 1960s to 1980s, sexual encounters between unmarried adults became more acceptable (Robinson et al. 1991; Greenwood and Guner 2010). Sex between unmarried adults increased, and casual sex occurred at a level never seen or heard of before. Feminist thinkers urged women to initiate sexual advances, enjoy sex, and experiment with new forms of sexuality (Ferguson 1984). Some men and women sought more sexual satisfaction than traditional marriage by instilling new institutions of open marriage, mate swapping, swinging, and communal sex (Cole and Spaniard 1974; Rubin 2001). These arise because social conventions on sexual relations have changed. The fourth type of desire is for intimate relations between individuals, especially between opposite sexes. Inter-sex relations in animals are mainly for reproduction and rearing their young. Interpersonal relations become more important. People gain satisfaction or utility from personal relationships. This relationship can be real or imagined. For example, fans gain satisfaction from following their idols, and fictional characters move readers of fiction. The relationship with fictional characters or remote heroes is a substitute for desired relationships in the real world. Love and associated sexual relationships have been an eternal literature theme and essential factors influencing political and economic outcomes. Libido, or sex drive, is necessary for human/animal physiological functions. Intimacy and sexual relationship between opposite sexes, not in exchange for payment, are essential sources of satisfaction and happiness in modern society. A harmonious sexual relationship between married couples is one of the foundations for a happy and stable marriage. Both partners will get

Consumption, Satisfaction, and Life’s Purpose

207

satisfaction from their sex life. Since many societies have permitted prostitution, men unable to satisfy their sexual desires could pay for sex with sex workers. Prostitution is often referred to euphemistically as “the world's oldest profession” and is estimated to have annual worldwide revenue of over $100 billion (Lehrer 2014). Prostitution is one branch of the sex industry, along with pornography, stripping, erotic dancing, etc. The fifth type of desire is for leisure, convenience, and ease of life. While animals and humans are mobile, they can become tired physically or mentally. Since they have to try to satisfy their other desires, they treasure the opportunities and facilities that let them enjoy their desired object, relations, and activities without any effort or with reduced efforts. Because of this desire, people want goods and services that make their lives more convenient. For example, cars improve their mobility; television and computers connected to the Internet enable them to see remote lands at home and watch entertainment without leaving home. The sixth type of desire is for a higher social status than their peers and, at the extreme, to become the head of their group. Many social animals have leading individuals in their communities, who have preferential access to resources such as opportunities for food and reproduction. A monkey king often monopolizes the right to mate with female monkeys in the herd. Only alpha wolves have the right to reproduce in a wolf pack. Some humans have a strong tendency to animality and enjoy bossing people around. Monarchs in ancient oriental countries had many consorts and absolute power over their subjects, who were probably more submissive to the monarchs than subordinate animals to the dominant members of the herd. Most people in developed countries enjoy better material and cultural goods and services than monarchs one thousand years ago. Would someone rather be an ancient monarch than an average modern citizen? People may have different answers to this question, reflecting their preferences and utility function. Some people might prefer to be an ancient monarch because they value a status much higher than their peers and the relative quality of their standard of living instead of the absolute quality of their standard of living. Their utility or happiness is mainly determined by the relative quantity of the goods and services they enjoy compared with their peers rather than the absolute amount of the goods and services they have. Other people may value the absolute quality of their standard of living more and be happier to see their peers enjoy the same or higher level of living. Whether an individual’s utility is determined by the relative or absolute quantity of goods and services affects her economic and political behavior.

208

Chapter 4

The seventh type of desire is for beauty, associated mainly with the visual and perhaps auditory senses. The desire for satisfaction with other sensory modalities also belongs to this type. Although this type of desire appears to have its origin in animals, for example, a peahen is attracted by a peacock in its pride, this desire is more of a human feature. Clothes arose as some shelter, but they have also served as decorations later. As discussed in Chapter One, humans made decorative objects early in history. People tried to use these objects to make themselves look prettier. In addition to appreciating beauty in themselves and their counterparties, people also started to appreciate the beauty in objects, activities, sounds, and lights with regularities and environments, which led to the emergence of art. Deliciousness to the gustatory sense and fragrance to the olfactory sense can be viewed similarly as beauty to the visual sense. The desire for delicacies derives from the need for food for the body’s energy. It develops into the need to satisfy the gustatory and olfactory senses and their associated advanced brain functions. Soft materials for clothes and other domestic goods can satisfy the tactile sense. The eighth type of desire is to possess resources beyond the need justified by other desires. Here resources denote anything of use or value, so wealth is vital. Many animals have their territories and would attack other animals entering their parts. The human desire to possess resources probably is an inherited animality, but this desire appears stronger in humans than animals. Ancient emperors and kings wanted to conquer as many lands and people as possible. Ordinary people want to save and accumulate a large sum of wealth, partly explained by the desire for security of future standard of living. However, there is still a component of the desire to possess more resources. The super-rich who have accumulated enough wealth for consumption by their future generations are still working hard to earn more, which can be explained by the desire to possess resources or self-fulfillment. The ninth type of desire is to understand unknowns. Many animals are inquisitive, and they want to explore new environments. One factor underlying their inquisitive behavior might be to find new sources of food. Humans probably are more inquisitive than animals, and human curiosity is the most critical drive for scientific research. Some are curious about nature and how to make new things. Some are interested in what other people are doing, gossiping about people around them, or talking about celebrities. With progress in information and communications technology (ICT), it has become much easier to follow and discuss what celebrities do. The desire to understand unknowns also underlies people’s learning behaviors.

Consumption, Satisfaction, and Life’s Purpose

209

The tenth type of desire is for self-fulfillment. This desire may have different levels of manifestation. At its lower level, people want confidence, self-esteem, and respect from others. The demand for equality, freedom, and justice can be viewed as an extension of the desire for self-fulfillment. Vanity can also be considered a distorted form of the desire for selffulfillment. Most people would continue to engage in competitive, productive, or creative activities because they desire self-fulfillment, even if they need not earn a living. These competitive, productive, or creative activities could be simply playing chess with friends and making toys or furniture at home. At its middle level, people want recognition of their ability, contribution, and achievements by society or those surrounding them. At its highest level, people get satisfaction from helping people and creating wealth or knowledge for humanity out of spontaneity, not being concerned by whether other people recognize their deeds. The above desires are not mutually exclusive. Although some desires are more important to life, they do not form a clear-cut hierarchy of human needs. Concerning different levels of human needs, Abraham Maslow presented his theory on the hierarchy of needs in 1943, which was further elaborated by him in later publications and widely used in management theories to illustrate how to motivate workers (Pinder 1998; Szilagyi and Wallace 1990). The hierarchy of needs is often portrayed in the shape of a pyramid, with the largest, most fundamental levels of needs at the bottom and the need for self-actualization at the top. The more fundamental four layers of the pyramid contain what Maslow called "deficiency needs" or "dneeds": esteem, friendship and love, security, and physiological needs. The most basic needs must be met before the individual strongly desires the secondary or higher-level needs. Later, Maslow (1969) added selftranscendence above self-actualization as another level of human needs (Fig.4-1). Physiological needs are the physical requirements for human survival. Air, water, and food are metabolic requirements for all animals, including humans. Clothing and shelter provide necessary protection from bad weather. According to Maslow, when an individual’s physical needs are relatively satisfied, their safety/security needs take precedence and dominate behavior. The next level of human needs is interpersonal and involves feelings of belongingness. Humans need to feel a sense of belonging and acceptance among their social groups. This need for belonging may overcome the physiological and security needs, depending on the strength of the peer pressure. After belongingness has been met, the next level of needs is esteem. People will want to be respected by others and

210

Chapter 4

need self-esteem first. A sense of achievement will give people confidence and self-esteem, making them feel respected by their peers. With selfesteem and respect from others, self-actualization is the next level of needs. People try to influence the world around them (for the better) by projecting their morality, creativity, and capacity to solve problems.

Selftranscendence Self-actualization Esteem Love/belonging Safety Physiological

Fig.4-1 Maslow’s hierarchy of human needs is portrayed as a pyramid.

In empirical studies, isolating all these levels of human needs may be challenging. People’s perception of the priority order in their needs varies with their circumstances. Many researchers used factor analytic techniques to test Maslow’s hierarchy theory of needs (Henne and Locke 1985; Locke, Sirota, and Wolfson 1976) and failed to find all five levels of human needs. Some only found lower-order needs (physiological needs) and higher-order needs (psychological needs) (Wahba and Bridwell 1976). Some studies on the influence of the Persian Gulf War on people’s satisfaction of needs showed that in US citizens, two factors of needs were identified during peacetime and three factors during wartime. In comparison, in Middle East employees, three factors were identified during peacetime and two during wartime. There were differences in these factors between US citizens and Middle East employees and between peacetime and wartime (Tang and West 1997; Tang and Safwat Ibrahim 1998). Children appeared to have higher physical need scores than the other groups, while older people had the highest level of security needs (Goebel and Brown 1981).

Consumption, Satisfaction, and Life’s Purpose

211

Although there are controversies regarding the details of Maslow’s hierarchy theory of human need (Wahba and Bridwell 1976; Hofstede 1984; Gambrel and Cianci 2003; Villarica 2011; Tay and Diener 2011), the existence of universal human needs and different levels of human needs are generally accepted. In this chapter, while acknowledging that other human desires may become dominant under different conditions and situations, we do not try to fit them into a clear-cut hierarchy because, except the basic physiological needs for survival, other needs tend to emerge in parallel though at different intensities for different individuals under different scenarios. We will directly consider the above ten desires or classify them roughly into physiological, psychological, sociological, methodological, and spiritual needs.

2. Ways to Satisfy Human Desires Humans first inherited from the ape-men hunter-gatherer skills and interpersonal relationships, and their desires were met through these skills and relationships. For almost all time in human history, nearly all people’s primary desires are food, shelter, possessing wealth, acceptance by their peers, and intimate relations. However, other desires such as leisure, understanding unknowns, and pursuing beauty also have their influences. The desire for food, shelter, and wealth can be met by production, exchange with others, or seizing from others by force. Taking resources from others by force is costly to everyone in the community. Therefore, communities tend to set up laws through agreement or accept rules imposed by the most powerful robbers, who begin to levy taxes in place of robbery to enforce the rules to make robbery illegal. With some level of law and order, people can get their food, shelter, and wealth through production by themselves, working for others to earn an income, and exchanging with others to obtain goods they do not have. In a hunter-gatherer society, people relied on the output of the local fauna and flora for their production and consumption. They worked to meet their desire for food and shelter. After catching or collecting what nature produced, they used the rest of their time for leisure or sleep time. The primitive tools made their life easier and more convenient than their bare hands. The desire for acceptance by peers and intimate relationships was achieved through collaboration in obtaining food, shelter, and connection between opposite genders. The stronger or smarter among them might satisfy their desire for higher social status by force, stratagem, or contribution

212

Chapter 4

to the herd. The herd had to work together to control the resources they needed. They could not satisfy their other desires in the modern sense. In an agricultural society, people had more control over what they could obtain from nature than the hunter-gatherers because they could choose the crops and animals to be raised in their regions to a certain extent, and the increased productivity for food and clothes enabled them to desire decorations for beauty and more sophisticated tools to make life easier. The increasing demand for decorations and tools made it possible to support some full-time craftsmen. With the development of agricultural technologies, one farmer could provide food for more and more people, such that more and more people could work on non-farming jobs and exchange food with farmers, and the social structure became more complex. Some elites could take spreading knowledge as their job to meet the demand for learning new knowledge from others. An elite class emerged to manage or rule the community. While the ruling elites cultivated their desires for beauty, possessing resources, understanding unknowns, and fulfilling their purpose in life, which usually looked primitive to the modern eye, the common majority struggled to satisfy their desire for food, clothes, and shelter. Poor people could only expect acceptance and intimate relationships among their peers. With the desire to control more resources and rule more people, monarchs waged wars to conquer other lands, which on the one hand, resulted in heavy losses of human life and properties and, on the other hand, facilitated the exchange of goods and information in expanded markets and economies. Probably as countermeasures against ruling elites’ raising social status by grabbing more resources and oppressing ordinary people, early thinkers developed religious and moral philosophies that promoted good deeds toward other people, especially ordinary people. In ancient China, having virtuous deeds as a role model for others, outstanding achievements in serving the community or state, and important ideas that promoted the progress of human society and civilization were viewed as three things that would make eternal fame and high status for a person. Confucius developed his moral philosophy with benevolence (ren) as the core concept, which later became the dominant ideology in China for over 2000 years. In the West, St. Augustine considered that happiness or the good life could be brought about by possessing the greatest good in nature that humans could attain and that one could not lose against one’s will, which greatly influenced Western society for over 1000 years. With such dominant ideologies, elites and ordinary people could try to raise their social status

Consumption, Satisfaction, and Life’s Purpose

213

and fulfill their purpose in life by being ethical and advancing humanity's causes. With the Industrial Revolution, agriculture's productivity increased even faster because the industry introduced new tools, machines, and technologies into agricultural production. Satisfying desires for food in terms of energy and shelter for protection is no longer a big problem for people in developed countries. Ordinary people can also aspire to pursue their desire for acceptance, respect, and self-esteem, desire for beauty, desire for ease, and convenience, desire to possess more resources, desire to understand unknowns, and desire for self-fulfillment. With the rise of commoners, liberty, equality, and fraternity have become essential parts of the value system in some societies. Although modern people no longer agree with the strict moral codes concerning family, marriage, and loyalty to monarchs promoted by ancient sages, noble deeds and a sense of justice are still highly regarded and practiced by people. The consumption of goods and services alone can no longer fully satisfy consumers. Many also want the producers to be ethical, practice fair trade, and engage in corporate social responsibility (CSR) activities. Consumers’ desire for these practices forces producers to adopt them. Inappropriate behaviors often enable people to satisfy their desires for intimate relationships, more resources, and higher social status. Some people give expensive gifts or money to their superiors, expecting the leaders to help their career progress. In the movie industry, some film directors and filmmakers use leading roles as bait to seduce or force actresses to have sexual relationships with them. In China, many corrupt male government senior officials, especially party chiefs or local chiefs, would promote female subordinates who have a sexual relationship with them. Some female officials became their local chief’s mistresses to get a promotion. Some businesswomen became local chief’s mistresses to get government purchase orders or to run government development projects. In some situations, the sexual relationship between a famous or influential person and an ordinary person is an idol-fan relationship without much intending financial or career gains. For example, Monica Lewinsky did not make economic or career gains from her relationship with former US President Bill Clinton. The position of national leaders can best meet the desire for high social status. Without limits on terms in office, national leaders tend to hold on to their jobs even if they are given the same pay during retirement. Some national leaders could be corrupted and want to stay on for personal or

214

Chapter 4

familial financial gains, but many want to stay on to serve their countries. The status as a national leader and the respect from average citizens give more utility than money can buy. Therefore, many want to be in prestigious or influential jobs without being paid. US President Donald Trump wanted only a $1 salary and donated all the rest. Other (former) senior government officials with a $1 salary include former US President Herbert Hoover, former US President John F. Kennedy, former California Governor Arnold Schwarzenegger, former Massachusetts Governor Mitt Romney, former New York City Mayor Michael Bloomberg, and former Los Angeles City Mayor Richard Riordan. Many senior business executives also have a $1 salary, although they might aim for a large bonus or a large number of shares or stock options (Tuttle 2016). For people working for no pay or only nominal pay, it becomes unclear whether their work is work or consumption. People pay to do physical exercises in gyms, which is generally considered consumption. There are many jobs for which people are interested in doing voluntarily if they have sufficient other incomes. These jobs include those with high social status, those with considerable executive power, those that make job holders wellrespected by fellow citizens, and those consistent with job holders’ pursuits. For people doing a job wholeheartedly without pay, we might view their work as consumption rather than work because these jobs satisfy their desires. Even for jobs opposite to the above described, many people may still volunteer to do them because the jobs are needed by society, and benefiting society is a desire of the volunteers.

3. The Process and Determinants of Consumption People have many different desires which can be satisfied by various means. Mainstream economics has traditionally focused on consuming goods to satisfy people’s desires and produce utility, i.e., satisfaction or happiness in economics. Generally, a good satisfies human desires and provides utility (satisfaction) to a person who consumes it. A desire for something is called a want in economics. Goods include properties (also called goods in a narrow sense) and services that are some forms of activities. Properties (goods) can be classified further into tangible and intangible properties. Information is an example of intangible property, which can be perceived only through print, broadcast, or computer. Consumption, in its narrow sense, is to acquire and use goods that are supplied through the market, and in its broad sense can be to use any goods. Although traditional economics

Consumption, Satisfaction, and Life’s Purpose

215

has reduced consumption to expenditure on goods, consumption has diverse forms and rich contents.

3.1. The Process of Consumption Traditional economics reduces consumption to purchasing final goods and services, but satisfaction or utility is not usually achieved by paying. People get satisfaction from eating food rather than buying food. The whole process for a good to be consumed may include the following components: planning, selecting, purchasing, consuming per se, and sometimes disposing of the rubbish produced by consumption. The planning stage may start with an idea of some consumption need and then find the existence of a good that meets this need. If it is a routinely consumed good, planning takes minimum time. The larger the money to be spent and the less frequently the good to be purchased, the more time is needed for planning. The selection period involves going to shops, making telephone calls to shops, or searching on the Internet. Before the advent of the Internet, consumers often visited several shops to compare and select a good of a particular type, which could take substantial time. Therefore, high-income people would hire servants for this type of selection, which became a service job instead of part of the consuming experience. For some individuals, planning and selecting goods are parts of the consuming experiences consumers actually enjoy. With the advent of the Internet, the planning and selecting process can be performed online for many goods, which saves searching time and may increase selection time due to information overload. In the future, it is possible that computers can make choices for consumers in their purchase decisions. The purchasing stage can be an enjoyable experience; some consumers may buy things they never use. Some goods need to be transported physically to the venue of consumption. Some goods (mainly services) can or must be consumed almost on the spot of purchase, although advance payment is always possible. Furniture is consumed not at the spot of purchase, while restaurant foods tend to be consumed on the spot of purchase. Cooking at home is part of the process of consuming food bought by consumers, which can be converted to domestic work/production and separated from consumption. With Internet shopping, goods can be purchased online and delivered by delivery firms. E-commerce and Internet shopping have greatly stimulated the growth of delivery firms. This is a transformation of consumption components into services. With the Internet, computer science, especially software, automation, and robotics, many components of a

216

Chapter 4

consumption process are transformed into goods (services). What cannot be transformed into goods is the core of consumption, the personal experience of using the goods or services per se or consumption per se. The selection per se may also be difficult to delegate to other people. Using the goods or services and, to a lesser extent selecting the goods or services are the two components that cannot be replaced in a consumption process. With new technologies and higher personal incomes, online shopping and delivery services can replace travel to shops and carrying goods home. Even selecting goods and services will be much easier when intelligent software can track and mimic your preference and recommend goods you would choose yourself. Then how consumers use the goods and services will characterize the consumption process. Goods in their broadest sense (i.e., something that satisfies a want or desire) can be classified according to who can consume them and how they are consumed into the following categories (Table 4-1): A. Own goods, such as sleep and leisure, can only be consumed by oneself. The quality of consuming own goods often depends on the quality of some private goods their consumers possess and the quality of some public goods. Comfortable furniture and a quiet environment are conducive to good sleep. Noise can stop a person from consuming sleep. Sleep and leisure need space and time to consume, but people need not pay for their sleep and leisure. People often must pay for the space and associated services to sleep well. The production/supply and demand/consumption of own goods are the same person and supply and consumption are simultaneous. Consumption of own goods has little externality, although its impact may have severe consequences; for example, a driver falling asleep often causes a severe accident. B. Relational goods are consumed together, such as a couple's intimate relationship and friendship between people. Partners together produce relational goods; although one or more partners are the driving force, the others are more passive. The consumption of relational goods need not be simultaneous with their production. Consuming relational goods without supplementing them will exhaust them. Relational goods generally need time and space to produce, but their consumption usually needs no time or space except for intimate acts between partners, which might more appropriately be classified as private goods.

Consumption, Satisfaction, and Life’s Purpose

217

Table 1 Types of goods in a broad sense

Own Goods Relation Goods

Private Common Goods

Public Common Goods

Feature Consumed only by oneself

Subtype

Rivalry Not applicable

Excludability Not applicable

Examples Sleep, Leisure

Consumed only in relation to each other

Exclusive Relation

Yes

Yes

Romance

Inclusive Relation

No

No

Friendship

Consumed by anyone but not in a good’s entirety by two or more people simultaneously Consumed in a good’s entirety by two or more people simultaneously

Private Goods

Yes

Yes

Food

Common Resources

Yes

No

Fish in oceans

Public Goods

No

No

Safe environment

Club Goods

No

Yes

Internet service

Table 2 Types of private common goods Feature

Subtype

Consumed only once

On the spot of supply Not on the spot of supply Active

One-time Consumption

Multi-time Consumption

Continuous Consumption

Consumed multiple times

Passive No clear action involved

Need Time

Need Space

Physiological Constraint

Yes

No*

Yes

Services

Yes

No*

Yes

Disposable goods

Yes

Yes

Yes

Bikes, cars, books

No**

Yes

Yes

Clothes

No

Yes

No

Decoration, furniture

Examples

* May still need some space. ** May still need some time to prepare or choose.

218

Chapter 4

C. Private and common-resource goods: these goods are rival in consumption, which means that a good cannot be consumed in its entirety by two people; for example, if one consumer has eaten an apple, other consumers cannot eat the same apple again. Economics divides such goods into private goods and common resources according to their excludability. If people can be excluded from consuming certain goods rival in consumption, these goods are defined as private goods. If people cannot be excluded from consuming these goods, they are defined as common resources. The actual consumption process of private goods and common resources can be classified into three groups (Table 2): 1) one-time consumption goods, such as food, drink, and many personally-received services, which can be further divided into a) those to be consumed on the spot of supply and b) those not restricted to consumption on the spot of supply; 2) multi-time consumption goods which involve the action of using them, such as bikes, cars, books, and some furniture; and 3) decorative, protective and convenience-providing goods which improve the living environment and may not involve the action of using them. Consuming one-time goods needs time (and probably space). Using multi-time goods needs time and space, and storing them also needs space. Consuming decorative/protective goods requires space, whereas time is usually unnecessary. Clothes have both the characteristics of multi-time goods and decorative/protective goods. Choosing and putting on clothes take time, but wearing clothes takes little time. A good reputation and senior government offices are also decorative, protective, or convenience goods. The suppliers of good reputations and senior government offices are the general public or government. Individuals need to make efforts to earn them. D. Public and club goods: these goods are non-rival in consumption, which means that a good can be consumed by two or more people, such as radio broadcasts, websites, clean air, and a safe social environment. In economics, if people can be excluded from consuming such non-rival goods, the non-rival goods are defined as natural monopolies or club goods. If people cannot be excluded from consuming such goods, they are defined as public goods. Public and club goods, such as radio broadcasts and cable television, need consumption time. Free public roads, parks, and museums where there is some rivalry in consumption when

Consumption, Satisfaction, and Life’s Purpose

219

crowded also need time to consume. Clean air and a safe social environment as public goods need no time to consume.

3.2. Determinants of Consumption Although consumption is one of the most important economic activities, mainstream economics has paid little attention to the various features of consumption in investigating the rational choice of consumers. Consumption has been reduced to the individual or household expenditure on consumer goods and services, and the interest of economists has been focused more on what determines consumption expenditure than consumption per se. According to John Maynard Keynes, current consumption is determined primarily by current income, with the marginal propensity to consume between zero and one (Keynes 1936). The life-cycle theory of Franco Modigliani and Richard Brumberg posits that current consumption is determined by lifetime income, and individuals will smooth their consumption over their lifetime by saving and borrowing based on the expectation of their lifetime income (Modigliani and Brumberg 1954). The permanent income hypothesis proposes that current consumption is primarily determined by permanent income, typically defined as the average or expected income (Friedman 1957). Precautionary saving (Carroll 1997) and liquidity constraints (Deaton 1991) have been proposed to explain the buffering saving behaviors. As traditional economics treats consumption as an activity that needs no time, consumption by consumers is the spending on goods and services, i.e., the purchase of goods and services by individuals. If there was no transaction involved, for example, a partner cooked for and served their partner and children a delicious dinner, there was no consumption (other than materials and ingredients bought for preparing the dinner). If the other partner cut the grass in their garden, there was no consumption; but if they cut the grass for their neighbor and was paid, there was consumption by their neighbor. Taking goods or services without payments is not consumption in economics. Not all purchases are consumption; goods purchased by firms are divided into intermediate consumption, intermediate products used for producing final products, and investment (in physical capital). Goods a government purchases include government spending (on consumables) and investment (durable goods such as buildings and equipment). Durable goods such as buildings bought by individual consumers are also treated as investments, bringing notional incomes (investment returns) to owner-occupiers.

220

Chapter 4

Reducing consumption to its expenditure implies that the individual capacity to consume is infinite, and consumption does not take time or need space. This simplification was appropriate during an age of want when the quantities of goods and services an individual could consume were small compared with their maximum capacity (if there is one). As discussed in the preceding section, most goods and services require time to consume; therefore, with technological progress bringing human society into an age of affluence, more people will need to consider allocating their time to different goods. In the theory of rational consumer choice in economics, only leisure needs time to consume, and the individual time endowment is allocated between work to earn an income for consumption and leisure or home production (Gronau 1976). However, rational individuals must also spend time on goods they have purchased or acquired without pay. If leisure or home production includes the time required to consume the goods a consumer purchases, its optimum will be determined by the time needed to consume the goods bought and the wage rate. Thus, the individual labor supply curve based on leisure-wage analysis in traditional economics becomes invalid. The finite individual time endowment and the time costs of consumption suggest that the maximum individual capacity to consume should be finite. This could explain why many wealthiest people, such as Bill Gates, Jeff Bezos, and Warren Buffett, consume little more than average upper-middleclass people (Cain 2017b, c, a). In the future, more people will find that their need for material goods and services has been sufficiently met, and they want to spend more time on activities that meet their needs for selfactualization and self-transcendence (Maslow 1954, 1969). Besides time, human physiology also puts some internal constraints on the individual capacity to consume, which might be more critical in consumption decisions than monetary and time constraints in the age of AI and robots. For example, some physical space in the gastrointestinal tract is necessary for consuming food and drink. People get drunk or unconscious and become temporarily incapacitated after consuming a certain amount of alcoholic beverage. Patients with gout have to refrain from consuming many delicious foods. Those with cardiovascular issues often have to reduce their intake of cholesterol-rich food. Almost everyone is advised to reduce their high-energy or fatty food intake. Moreover, continuous physical and mental activities would make a person physically and mentally tired, which could be viewed as the person’s physiological capacity for that activity has been used up. In addition to time and physiological constraints, people need space to consume, to feel safe and secure, or to store goods for future consumption.

Consumption, Satisfaction, and Life’s Purpose

221

From the discussion above, income or wealth is not the unique determinant of consumption. Since the consumption process of most goods needs time and everyone’s time endowment is limited, time will become an important determinant of consumption for a person of affluence. Each type of consumption is to satisfy a desire underlaid by some human physiological mechanism. Hence it depends on such physiological capacity, especially for a person of affluence. In addition to income or money, time, and physiological capacity, some external space is also needed. For example, a banquet needs space, usually provided by a restaurant, and you are charged for the food and drinks. If you have a banquet at home or on free public land, you use your space resource or a public resource. Therefore, money, time, physiological capacity, and space are four critical determinants of consumption. In an age of want, time and physiological capacity can be ignored; they must be considered in an age of affluence.

3.3. Consumer Rationality and Maximum Sensible Consumption Economics was born in an age of want just before industrialization. Modern economics’ main concerns and way of thinking still have a deep imprint of that age. The basic concern of traditional economics is allocating scarce resources, capital, labor, natural resources, and products to maximize people’s utility. Underlying optimal allocations are the three axioms of the rationality of consumer preference, Completeness, Transitivity, and Nonsatiation. The three axioms of rationality can be defined as follows: 1) Completeness: either A>B, or B>A, or A=B. 2) Transitivity: if A>B and B>C, then A>C. 3) Non-satiation: The more goods that produce positive utility, the better. For people who prefer diversity in consumption rather than monotonicity, their preference can be characterized as 4) The convex indifference curve. Economists assume that people prefer variety in their consumption, which is true in general. The rationality axioms have two weaknesses due to economics’ birth in an age of want rather than affluence. One is the implicit assumption that utility is determined only by consumption (of material goods and services), and the other is neglecting the conditions of non-satiation. Both weaknesses bear the imprint of an age of want such that human desires could be simplified to the desire for food, clothes, material goods, and services that primarily meet people’s physiological needs (and, to some extent, psychological and

222

Chapter 4

social needs). When people’s physiological and psychological needs are met, the axioms of rationality will show their shortcomings. The completeness axiom describes a logical or mathematical relationship between two choices, which is true. The axiom of transitivity should always be valid if satisfaction or utility is determined only by choices of consumption bundles contained in the relations. We should know this from mathematical logic. However, many factors analyzed in the preceding sections determine people’s satisfaction or utility. Therefore, the rationality represented by the three axioms in traditional economics is partial rationality based on material goods/services. General rationality or total rationality should incorporate the satisfaction of all human desires. As individuals’ physiological and safety needs are met, other needs become more apparent, leading to a “violation” of the partial rationality of traditional economics. But this violation reflects true rationality, the general rationality accounting for all individual needs. When predictions based on the assumption that material goods and services or money determine utility are tested by experimental economics or behavioral economics, the results are substantially different from those predicted by the axioms of transitivity and non-satiation. The term “bounded rationality” is coined to describe the discrepancy between experimental results and the conclusions of mainstream economics. As pointed out earlier, these discrepancies arise because the utility in traditional economics is reduced to meeting the desire for material goods and services or money. Apart from ignoring other desires and needs, traditional economics also ignores that deciding on choices has its internal costs. Here internal costs of making decisions denote the time, the mental (and possibly physical) efforts, and the emotional disturbance required to reach a decision based on the total computation of expected utility using all available information. We may call the costs to acquire goods or services external costs. These internal costs may reduce a person’s effort in making an optimal decision, i.e., maximizing the expected utility without internal costs. This is similar to external costs reducing consumption of goods and services. When decisionmaking requires much computation in the brain, intuition and approximation will be used to assess the favorability of goods and services, which can give different preference profiles from those based on complete computation. For this reason, “bounded rationality” is, in fact, pragmatic or practical rationality, which is optimal when internal costs are considered. In contrast with the axioms of completeness and transitivity, which describe the mathematical and logical relationship, non-satiation describes one

Consumption, Satisfaction, and Life’s Purpose

223

aspect of human nature. In an age of shortage, non-satiation is practically true, but it may need some qualifier for periods where the lack of goods and services is no longer a problem. As discussed earlier, desires for goods and services are subject to time, physiological capacity, and space constraints. The third axiom, non-satiation, is valid only within the time, physiological capacity, and space constraints. When the quantity of a good exceeds the time, physiological capacity, and space constraints, it might cause a reduction in utility as well. When this author taught the consumption topic in a microeconomics course, students were asked whether non-satiation would be violated if apples were transported into someone’s home in such a quantity that leaves no free space anymore. Students agreed that more apples would decrease that person’s utility at a certain point. The nonsatiation axiom appears salvageable by selling unwanted apples to other people or exchanging them for other goods. However, this can be true only when most people do not have many apples. Nobody can sell apples if most people face the same abundance problem. Similarly, water is a precious resource in dryland, so more rain increases utility. However, more rain will decrease utility at a certain point and even become a disaster. Given that a person has a limited time endowment, limited physiological capacity, and limited private space, as well as many desires to satisfy other than those for food, shelter, and convenience, it can be postulated that a person has a limited capacity to consume material goods and services which are usually supplied through the market. We may call this type of goods market goods or economic goods. As we have also identified many desires that cannot be satisfied by simply consuming economic goods, activities to meet those desires will compete for time, space, and physiological capacity with the consumption of economic goods. Therefore, a person’s maximum consumption is smaller than the maximum consumption capacity allowed by their time endowment, physiological capacity, and private space. Some may argue against the maximum capacity to consume economic goods because human greed is infinite. A person may engage in wasteful consumption for vanity or other purposes (conspicuous consumption). Here we distinguish sensible consumption and unreasonable or nonsensical consumption. With increased labor productivity, the desire for goods and services that can be supplied in large quantities is sensible. For commodities such as gold and rare gems, a desire to have a large amount of them is unreasonable. Wasting goods to show off, which sometimes does happen, is nonsensical consumption. Desires concerning personal relationships could often be unreasonable; for example, many people may dream of having a personal relationship with movie stars. In the Robotic Age, society

224

Chapter 4

may provide robotic copies of movie stars to those who want to have a personal relationship with them. Still, it is impossible to have natural persons for all who wish. In the Robotic Age, robot or virtual actors and actresses may be so successful that they drive entirely human actors/actresses out of acting. Then, the desire for human actors and actresses would no longer exist. A person’s maximum capacity for sensible consumption of economic goods is finite. In the Robotic Age, even with much-increased productivity, there seems to be no need to produce goods and services more than the aggregate demand that satisfies everyone’s maximum sensible consumption. We may separate two types of consumption demand: functional and vanity. Functional consumptions satisfy some functional needs; for example, eating nutritious, healthy, and tasty food is functional consumption. So is wearing comfortable clothes. Fashion accessories and cosmetics that enhance people’s appearance are also functional consumption. Pursuing goods made from rare natural materials, which often cannot be distinguished from those made from artificial materials during consumption, is vanity consumption. For example, fake diamonds cannot be distinguished from natural diamonds for decoration, such as in La Parure by the French novelist Guy de Maupassant. Similarly, eating particular foods, such as wild caviar, could be considered vanity consumption, as are clothes of expensive designer brands. In the Robotic Age, the market can still be the solution for matching supply with demand, and there is no particular need for central planning. All consumption needs, whether functional or vanity, can be met by market mechanisms if humans are still responsible for what to produce and consume. There is a dividing time between a society where the income budget constraint plays a role and a society where the income budget constraint does not play a role in determining functional consumption bundles. To reach the economic development stage of not considering income budget constraints for functional consumption, labor productivity must be at a level that can produce all goods and services for functional consumption. The income budget constraint may still apply to vanity consumption. It is quite possible in the Robotic Age that people are no longer interested in vanity consumption. Because the real limit for vanity consumption is the rarity of some particular natural resources, vanity consumption will put little production pressure on the economy. The high prices of designer clothes are artificial due to their limited output, and their high prices cannot be sustained if they are mass-produced.

Consumption, Satisfaction, and Life’s Purpose

225

4. Modeling Constraints of Consumption Besides consumers and goods, consumption has four more critical elements: monetary costs, time, physiological capacity, and space. They represent four constraints on the individual capacity to consume.

4.1. The Monetary Costs of Consumption In economics, consumption (except that of leisure) is implicitly assumed to take no time and space, so a rational consumer chooses among different consumption bundles under their monetary budget constraint. ‫ܲ = ܯ‬௑ ܳ௑ + ܲ௒ ܳ௒

(1)

In equation (1), M is the monetary budget constraint; ܲ௑ and ܲ௒ are the prices of goods X and Y, respectively; ܳ௑ and ܳ௒ are the quantities of goods X and Y, respectively. Without time, physiological capacity, and space constraints, a consumer’s capacity to consume is infinite. The more constituent goods, the higher the bundle’s utility. The axiom of non-satiation describes this feature of consumption. ܷ൫‫ܥ‬ሺܳଶ௑ , ܳଵ௒ ሻ൯ > ܷ൫‫ܥ‬ሺܳଵ௑ , ܳଵ௒ ሻ൯, ‫ܳ׊‬ଶ௑ > ܳଵ௑

(2)

In equation (2), ܷሺ‫ܥ‬ሻ is the consumption C-derived utility, ܳଵ௑ and ܳଶ௑ the quantities of good X, and ܳଵ௒ the quantity of good Y. For the work and leisure decision, the time endowment multiplied by the wage rate becomes its monetary budget constraint. The consumer choice is between consumption of all other goods, equal to the income from work (hours × wage rate), and leisure priced at the wage rate.

4.2. Time Costs of Consumption Consumption costs money and time (Goolsbee and Klenow 2006). Watching a movie or eating at a restaurant can be quite time-consuming. Consumption is not a transient action of purchase. Purchasing a good or service only acquires ownership, a condition for consumption rather than consumption per se. Although the shadow value of time used in consuming goods and services can be estimated from people’s consumption of timesaving products (Phaneuf 2011), the time cost of consuming a particular

Chapter 4

226

good cannot be substituted with an additional monetary cost. Consumers have to spend time enjoying a delicious meal. Y 8

B

C 4

A 0

6

12

X

Fig.4-2 A consumer’s time budget constraint determines the consumption bundles that the consumer has enough time to consume. The points on or below the line from A to B are feasible timewise.

When consumption has a time cost, money alone is insufficient to complete consumption. Consumers must allocate their time between work, leisure, and consumption of different goods. If good X takes two hours per piece to consume and good Y takes three hours per piece to consume, an individual can consume a maximum of 12 pieces of X or eight pieces of Y in a day without any sleep or doing other things (Fig.4-2), ܶ௑ ௔௡ௗ ௒ = ߬௑ ܳ௑ + ߬௒ ܳ௒

(3)

In equation (3), ܶ௑ ௔௡ௗ ௒ is the time budgeted for goods X and Y, ߬௑ the time cost of consuming one piece of X and ߬௒ the time cost of consuming one piece of Y.

4.3. Space Constraints of Consumption Although the consumption of goods or services does not “wear out” space, some external space may be needed for consumption to be carried out such that the space may manifest as a constraint for consuming certain goods. The external space constraints have two meanings. First, most goods occupy a certain area, implying that there can be a finite number of physical goods in a finite private space. Second, as an individual’s attention is only effective within a finite and relatively small space, their expenditure on goods and services consumed outside the area of reach is not their consumption except for vanity (and charity which will be considered consumption by the

Consumption, Satisfaction, and Life’s Purpose

227

beneficiaries). Both aspects of the external space constraints will present some limits on an individual’s maximum capacity for non-vanity consumption. There are three types of space needed for consumption in a home: the storage space, the furniture and decoration space, and the access and function space, which includes the consumption space required temporarily for goods immediately before and during their consumption. The storage space stores food, clothes, and other consumables for future consumption. These goods will share the storage space, ܸ௦௧௢௥௘ = ‫ݒ‬௦௧௢௥௘,ଵ ܺଵ + ‫ݒ‬௦௧௢௥௘,ଶ ܺଶ + ‫ ڮ‬+ ‫ݒ‬௦௧௢௥௘,௡ ܺ௡

(4)

In equation (4), ܸ௦௧௢௥௘ is the storage space; ‫ݒ‬௦௧௢௥௘,ଵ , ‫ݒ‬௦௧௢௥௘,ଶ , ‫ڮ‬, and ‫ݒ‬௦௧௢௥௘,௡ the space needed for storing one unit of goods 1, 2‫ڮ‬, and n, respectively; ܺଵ , ܺଶ , ‫ڮ‬, and ܺ௡ the quantities of goods 1, 2‫ڮ‬, and n, respectively. Although many types of goods share the storage space, consumers usually allocate subspaces of certain volumes for different goods. For example, they may assign some space for storing food, some space for clothes, and some space for books and other information media.

4.4. The Physiological Constraints of Consumption An internal space constraint means that space inside a consumer’s body is necessary for consumption. This is especially true for food and drink because some physical space in the gastrointestinal tract is essential for consuming them. The internal space for food is just one instance of the physiological capacity constraints, and violating it would damage our health. For example, the upper limit of the safe blood ethanol level is a constraint. People will have physical or mental fatigue after engaging in physical or mental activities for long hours, such as muscle fatigue, reading fatigue, learning fatigue, and probable fatigue from various physical, mental, and entertaining activities (Enoka and Duchateau 2008; Marcora, Staiano, and Manning 2009). We may view these fatigues as physiological capacity constraints on consuming various goods and services. The above analysis shows multiple physiological capacity subspaces for individual consumption and other activities. Food consumption is constrained by the internal volume of the gastrointestinal tract, primarily the stomach volume in the short term. In the long term, the blood sugar level, cholesterol level, etc., can all work as physiological capacity constraints for eating foods containing these nutrients (Grundy et al. 2005). We can

228

Chapter 4

describe the physiological space budget for consuming goods similarly as the monetary budget and the time budget, Ȱ௑ ௔௡ௗ ௒ = ߶௑ ܳ௑ + ߶௒ ܳ௒

(5)

In equation (5), Ȱ௑ ௔௡ௗ ௒ is the physiological subspace budgeted for goods X and Y, ߶௑ the physiological space needed for consuming one piece of good X, ߶௒ the physiological space needed for consuming one piece of good Y, ܳ௑ the number of good X, and ܳ௒ the number of good Y. A feasible consumption bundle for a consumer must be on or within all four budget lines. If the consumption bundle is above the monetary budget line, they cannot afford to consume it. They do not have enough time to consume it if it is above the time budget line. They do not have enough space to consume if it is above the space budget line. If it were above the physiological capacity budget line, it would cause damage to their health and reduce their utility. The combination of time, space, and physiological capacity constraints will break the non-satiation axiom.

5. A New Hierarchical Model of Human Needs Human desires for survival and a happy life are the internal drives for consumption and other utility-producing activities. We have examined various human desires in Section 1 of this chapter; some can be satisfied by economic goods, but economic goods may not meet other demands. Our list of desires is somehow different from Maslow’s theory of the hierarchy of needs (Maslow 1943, 1954, 1969). Although the existence of universal human needs and different levels of human needs are generally accepted, and Maslow’s hierarchy theory has been widely used in management studies (Pinder 1998; Szilagyi and Wallace 1990), the theory’s details are disputed (Wahba and Bridwell 1976; Tay and Diener 2011). Since higher-level human needs have not attracted enough attention from economics, we try to incorporate other desires into the consumer’s utility function in the present chapter. We classify human desires into physiological, psychological, sociological, methodological, and spiritual needs.

5.1. Physiological Needs Physiological needs include the desire for food and shelter. They are the physical requirements for human survival and essential health, which form the foundation for other needs (Martin and Hill 2012; Gupta and Srivastav 2016). Oxygen, water, and food are metabolic requirements for survival by

Consumption, Satisfaction, and Life’s Purpose

229

all known animals except three species of anaerobic Porifera, the tiny animals discovered at the bottom of the Mediterranean Sea (Danovaro et al. 2010). Breathing air is the most fundamental need for humans, but it is not an activity of consumption in economics because of the abundance of air and the difficulty of asserting its ownership. Drinking water is the second most important need, as animals can survive longer without food than without water (Campbell and Cicala 1962; Campbell, Teghtsoonian, and Williams 1961; Dufort 1963). Food provides the energy to drive all body activities and the nutrients for growth and repair. Keeping body temperature in the normal range is essential for survival and basic health. Shelters and clothes can help humans keep their body temperature in cold weather, and shelters also protect people from heatstroke. Visceral senses and the relevant neural centers in the brain primarily regulate these essential physiological needs. There are sensors (neural receptors) inside our body to sense deep body temperature (by thermoreceptors), blood volume (by volumetric receptors), osmotic pressure (by osmoceptors), and blood levels of glucose, amino acids, fatty acids, and hormones (by various chemoreceptors). Neural signals from those receptors are sent to the brain centers regulating food, water, salt intake, and temperature (Hall 2015). When an individual’s deep body temperature decreases, the temperature center in the brain will activate body tissues that produce heat, make the individual put on more clothes, and seek warm shelter. When blood glucose, amino acids, and fatty acids decrease to certain levels, individuals will feel hungry and look for food. When blood volume decreases or osmotic pressure increases, individuals will feel thirsty and seek water. The physiological space for food makes up its physiological constraint. Ȱ௙௢௢ௗ = ߶௙ଵ ܺଵ + ߶௙ଶ ܺଶ + ‫ ڮ‬+ ߶௙௡ ܺ௡

(6)

In equation (6), Ȱ௙௢௢ௗ is the physiological space for food; ߶௙ଵ , ߶௙ଶ , ‫ڮ‬, and ߶௙௡ the physiological space needed for consuming one unit of foods 1, 2‫ڮ‬, and n, respectively; ܺଵ , ܺଶ , ‫ڮ‬, and ܺ௡ the quantities of foods 1, 2‫ڮ‬, and n, respectively. The consumption of clothes also has a space constraint because the quantity and types of clothes an individual can wear simultaneously are finite and limited. Wearing too many clothes will make a person clumsy in action and laughable in the eyes of their peers. Dressing protects people from adverse weather, symbolizes social conventions, and improves wearers’ appearance

230

Chapter 4

and charm. Individuals need clothes for different seasons (physiological needs) and occasions (psychological and sociological needs). A subspace for clothes exists as part of the overall storage space. When the shortage of goods is no longer a problem, the axiom of nonsatiation may become invalid due to time and space constraints and the satisfaction of corresponding human physiological needs. The European Union and the United States have paid farmers to cut their planted areas to reduce their agricultural output. When the quantity of a good exceeds the time and space constraints, it might reduce utility. When apples are transported into someone’s home in such a quantity that leaves little free space, more apples will decrease that person’s utility.

5.2. Psychological Needs Psychological needs include the desires for leisure, ease and convenience, beauty, intimate relations, understanding unknowns, and possessing resources. They are the physical and non-physical requirements for safety, security, curiosity, pleasure, and fun. Some people read novels for fun, some listen to music, some watch soaps on TV or movies in cinemas, some play sports or games, some play with toys and various gadgets, and some enjoy delicious foods (Mulvey 1975; Small et al. 2001; Kringelbach, Vuust, and Geake 2008; Salimpoor et al. 2011; Perlovsky et al. 2013). Many people enjoy sexual intimacy (Rye and Meaney 2007). All these activities can be considered to satisfy psychological needs, which generally correspond to favorable stimuli in human sensory and cognitive modalities. Somatesthesia (which includes tactile, cutaneous temperature, and nociceptive senses) and special senses such as taste, smell, hearing, balance, and vision can evoke emotional and cognitive responses in the forms of pleasure, joy, euphoria, worry, anxiety, fear, depression, despair, dignity, humiliation, and curiosity. These emotional and cognitive responses are outcomes of evolutionary adaptation that help humans survive and live better lives. Materials that adequately excite these sensory and cognitive systems and signal favorable external and internal conditions become goods that produce satisfaction and pleasure via various sensory and cognitive modalities. Often there is no clear demarcation between consumption for physiological needs and that for psychological needs. Food and drink are essential for survival and having fun and pleasure because delicious foods satisfy people by acting on the senses of taste and smell. Some substances in foods and drinks can work on the central nervous system to cause joy, euphoria, and hallucination (Hindmarch 2004). In addition to delicious foods, fresh air and

Consumption, Satisfaction, and Life’s Purpose

231

aroma from plants or artificial fragrances in the environment also evoke satisfaction via smelling (Schleidt, Neumann, and Morishita 1988). People may also feel satisfaction by hearing dulcet voices, crisp birdsong, a breeze, light rain, or music. The right mixture of colors, brightness, and geometric shapes in pictures, landscapes, and surrounding environments and objects can evoke satisfaction via vision (Fiore 2006). Clothes provide warmth and make the wearers comfortable via tactile sense and more pleasing to the eye. Besides sensing relatively simple musical sounds and pleasant views, hearing and vision are the pathways for the more sophisticated cognitive system. Objects and materials that evoke satisfaction via those sensory modalities also take time and attention from those senses. Because of finite time endowment, the consumption of goods for producing satisfaction via those sensory modalities will also be finite and have some limits. Many consumption activities for psychological needs are entertainment. Entertainment requires time and physiological space in terms of fatigue growing out of prolonged engagement in one activity, so there is a maximum consumption even if they can be addictive. For people with sufficient income for food and clothes, emotional and cognitive satisfaction will be their major demand, and they will spend more time on entertainment and knowledge acquisition. ܶ௘௡௧௘௥௧௔௜௡௠௘௡௧ = ߬ாଵ ‫ܧ‬ଵ + ߬ாଶ ‫ܧ‬ଶ + ‫ ڮ‬+ ߬ா௡ ‫ܧ‬௡

(7)

In equation (7), ܶ௘௡௧௘௥௧௔௜௡௠௘௡௧ is the time budgeted for entertainment ߬ாଵ the time needed for consuming 1 unit of entertainment 1, ‫ܧ‬ଵ the quantity of entertainment 1, etc. Reading non-fiction and watching documentaries for general knowledge is entertainment rather than learning here. In contrast to physiological constraints due to fatigue, addiction to some activities and materials are phenomena where an action or consumption is self-reinforcing. Addiction is commonly associated with drug abuse, but some people are addicted to more innocuous activities and substances. Drug addicts become dependent on alcohol, cannabis, barbiturates, benzodiazepines, cocaine, methaqualone, opioids, or amphetamine derivatives (Hindmarch 2004; Wise 1996). Some people may be somehow addicted to food, which could lead to obesity and various obesity-related diseases (Pelchat 2009; Gearhardt et al. 2011; Davis et al. 2011). Many people are addicted to gaming (Rooij et al. 2015; Fisher 1994). However, there are also physiological constraints for various addictive substances because overdosing will cause health damage, incapacity, and even death. Moreover, addiction to some

232

Chapter 4

substances and activities will reduce the time (and money) budgeted for all other activities and decrease the ability to consume other goods, significantly reducing the overall demand for goods and services. Therefore, addiction does not contradict the view that the individual capacity to consume has a finite maximum.

5.3. Sociological Needs Sociological needs include seeking peer acceptance, intimate relations, high social status, and possessing resources. They are the requirements for empathy, love, friendship, belonging, being ethical, being heard by others, attention from others, respect by others, esteem, vanity, altruism, and achievements recognized by others (Batson 1987; Eisenberg and Miller 1987; Weiner 1985). Many people might think that these needs are also psychological. The difference between psychological and sociological needs in the present framework is that psychological needs are personal enjoyment from consuming goods and services. In contrast, sociological needs are appreciation by, and caring for, others. Unlike physiological and psychological needs, sociological needs usually do not have goods to satisfy them directly. Many consumption activities involving goods and services for physiological and psychological needs are rituals for promoting and maintaining close relationships between partners, family members, friends, or colleagues. Human beings treasure love, friendship, and belonging, and possessing these can give individuals great satisfaction that deserves the money and effort spent cultivating and nurturing. However, spending money on goods cannot buy love, friendship, and belonging. More attention and time have to be spent on activities that demonstrate good personality and good deeds because working hard and having a noble character is more likely to bring in friendship, love, attention from others, respect from others, esteem, acceptance by one’s social group, and achievements recognized by others. Attention from others and being heard are essential aspects of sociological needs. The thriving social media, blogs, personal websites, and files uploaded to websites such as YouTube reflect the needs to be heard and to get attention from others. Most bloggers and self-media journalists spend substantial time posting information online without receiving income. Their expected compensation for their efforts is to have Internet users’ attention and be heard. Similarly, Facebook, Twitter, WhatsApp, and WeChat users post their information and share files to connect with people.

Consumption, Satisfaction, and Life’s Purpose

233

Sociological needs can be classified into two categories: those consuming goods and services and those without. Promoting friendship and love by eating together, going to entertainment together, playing together, or having a holiday together belongs to the first category, which requires time and physiological space for consumption to satisfy the needs. Ȱ௙௢௢ௗ = Ȱ௙௢௢ௗ,௣௛௬ + Ȱ௙௢௢ௗ,௣௦௬ + Ȱ௙௢௢ௗ,௦௢௖

(8)

ܶ௘௡௧௘௥௧௔௜௡௠௘௡௧ = ܶ௘௡௧,௣௦௬ + ܶ௘௡௧,௦௢௖

(9)

In equations (8) and (9), Ȱ௙௢௢ௗ,௦௢௖ is the physiological space budgeted for food to meet sociological needs, ܶ௘௡௧,௣௦௬ the time for entertainment to meet psychological needs, and ܶ௘௡௧,௦௢௖ the time for entertainment to meet sociological needs. We will treat those needs with consumer goods and services as sociological needs. When physiological needs have been met with consumption, people will demand products that also meet their sociological needs. One example is ethical consumption (Adams and Raisborough 2010). Many consumers in developed countries buy fair trade goods at a price premium (Pelsmacker, Driesen, and Rayp 2005; Loureiro and Lotade 2005) because they become increasingly concerned with whether socially responsible firms produce the goods they purchase. They are willing to pay extra for socially responsible consumption (Webb, Mohr, and Harris 2008). Doing good can increase perceived well-being (Woodyard and Grable 2014). This ethical consciousness could also explain the popularity of the concept and practice of corporate social responsibility (CSR). In the past few decades, CSR has become a popular concept for business people and academics (Porter and Kramer 2006; Carroll 1991; Matten and Moon 2004). A special form of sociological need is the consumption for vanity. Vanity consumption satisfies a consumer’s excessive need to show good appearance, achievements, status, wealth, or generosity (Netemeyer, Burton, and Lichtenstein 1995; Workman and Lee 2011). Vanity is a psychological construct, and we list it among sociological needs because it relates to assumed perceptions by others. Marketing has exploited consumers’ vanity (Grilo, Shy, and Thisse 2001; Durvasula and Lysonski 2008). In contrast with vanity consumption, we may define functional consumption as satisfying some functional needs. Eating nutritious, healthy, and tasty foods and wearing comfortable quality clothes are functional consumption. Using fashion accessories and cosmetics that enhance a person’s appearance is also functional consumption. Using goods made from rare natural materials,

234

Chapter 4

which often cannot be distinguished from those made from artificial materials, and eating some special foods such as wild caviar (Raymakers 2007) are vanity consumption. Owning goods in quantities excessive of one’s capacity to use and spending beyond one’s affordability are also vanity consumption. Although vanity consumption plays an essential part in many consumer goods, the spending on rare gems and foods from endangered species, which can be afforded only by the super-rich, makes up only a small part of the aggregate consumption in the economy (Featherstone 2013; Hirschman and Steinberg 1990). The value of these luxury goods depends on the rarity of the particular natural resources and the artificial exclusivity of the brands with expensive marketing. It is impossible to achieve fast economic growth by meeting everybody’s vanity consumption needs. The high prices of designer clothes are artificial due to limited output, and their high prices are not sustainable if they are mass-produced. Therefore, growth in vanity consumption would not be sufficient for sustaining fast economic growth.

5.4. Spiritual, Self-Actualization, and Self-Transcendence Needs We define the sociological needs without consumption as spiritual needs and also include the desire for understanding unknowns and self-fulfillment in the spiritual needs. Promoting friendship, attention from others, and respect by others by doing good work and helping other people belong to activities for spiritual needs. These activities are also subject to time constraints. The difference between work and activities for meeting spiritual needs is that work is to earn an income for consumption, while activities for meeting spiritual needs are to earn respect, friendship, and love from others through personal achievements and good deeds. ܶௌ = ߬ௌଵ ܵଵ + ߬ௌଶ ܵଶ + ‫ ڮ‬+ ߬ௌ௡ ܵ௡

(10)

In equation (10), ܶௌ is the time budgeted for spiritual activities, and ߬ௌଵ is the time needed for performing one unit of spiritual activity 1, ܵଵ the quantity of spiritual activity 1, etc. People’s internal drive to meet their spiritual needs is more likely to create social wealth and public goods than consume them. In Maslow’s hierarchy, self-actualization is the next level above esteem, and self-transcendence is above self-actualization. For different people, selfactualization may mean different things. We can see that many government leaders try to hold onto their jobs for as long as possible, and entrepreneurs

Consumption, Satisfaction, and Life’s Purpose

235

continue to lead their firms even though their wealth has been more than enough to support their families for many generations. Many of these entrepreneurs promise to donate most of their wealth during their lifetime. At this level of human needs, work no longer causes disutility; on the contrary, work produces more utility than food, clothes, and entertainment can. Self-transcendence implies full spiritual awakening or liberation from egocentricity, which leads to working more for society and humanity. The need for self-actualization and self-transcendence requires time and possibly space to achieve, which reduces the maximum capacity to consume goods and services. This study will also treat self-actualization and selftranscendence as spiritual needs.

5.5. Methodological Needs Methodological needs are the desire for leisure, convenience, and ease of life. They are the requirements for facilities and skills that help meet other needs. On many occasions, especially in modern times, people need transport, furniture, and other facilities that enable them to consume goods and services to meet their physiological, psychological, and sociological needs. People buy cars, but probably only a few gain satisfactions from driving. Most drivers purchase a vehicle because they need it as a tool for mobility. People use public transport not because they enjoy being a bus, train, or airplane passenger. They travel because it is the means to achieve their objectives. The goods and services that meet methodological needs are extensions of the human body's sensory, locomotor, and cognitive systems. Cutlery is to help people’s fingers and teeth for better digestion and absorption of food in the digestive system. Bikes, cars, ships, and airplanes help the locomotor system improve mobility. Communications technology and equipment enhance human vision and hearing while writing and information technology (IT) supplement the brain's memory, learning, and analytical function. By the same logic, purchasing a television meets the methodological need for watching television broadcasting and using it as a monitor for electronic games rather than looking at the TV set. A TV set as a tool for managing news and entertainment is very different from a picture or a bottle of wine, which are consumable. A subscription to Internet access meets the methodological need for using various services on the Internet. Both physical and non-physical properties can be tools to enhance the function of some human body systems so that owners can better meet their physiological, psychological, and sociological needs. Medical care meets the methodological need for physical and mental health to facilitate other needs. Using tools needs time and space, so consumption for

236

Chapter 4

methodological needs is also subject to time and space constraints, which are part of the time and space required to consume goods for physiological, psychological, and sociological needs. When not used, tangible tools need storage space, which is part of the overall storage space. ܸ௦௧௢௥௘ = ܸ௦௧௢௥௘,௙௢௢ௗ + ܸ௦௧௢௥௘,௖௢௡௦௨௠௔௕௟௘௦ + ܸ௦௧௢௥௘,௧௢௢௟௦

(11)

In equation (11), ܸ௦௧௢௥௘,௙௢௢ௗ is the storage space for food, ܸ௦௧௢௥௘,௖௢௡௦௨௠௔௕௟௘௦ the storage space for other consumables, and ܸ௦௧௢௥௘,௧௢௢௟௦ the storage space for tools. Again, methodological and sociological needs are often blurred, and vanity consumption is often characterized by buying expensive equipment. People buy high-end or expensive brand cars to show their social status and belonging to a particular social group. Similarly, people buy furniture and many other goods. Furniture may be considered a tool to facilitate the consumption of other goods or as a decoration for meeting psychological and sociological needs. People may need to learn how to use a tool before they can use it safely and efficiently. Acquiring skills such as reading, writing, and driving can also be deemed methodological needs. Being able to read is a prerequisite for enjoying a book by oneself. Skill acquisition often costs the learner money and time and needs some space. Meeting methodological needs per se might give many individuals a sufficient incentive to acquire the targeted abilities. Many economists view meeting methodological needs regarding ability as investing in human capital. Learning skills for improving consumption can be considered an investment in human capital for home production, while buying physical or non-physical tools is an investment (in capital stock).

6. Utility Function and the Utility Maximization Problem Since goods and services need various amounts of time, space, and physiological space to consume, they have different levels of time, space, and physiological space costs. We have shown that human needs are segmented with different priorities. Therefore, an individual’s utility is a function of her/his consumption of goods and services for physiological, psychological, sociological, and methodological needs and her/his spiritual activities, ‫ݑ‬൫‫ܥ‬௣௛௬ , ‫ܥ‬௣௦௬ , ‫ܥ‬௦௢௖ , ‫ܥ‬௠௘௧ , ܵ൯ . A rational person bases their consumption and resource allocation decision on the lifetime utility maximization. The objective of an individual’s decisions is to maximize

Consumption, Satisfaction, and Life’s Purpose

237

expected utility, subject to monetary budget constraints, time constraints, external space constraints, and physiological constraints, Max ‫ݑ‬൫‫ܥ‬௣௛௬ , ‫ܥ‬௣௦௬ , ‫ܥ‬௦௢௖ , ‫ܥ‬௠௘௧ , ܵ൯

(12)

Subject to ‫ ܯ‬൒ ‫ܯ‬௣௛௬ + ‫ܯ‬௣௦௬ + ‫ܯ‬௦௢௖ + ‫ܯ‬௠௘௧

(13)

ܶ஼ௌ = ܶ௣௛௬ + ܶ௣௦௬ + ܶ௦௢௖ + ܶ௠௘௧ + ܶௌ

(14)

ܸ ൒ ܸ௣௛௬ + ܸ௣௦௬ + ܸ௦௢௖ + ܸ௠௘௧ + ܸௌ

(15)

Ȱ ൒ Ȱ௣௛௬ + Ȱ௣௦௬ + Ȱ௦௢௖ + Ȱ௠௘௧ + Ȱௌ

(16)

The above maximization problem is a one-period optimization and can be extended to a multi-period optimal intertemporal consumption and portfolio strategy problem. In the above function and equations, ‫ܥ‬௣௛௬ , ‫ܥ‬௣௦௬ , ‫ܥ‬௦௢௖ and ‫ܥ‬௠௘௧ are consumptions of goods and services for physiological, psychological, sociological, and methodological needs, respectively, and S activities for spiritual needs. They can be considered as vectors whose elements are determined by four factors: the monetary budget ‫ ܯ‬, time budget ܶ஼ௌ , space budget ܸ, and physiological space budget Ȱ, as well as individual preference. For simplicity, we classify sleep as a physiological need and leisure as a psychological need, which both take time and space but not money (except the opportunity cost of using the time for earning a wage). ‫ܯ‬௣௛௬ , ‫ܯ‬௣௦௬ , ‫ܯ‬௦௢௖ and ‫ܯ‬௠௘௧ are the money spent on the four types of consumption needed ܶ௣௛௬ , ܶ௣௦௬ , ܶ௦௢௖ , ܶ௠௘௧ and ܶௌ are the time spent on the five types of need (including the time for spiritual needs ܶௌ ); ܸ௣௛௬ , ܸ௣௦௬ , ܸ௦௢௖ , ܸ௠௘௧ and ܸௌ are the external space required for the five types of need; and Ȱ௣௛௬ , Ȱ௣௦௬ , Ȱ௦௢௖ , Ȱ௠௘௧ and Ȱௌ are the physiological space required for the five types of need. In equation (13), there is no money spent on spiritual activities because we define spiritual activities as those without the consumption of goods and services. Those spiritual activities requiring personal consumption are defined as consumption for meeting sociological needs. The physiological space Ȱ௣௛௬ , Ȱ௣௦௬ , Ȱ௦௢௖ , Ȱ௠௘௧ and Ȱௌ as well as Ȱ can also be considered vectors since there are many mutually independent physiological constraints. The time budget ܶ஼ௌ includes only time allocated for consumption and spiritual activities, so the monetary budget becomes

Chapter 4

238

exogenously determined. If the time for work is included in the maximization problem, the monetary budget ‫ ܯ‬will become endogenous. ܶ = ܶ௣௛௬ + ܶ௣௦௬ + ܶ௦௢௖ + ܶ௠௘௧ + ܶௌ + ܶ௪௢௥௞ ‫ܶ߱ = ܯ‬௪௢௥௞

(17) (18)

In equations (17) and (18), T is the time endowment, ܶ௪௢௥௞ the time spent on work to earn income M and ߱ the wage rate per unit of time. The constraint equations show that when consumption is small (and the income is low), the time, space, and physiological constraints would have little impact. The utility maximization problem reduces to the maximization under only the monetary budget constraint. When the income is high and consumption large, these constraints will work similarly to the monetary budget constraint. We have not specified the utility function form, which is unlikely to be a simple function, but we might approximate it with relatively simple procedures. One possibility is that the utilities derived from meeting the five types of needs are additively separable: ି ൯‫ݑ‬൫‫ܥ‬௣௦௬ ൯ + ‫ݑ‬൫‫ܥ‬௣௛௬ , ‫ܥ‬௣௦௬ , ‫ܥ‬௦௢௖ , ‫ܥ‬௠௘௧ , ܵ൯ = ‫ݑ‬൫‫ܥ‬௣௛௬ ൯ + ‫ܦ‬ଵ ൫‫ܥ‬௣௛௬ െ ‫ܥ‬௣௛௬ ି ି ି ‫ܦ‬ଵ ൫‫ܥ‬௣௛௬ െ ‫ܥ‬௣௛௬ ൯‫ݑ‬ሺ‫ܥ‬௠௘௧ ሻ + ‫ܦ‬ଵ ൫‫ܥ‬௣௛௬ െ ‫ܥ‬௣௛௬ ൯‫ܦ‬ଶ ൫‫ܥ‬௣௦௬ െ ‫ܥ‬௣௦௬ ൯ሾ‫ݑ‬ሺ‫ܥ‬௦௢௖ ሻ + ‫ݑ‬ሺܵሻሿ

(19) In equation (19), ‫ݑ‬൫‫ܥ‬௣௛௬ ൯ , ‫ݑ‬൫‫ܥ‬௣௦௬ ൯ , ‫ݑ‬ሺ‫ܥ‬௠௘௧ ሻ , ‫ݑ‬ሺ‫ܥ‬௦௢௖ ሻ and ‫ݑ‬ሺܵሻ are utilities derived from consumption for physiological, psychological, methodological, and sociological needs and spiritual activities, ି is the minimum physiological consumption required for respectively. ‫ܥ‬௣௛௬ ି the minimum psychological consumption before survival, and ‫ܥ‬௣௦௬ ି ൯ is a binary function with a value sociological needs arise. ‫ܦ‬ଵ ൫‫ܥ‬௣௛௬ െ ‫ܥ‬௣௛௬ ି ି of 1 if ൫‫ܥ‬௣௛௬ െ ‫ܥ‬௣௛௬ ൯ ൒ 0 and a value of 0 if ൫‫ܥ‬௣௛௬ െ ‫ܥ‬௣௛௬ ൯ (remaining) ME percentages x ȴMPG%

Fig.4-4 An illustration of why productivity growth has slowed since the 1970s. The increase in service sector employment with low productivity growth is larger than that of manufacturing employment proportion and manufacturing productivity growth. MPG% and SPG% are the current (or previous) period manufacturing and service productivity growth rates, respectively; ǻMPG% is the extra manufacturing growth rate in the next (or current) period.

7.3. The Role of Information Technology and Asset Price One of the most notable features in consumption growth since the 1970s is the transformation of the computer into a consumer product and the associated rise of ICT, especially mobile computing and smartphones. However, IT has yet to bring in functionally or mechanistically novel products (Table 4-3), whereas steam engines and electricity made them possible and created new production processes. The term functionally or categorically novel means a property that satisfies or extends a particular function of the human body from scratch. For example, horse-drawn carriages extend the human ability to move on land, and boats extend the ability to move on the water. The term mechanistically novel means a property that satisfies or extends a particular function of the human body with different mechanisms from the existing products. Steam enginepowered trains and internal combustion engine-powered cars are mechanistically novel compared with horse-drawn carriages.

Consumption, Satisfaction, and Life’s Purpose

243

Table 4-3 Types of novel products

Functionally Novel

Feature A new category of products extending one function of the human body

Mechanistically Novel

A new mechanism for an existing category of products

Characteristically Novel

An improvement of an existing mechanism for an existing category

Impact Create a new sector with new infrastructure and new demand Create a new subsector with additional new infrastructure and increase overall demand in the sector Replace the existing products with more customer satisfaction

Example Animaldrawn carriage

Railway train

High-speed train

Besides AI and automation, the current IT typified by the Internet has two major economic impacts. One is to let some traditional goods reach consumers who could not consume those goods previously because of information deficiency. The other is to destroy or undermine businesses that rely on information deficiency to be sustainable. The former includes the emergence of online shops using platforms such as eBay, Alibaba/Taobao, and Amazon. The latter includes the shutdown of many brick-and-mortar shops and the decline of traditional media. In both cases, competition based on transparent price information will drive the economic profits to zero and concentrate production to the most efficient firms or individuals (the winner takes all). Producers and retailers in a competitive market will make no economic profits, while Internet platforms and advertisement providers make huge profits through their near-monopoly power. The concentration of wealth and income exacerbated by the success of software and Internet firms is one factor that pushes up the asset price. Information-related products usually have zero marginal costs, meaning their prices should be zero according to economics. However, they often charge high fees because laws protect intellectual property rights. Those firms make extraordinary revenues and profits with little or no additional capital and labor inputs. Since goods with zero marginal cost cannot be a

244

Chapter 4

store of value, entrepreneurs and capital owners who made millions, even billions of dollars, have to store their wealth in capital assets and physical goods such as houses, pushing up asset prices. Concentrating income and wealth to an ever-smaller proportion of the population leads to more wealth being stored in assets and further increasing asset prices.

7.4. The Potential Future Economic Growth and Employment Fast economic growth might be initiated by new processes that produce large quantities of the previously undersupplied existing products more cheaply, (process innovation) or by functionally or mechanistically novel products with their production processes (product innovation and process innovation). The First Industrial Revolution began with new technologies in the textile industry and iron making to produce existing goods in much larger quantities at much lower costs. The rapid economic growth caused by new textile and iron-making processes stimulated the invention of functionally or mechanistically novel products, further promoting economic growth. The Second Industrial Revolution brought even more functionally novel products for consumers to meet their physiological, psychological, and sociological needs. The economic miracle created by the industrial revolutions arose from the previously unmet needs in the quantity of existing goods and the variety of newly invented functionally and mechanistically novel products. Our present framework suggests that employment will decrease sharply in the future because of rapid productivity growth in the service sector due to the wide application of AI and robots and the market saturation of manufactured goods. Rapid economic growth depends on the aggregate demand growth in quantities due to either lower prices enabled by new production technologies or new categories of products invented to satisfy previously unmet human needs. Suppose there is no functionally or mechanistically novel product introduced into the economy. In that case, the economy can still grow slowly with characteristically novel products (Table 3), improving the quality of existing products. Different versions of mobile phones are examples of characteristically novel products. Functionally novel products will be rare in the future. Therefore, future economic growth will depend more on the innovation of characteristically novel products than functionally or mechanistically novel products. There will likely be no increase in the quantity of goods and services for consumption per capita, but quality improvement will continue. Improving the quality of products will not lead

Consumption, Satisfaction, and Life’s Purpose

245

to a net increase in employment. As more and more machines become smart and need no or only minimum human attention, there will be fewer and fewer workers being employed. Eventually, there will be few human workers in the economy. As total consumption has its limit, with the progress of technologies and the increase of productivity, the number of jobs will decrease. At some point, working opportunities become so rare that people compete or bid for a period of working experience. Then work itself could become a consumer product. People pay to work. Human beings can always find something to do for themselves and by themselves. The working tasks as rare consumption goods are related to public services and producing goods in large quantities. At that time, most people will work as freelance artists doing whatever they want, but only a few can become successful superstar artists. Most people will live on the minimum guaranteed income for a decent and respectable life and find satisfaction in doing things they enjoy.

7.5. Social Welfare and Politics Because of time, space, physiological constraints, and segmented human needs, individual consumption capacity is finite and employment will decrease as AI and robotics progress. We have defined the historical period when employed and self-employed working-age households (i.e., those with at least one employed or self-employed working-age member) are fewer than half of all working-age households due to the widespread application of robots and AI, as the Robotic Age. When employed and self-employed households are fewer than half of all working-age households, the balance of power will tip toward those without a job or many capital assets. People without a job could become more politically active. Politicians and political parties advocating high-level social welfare will dominate parliaments and governments in all democracies. Every member of society will get a guaranteed income sufficient for living a respectable life. With a guaranteed income, everybody can work for their spiritual pursuits, self-actualization, and self-transcendence. Society needs to establish a political infrastructure appropriate for the Robotic Age progressively.

8. Summary Both human needs and constraints on them determine rational consumer choice. Human needs include physiological, psychological, sociological, spiritual, and methodological ones. There are limits on the maximum

246

Chapter 4

individual capacity of functional consumption because of time, space, and physiological constraints. ICT increases people’s access to information, knowledge, and entertainment but does not increase people’s time endowment or change human physiology. When the human-replacing technologies in the service sector mature, they will lead to faster labor productivity growth. With limits on the maximum individual capacity to consume, technological progress will inevitably lead to decreasing employment of human workers until, eventually, very few humans are employed.

References Adams, Matthew, and Jayne Raisborough. 2010. “Making a difference: ethical consumption and the everyday.” British Journal of Sociology 61 (2):256-274. Batson, C. Daniel. 1987. “Prosocial motivation: is it ever truly altruistic?” Advances in Experimental Social Psychology 20:65-122. Cain, Áine. 2017a. “A day in the life of the richest person in the world, Jeff Bezos — who made $6.44 billion in one day and still washes the dishes after dinner.” Business Insider, Last Modified 6 March 2018, accessed 7 June 2018. http://www.businessinsider.com/jeff-bezos-daily-routine2017-7. Cain, Áine. 2017b. “Inside the daily routine of 87-year-old Warren Buffett, who loves McDonald's, spends 80% of his workday reading, and unwinds by playing the ukulele.” Business Insider, accessed 7 June 2008. http://www.businessinsider.com/warren-buffett-daily-routine-2017-8/. Cain, Áine. 2017c. “Inside the daily routine of billionaire Bill Gates, who loves cheeseburgers, tours missile silos, and washes the dishes every night.” Business Insider, accessed 7 June 2018. http://www.businessinsider.com/bill-gates-daily-routine-2017-8. Campbell, Byron A., and George A. Cicala. 1962. “Studies of water deprivation in rats as a function of age.” Journal of Comparative and Physiological Psychology 55 (5):763-768. Campbell, Byron A., Robert Teghtsoonian, and Ruth Ann Williams. 1961. “Activity, weight loss, and survival time of food-deprived rats as a function of age.” Journal of Comparative and Physiological Psychology 54 (2):216-219. Carroll, Archie B. 1991. “The pyramid of corporate social responsibility: Toward the moral management of organizational stakeholders.” Business Horizons 34 (4):39-48.

Consumption, Satisfaction, and Life’s Purpose

247

Carroll, Christopher D. 1997. “Buffer-stock saving and the life cycle/permanent income hypothesis.” Quarterly Journal of Economics 112 (1): 1–55. Cole, Charles L, and Graham B Spaniard. 1974. “Comarital mateဨsharing and family stability.” Journal of Sex Research 10 (1):21-31. Danovaro, Roberto, Antonio Dell'Anno, Antonio Pusceddu, Cristina Gambi, Iben Heiner, and Reinhardt Møbjerg Kristensen. 2010. “The first metazoa living in permanently anoxic conditions.” BMC Biology 8:30. doi: 10.1186/1741-7007-8-30. David, Paul A. 1990. “The dynamo and the computer: an historical perspective on the modern productivity paradox.” American Economic Review 80 (2):355-361. Davis, Caroline, Claire Curtis, Robert D. Levitan, Jacqueline C. Carter, Allan S. Kaplan, and James L. Kennedy. 2011. “Evidence that 'food addiction' is a valid phenotype of obesity.” Appetite 57 (3):711-717. Deaton, Angus S. 1991. “Saving and liquidity constraints.” Econometrica 59 (5):1221-1248. Diewert, W. Erwin, and Kevin J. Fox. 1999. “Can measurement error explain the productivity paradox?” Canadian Journal of Economics 32 (2):251-280. Dufort, Robert H. 1963. “Weight loss in rats continuously deprived of food, water, and both food and water.” Psychological Reports 12 (2):307-312. Durvasula, Srinivas, and Steven Lysonski. 2008. “A doubleဨedged sword: understanding vanity across cultures.” Journal of Consumer Marketing 25 (4):230–244. Eisenberg, Nancy, and Paul A. Miller. 1987. “The relation of empathy to prosocial and related behaviors.” Psychological Bulletin 101 (1):91-119. Enoka, Roger M., and Jacques Duchateau. 2008. “Muscle fatigue: what, why and how it influences muscle function.” Journal of Physiology 586 (1):11-23. doi: 10.1113/jphysiol.2007.139477. Featherstone, Mike. 2013. “The rich and the super-rich: mobility, consumption and luxury lifestyles.” In Consumer Culture, Modernity and Identity, edited by Nita Mathur, 3-44 New Delhi: SAGE Publications. Ferguson, Ann. 1984. “Sex war: The debate between radical and libertarian feminists.” Signs: Journal of Women in Culture and Society 10 (1):106112. Fiore, Ann Marie. 2006. “Effects of experiential pleasure from a catalogue environment on approach responses toward fashion apparel.” Journal of Fashion Marketing and Management 6 (2):122-133. Fisher, Sue. 1994. “Identifying video game addiction in children and adolescents.” Addictive Behaviors 19 (5):545-553.

248

Chapter 4

Friedman, Milton. 1957. A theory of the consumption function. Princeton: Princeton University Press. Gambrel, Patrick A., and Rebecca Cianci. 2003. “Maslow's hierarchy of needs: Does it apply in a collectivist culture?” Journal of Applied Management and Entrepreneurship 8 (Apr):143-161. Gatti, Domenico Delli, Mauro Gallegati, Bruce C. Greenwald, Alberto Russo, and Joseph E. Stiglitz. 2012. “Mobility constraints, productivity trends, and extended crises.” Journal of Economic Behavior and Organization 83 (3):375-393. Gearhardt, Ashley N., Sonja Yokum, Patrick T. Orr, Eric Stice, William R. Corbin, and Kelly D. Brownell. 2011. “Neural correlates of food addiction.” Archives of General Psychiatry 68 (8):808-816. Goebel, Barbara L, and Delores R Brown. 1981. “Age differences in motivation related to Maslow's need hierarchy.” Developmental Psychology 17 (6):809. Goolsbee, Austan, and Peter Klenow. 2006. “Valuing consumer products by the time spent using them: an application to the internet.” Social Science Electronic Publishing 96 (2):108-113. Gordon, Robert J. 2000. “Does the “new economy” measure up to the great inventions of the past?” The Journal of Economic Perspectives 14 (4):49-74. Greenwood, Jeremy, and Nezih Guner. 2010. “Social change: The sexual revolution.” International Economic Review 51 (4):893-923. Grilo, Isabel, Oz Shy, and Jacques François Thisse. 2001. “Price competition when consumer behavior is characterized by conformity or vanity.” Journal of Public Economics 80 (3):385-408. Gronau, Reuben. 1976. “Leisure, home production, and work--the theory of the allocation of time revisited.” Journal of Political Economy 85 (6):1099-1123. Grundy, Scott M., James I. Cleeman, Stephen R. Daniels, Karen A. Donato, Robert H. Eckel, Barry A. Franklin, David J. Gordon, Ronald M. Krauss, Peter J. Savage, Sidney C. Smith, John A. Spertus, and Fernando Costa. 2005. “Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute scientific statement: Executive Summary.” Critical Pathways in Cardiology 4 (4):198-203. Gupta, Shruti, and Pratish Srivastav. 2016. “An exploratory investigation of aspirational consumption at the bottom of the pyramid.” Journal of International Consumer Marketing 28 (1):2-15. Hall, John E. 2015. Guyton and Hall Textbook of Medical Physiology. 13 ed. Philadelphia, PA: Saunders.

Consumption, Satisfaction, and Life’s Purpose

249

Henne, Douglas, and Edwin A Locke. 1985. “Job dissatisfaction: what are the consequences?” International Journal of Psychology 20 (2):221-240. Hindmarch, Ian. 2004. “Psychomotor function and psychoactive drugs.” British Journal of Clinical Pharmacology 58 (7): S720-S740. Hirschman, Elizabeth C., and Saul Steinberg. 1990. “Consumption styles of the rich and famous: the semiology of Saul.” Advances in Consumer Research 17:850-855. Hofstede, Geert. 1984. “The Cultural Relativity of the Quality of Life Concept.” Academy of Management Review 9 (3):389-398. Keynes, John Maynard. 1936. The General Theory of Employment, Interest Rate and Money. London: Palgrave Macmillan. Reprint, 2007. Kringelbach, Morten L., Peter Vuust, and John Geake. 2008. “The pleasure of reading.” Interdisciplinary Science Reviews 33 (4):321-335. Lehrer, Keith Akiva. 2014. “Sustainability and Its Constraints in the Global Tourist Industry: Exploring the Dark Side in the Growth of International Tourism.” Journal of Tourism and Hospitality Management 2 (5):183197. Locke, Edwin A, David Sirota, and Alan D Wolfson. 1976. “An experimental case study of the successes and failures of job enrichment in a government agency.” Journal of Applied Psychology 61 (6):701. Marcora, Samuele M., Walter Staiano, and Victoria Manning. 2009. “Mental fatigue impairs physical performance in humans.” Journal of Applied Physiology 106 (3):857-64. doi: 10.1152/japplphysiol.91324.2008. Martin, Kelly D., and Ronald Paul Hill. 2012. “Life satisfaction, selfdetermination, and consumption adequacy at the bottom of the pyramid.” Journal of Consumer Research 38 (6):1155-1168. Maslow, Abraham H. 1943. “A theory of human motivation.” Psychological Review 50 (4):370–396. Maslow, Abraham H. 1954. Motivation and personality. New York, NY: Harper. Maslow, Abraham H. 1969. “The farther reaches of human nature.” Journal of Transpersonal Psychology 1 (1):1-9. Matten, Dirk, and Jeremy Moon. 2004. “Corporate social responsibility.” Journal of Business Ethics 54 (4):323-337. Modigliani, Franco, and Richard Brumberg. 1954. “Utility analysis and the consumption function: an interpretation of cross-section data.” In PostKeynesian Economics, edited by Kenneth K. Kurihara, 388-436. New Brunswick, New Jersey: Rutgers University Press. Mulvey, Laura. 1975. “Visual pleasure and narrative cinema.” Screen 16 (3):6-18.

250

Chapter 4

Netemeyer, Richard G., Scot Burton, and Donald R. Lichtenstein. 1995. “Trait aspects of vanity: Measurement and relevance to consumer behavior.” Journal of Consumer Research 21 (4):612-626. Pelchat, Marcia Levin. 2009. “Food addiction in humans.” Journal of Nutrition 139 (3):620-2. Perlovsky, Leonid, Arnaud Cabanac, Marie-Claude Bonniot-Cabanac, and Michel Cabanac. 2013. “Mozart effect, cognitive dissonance, and the pleasure of music.” Behavioural Brain Research 244:9-14. Phaneuf, Daniel J. 2011. “Can consumption of convenience products reveal the opportunity cost of time?” Economics Letters 113 (1):92-95. Pinder, Craig C. 1998. Work Motivation in Organizational Behavior. Upper Saddle River, New Jersey: Prentice-Hall. Porter, Michael E., and Mark R. Kramer. 2006. “Strategy and society: the link between competitive advantage and corporate social responsibility.” Harvard Business Review 84 (12):78-92. Raymakers, Caroline. 2007. “CITES, the convention on international trade in endangered species of wild.” Journal of Applied Ichthyology 22 (s1):53-65. Robinson, Ira, Ken Ziss, Bill Ganza, and Stuart Katz. 1991. “Twenty years of the sexual revolution, 1965-1985: An update.” Journal of Marriage and the Family:216-220. Rooij, Antonius J. Van, Tim M. Schoenmakers, Ad A. Vermulst, Regina J.J.M. Van Den Eijnden, and Dike Van De Mheen. 2015. “Online video game addiction: identification of addicted adolescent gamers.” Addiction 106 (1):205-212. Rubin, Roger H. 2001. “Alternative lifestyles revisited, or whatever happened to swingers, group marriages, and communes?” Journal of Family Issues 22 (6):711-726. Rye, B. J., and Glenn J. Meaney. 2007. “The pursuit of sexual pleasure.” Sexuality and Culture 11 (1):28-51. Salimpoor, Valorie N., Mitchel Benovoy, Kevin Larcher, Alain Dagher, and Robert J. Zatorre. 2011. “Anatomically distinct dopamine release during anticipation and experience of peak emotion to music.” Nature Neuroscience 14 (2):257-262. Schleidt, Margret, Peter Neumann, and Harumi Morishita. 1988. “Pleasure and disgust: memories and associations of pleasant and unpleasant odours in Germany and Japan.” Chemical Senses 13 (2):279-293. Small, Dana M., Robert J. Zatorre, Alain Dagher, Alan C. Evans, and Marilyn Jones-Gotman. 2001. “Changes in brain activity related to eating chocolate: from pleasure to aversion.” Brain 124 (9):1720-1733.

Consumption, Satisfaction, and Life’s Purpose

251

Szilagyi, Andrew D., and Marc J. Wallace. 1990. Organizational behavior and performance. Glenview, Illinois: Scott Foresman & Co. Tang, Thomas Li-Ping, and Abdul H Safwat Ibrahim. 1998. “Importance of human needs during retrospective peacetime and the Persian Gulf War: Mideastern employees.” International Journal of Stress Management 5 (1):25-37. Tang, Thomas Li-Ping, and W Beryl West. 1997. “The importance of human needs during peacetime, retrospective peacetime, and the Persian Gulf War.” International Journal of Stress Management 4 (1):47-62. Tay, Louis, and Ed Diener. 2011. “Needs and subjective well-being around the world.” Journal of Personality and Social Psychology 101 (2):354365. Triplett, Jack E. 1999. “The Solow productivity paradox: what do computers do to productivity?” Canadian Journal of Economics/Revue Canadienne D`Economique 32 (32):309-334. Tuttle, Brad. 2016. 12 leaders who took $1 salaries - and how they fared. 2022. Accessed October 25, 2022. https://money.com/donald-trump-1salary-president-ceo. Villarica, Hans. 2011. “Maslow 2.0: A new and improved recipe for happiness.” The Atlantic 308 (1). Wahba, Mahmoud A., and Lawrence G. Bridwell. 1976. “Maslow reconsidered: A review of research on the need hierarchy theory.” Organizational Behavior and Human Performance 15 (2):212-240. Webb, Deborah J., Lois A. Mohr, and Katherine E. Harris. 2008. “A reexamination of socially responsible consumption and its measurement.” Journal of Business Research 61 (2):91-98. Weiner, Bernard. 1985. “An attributional theory of achievement motivation and emotion.” Psychological Review 92 (4):548-573. Wise, R. A. 1996. “Addictive drugs and brain stimulation reward.” Annual Review of Neuroscience 19 (1):319-340. Woodyard, Ann, and John Grable. 2014. “Doing Good and Feeling Well: Exploring the Relationship Between Charitable Activity and Perceived Personal Wellness.” Voluntas International Journal of Voluntary and Nonprofit Organizations 25 (4):905-928. Workman, Jane E., and Seung Hee Lee. 2011. “Vanity and public selfconsciousness: a comparison of fashion consumer groups and gender.” International Journal of Consumer Studies 35 (3):307–315.

CHAPTER 5 PRODUCTION IN THE ROBOTIC AGE

How a society organizes its production processes and how the members of society participate in the production processes largely determine how society functions as a community of human beings. As discussed earlier, human society has evolved from the Manual Age to the Machine Age, where human (and animal) power is replaced by machine power. Humanity is now entering the Robotic Age, in which human intelligence used in production is being replaced by machine intelligence. Because of the time, space, and physiological capacity constraints, there is a limit to an individual’s non-vanity consumption. The consumption limit will constrain, in turn, the jobs available for human workers, especially with progress in artificial intelligence (AI) and robotics. As reviewed in Chapter 3, industrial robots have replaced many human workers in several sectors, such as the automobile industry. AI systems and robots are further externalization, expansion, scale-up, separation, and alienation of human body functions and production processes, as discussed in Chapter 1. However, in contrast with technologies in the Machine Age, which tied human workers to some specific jobs, the incoming technologies will liberate humans from the production processes. AI systems and robots will be able to replace humans and perform complete job systems in the Robotic Age. For AI systems and robots to replace human workers, they must satisfy three criteria: 1) cost-effective: they should be cheaper than human workers for the same job; 2) comparable quality: the products made by them have quality better than or comparable to those made by human workers; 3) controllable: they can be controlled by humanity. In this chapter, we will examine the Robotic Age production factors, types of products, and producers and analyze the trend of human-replacing technology in different sectors.

1. Production Factors in the Robotic Age In economics, resources can be divided into three categories: natural resources, capital stock, and labor. Generally, any production needs these

Production in the Robotic Age

253

three types of resources. Natural resources are endowed by nature to a particular region or country. People cannot change where natural resources are located, but they can make better or more efficient use of them. Capital stock and labor are the resources over which people have more control. In traditional economics, factors in production are often simplified into labor and capital. In the simplest form of a production theory, the output Q of a firm or an economy is determined by the quantities of the labor L and the capital K it has employed. ܳ = ܳ(‫ܮ‬, ‫)ܭ‬

(1)

In a straightforward interpretation, labor is the number of workers, and capital is the number of machines and materials used in production. The machines contain and embody the technologies at the time. The workers need to be able to operate the machines. In the early days of human history, labor was more important, and there was little capital stock to speak of. As production technology develops, more capital is needed in the production process.

1.1. Capital Capital is the means of production, including the materials to work with, i.e., the non-human assets used in production. Capital can be further divided into tangible and intangible assets according to the capital item being physical or non-physical, although non-physical capital usually needs a physical carrier. It can also be divided into capital items originating from human efforts and capital items endowed by nature (natural resources). However, human-made capital items are often built upon natural resources. Capital items arising from human actions are called capital stocks (machinery, buildings, etc., made by humans). Table 5-1 summarizes the capital types for different ages. Human-made capital is not all the same for different ages. The intangible assets, such as patents and franchises, appeared in the late Manual Age. A key feature of economic growth over three million years of human history is the accumulation of capital stocks. It started with some simple tools to facilitate hunting and farming. There will be more capital stocks or better machinery in the Robotic Age than in previous eras. Moreover, capital stocks include AI systems and intelligent robots that can automate complete job systems. A complete job system means all roles in making a final finished product; it is not just one of many specialized roles in creating a complete product. For example, although a worker at an assembly line for

Chapter 5

254

cars used to do just welding, which can be replaced entirely with a welding robot, we do not think the welding robot has automated a complete job system because it cannot perform other roles in making a car. AI and intelligent robots that can automate complete job systems will challenge the traditional dichotomy of labor and capital. Table 5-1 Types of capital in different ages Human-made assets

Manual Age

Machine Age

Tangible Buildings, tools, farm animals, farm infrastructure

Intangible Franchises, patents, trademarks

Machines, public utilities, buildings, tools, farm animals, farm infrastructure

Franchises, patents, trademarks

Natural resource

Human Resource*

Land and its fauna and flora, minerals, water, water power, wind power

Uneducated inexperienced human ability, work experience, expertise from training, ability from education Uneducated inexperienced human ability, work experience, expertise from training, ability from education Uneducated inexperienced human ability, work experience, expertise from training, ability from education

Land and its fauna and flora, minerals, water, water power, wind power

Multifunctional Franchises, Land and its robots, public patents, fauna and utilities, trademarks, flora, Robotic machines, software, AI minerals, Age buildings, systems water, water tools, farm power, wind animals, farm power infrastructure * Human resources are included here for comparison.

1.2. Labor In classical economics, the labor theory of value considers labor as the source of value contained in goods. In his Wealth of Nations, Adam Smith emphasized the importance of labor in determining the value of goods (Smith 2010). Now that we are at the dawn of the Robotic Age, with the benefits of hindsight, we can see that production needs raw materials, tools,

Production in the Robotic Age

255

energy, and the intelligence that guides production processes. In the Manual Age, labor, the driving force of a production process and credited with the source of value by classical economists, provides energy and intelligence (Fig.5-1). In the Machine Age, although power is not mainly supplied by labor, labor is still the intelligence provider that guides the production processes, and some manual power is still required to switch on machines, adjust their performance, service them, and fix them when they break down. Automation reduces the contribution from labor to the necessary intelligence for most production processes, but labor (human input) is still required at some stages of a complete production process. Although modern mainstream economics no longer considers labor as the sole source of value and adopts the subjective theory of value or exchange theory of value, labor is still considered an irreplaceable but partially substitutable production factor.

Manual Power

Products Tools/ Machines

Labor Intelligence

Capital

Materials

Fig.5-1 Production factors in the Manual and Machine Ages include materials, tools/machines, and labor which consists of manual power and intelligence.

In the Robotic Age, the intelligence needed to guide a production or service process can be provided by AI, which removes the last necessity derived from the human being in production and service processes, hence to a large extent, human labor is no longer an irreplaceable production factor and nor is it the source of the value contained in goods. Production technology and intelligence are embodied in the tools, including robots and AI systems, as well as the materials to be worked on. The product designs, production procedures, and control processes are all encompassed by human intelligence. If we stick to the labor theory of value, we will attribute all value creations to human intelligence; otherwise, all four factors contribute

256

Chapter 5

to value creation, and we can investigate their relative contribution based on marginalism. Labor is the capacity of human workers to produce goods and services with tangible and intangible means of production. With the introduction of “human capital” as a key concept in economics, the meaning of labor becomes a bit vague. Should labor include the workers' intelligence, knowledge, and skills involved in the production? If it does not have the workers' intelligence, knowledge, and abilities, should a performer of labor be considered a biological machine with zero intelligence? The relationship between labor and human capital may become irrelevant in the Robotic Age. The three main factors of production, labor, human-made capital, and natural resources, are affected differently by automation and robotics. Natural resources would play more or less the same role as in the Machine Age; human-made capital would be the incarnation of advanced technologies, and labor will be the most affected factor. When robots and AI systems can fully automate complete job systems, it is difficult to say that labor is still a production factor. Some economists have argued that technological progress will create more new jobs as in the past, and the concern with technological unemployment is the lump of labor fallacy (Walker 2007; Brynjolfsson and McAfee 2015). Although this had been the case in the past, as we have analyzed in the preceding chapters, it will be different this time primarily because almost all human sensory modalities have enough goods to satisfy. AI and robots are replacing human workers in production processes. Of course, some human labor may still be needed at the early or even the mature stage of the Robotic Age. There is a high probability that AI systems or robots can perform entirely almost all jobs in the future. A vast factory could well be operated by only one or two managers who keep an eye on the performance of robots. At that stage, labor provided by one or two managers may no longer have the same meaning as described by traditional economics.

1.3. Human Capital: A Critique Since the 1950s, the term “human capital” has become popular in economics. It refers to the stock of knowledge, habits, and social and personality attributes embodied in the ability to perform labor to produce economic value. Jacob Mincer's article "Investment in Human Capital and Personal Income Distribution" in the Journal of Political Economy in 1958 is probably the first use of the term in the modern neoclassical economic literature (Mincer 1958). Gary Becker further popularized the term with his

Production in the Robotic Age

257

book Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, first published in 1964 (Becker 2009). Adam Smith, in 1776, already defined acquired and useful abilities as one of the four types of fixed capital: instruments of trade, buildings as the means of procuring revenue, improvements of land, and the acquired and useful abilities of all the inhabitants or members of the society (Smith 2010). There are controversies on whether human capital is an appropriate concept. Unlike physical or intangible capital, a person’s human capital cannot be owned by another person or organization except when slavery is permitted (Piketty 2014). What is the relationship between labor and human capital? Adam Smith commented in 1776 that “[t]he improved dexterity of a workman may be considered in the same light as a machine or instrument of trade which facilitates and abridges labor, and which, though it costs a certain expense, repays that expense with a profit” (Smith 2010). From this comment, labor would be viewed as the capacity to work before training, learning, and doing the job. Human capital is the additional capacity to produce economic value above the ability of an inexperienced and untrained worker. A division of labor will facilitate the accumulation of experience and expertise and reduce the time needed to train a person to perform their job. Human capital includes the knowledge of how to operate the equipment. In development economics, human capital is often measured by the average years of schooling, the number of university graduates, or their proportion in the population. It might be obscure whether a high human capital measured as such is the cause or the consequence of sustained economic growth. One fact is that China, during 1978–2008, had much less human capital per capita than the United States, although its economy grew much faster. Training employees to operate a machine is relatively easier than teaching them to pass college courses. A few days of practice using a device is more effective than a few years of college study to train a machine operator. Most administrative and industrial positions do not need university degrees. Only academic, professional (like medical doctors), and industrial R&D jobs require advanced university education. On-job training is more cost-effective for society, even if it is not for companies. More education, of course, can usually let job candidates have a high ability to master the operational skill of a sophisticated machine. However, secondary school graduates need only a few weeks or months of training to use and operate most machines. People with bachelor’s, master’s, or doctorate degrees might not have many advantages. This indicates that

258

Chapter 5

technology rather than human capital, as measured by years of schooling, determines economic growth. More advanced and sophisticated machines need less training and fewer years of education for proper operation. In molecular biology research, DNA sequencing was a task for doctorate students or doctoral researchers in the 1970s and 1980s, but a secondary school leaver can do it well with automated DNA sequencers. The technology used in their job determines a worker’s productivity and income, which usually does not require what they have studied during their schooling years beyond secondary school. University education has dramatically expanded in China since the mid-1990s, and the current annual graduate number is about 30 to 40 times that around 1980. Many graduates work as shop assistants, supermarket checkout cashiers, or takeaway deliverers nowadays in China. The causal relationship between human capital (as measured by years of schooling) and economic growth level found or assumed by researchers could be a reversed one, i.e., economic growth increased average years of schooling. Expertise and experiences acquired through work or special education are more effective than a general-purpose university education. Expertise and experiences might be capital for the person who owns them or for the nation, but for firms, they are just features of labor worthy of wage premiums. With AI systems and robots taking over responsibilities previously done by professionals, human capital will become irrelevant in production processes. It costs almost nothing to copy the software of an AI system, while training another human specialist costs a lot more. In the Robotic Age, the value of human capital will depreciate in the production processes because human intervention is optional when AI systems and robots take control.

1.4. Production Technology and Productivity In a modern economy, production technology is mainly embedded in the equipment and machinery (the capital stock) used in production. In the Machine Age, human beings with knowledge of how to use machines still play a crucial role in realizing the potential of technology. In neoclassical growth theories, technological progress is essential for raising the economy above its current dynamic equilibrium. Steam engines, internal combustion engines, and electrical motors have fundamentally transformed production and the economy. With AI systems and robots that automate complete job systems, labor and human capital will have little or no role in production.

Production in the Robotic Age

259

An economy's development level or living standard is usually measured by labor productivity. Labor productivity is the output or value created by one worker during one unit of time. Two types of labor productivity are often used: labor productivity per annum and labor productivity per hour. Labor productivity per annum is the product of labor productivity per hour and working hours per annum. However, labor productivity will lose its original significance if AI systems and robots replace humans. The importance of the service sector in employing since World War II illustrates that automation and the application of robots are far more advanced in manufacturing than in the service sector. The high productivity in manufacturing (and mechanized agriculture) supported the development of the service sector and created a surplus labor force to be employed by the service sector. Without the fast productivity growth in manufacturing (in a broad sense, including the software sector) and agriculture, the service sector would not currently have a high employment proportion. After all, the essence of the service sector is to deliver the primary products of the first and second industries to the final consumers, to transfer the production factors owned by members of the society to the first and second industries, and to provide entertainment and hospitality to people. Modern society has adapted well to the increasingly high productivity of manufacturing and agriculture by increasing the weight of the service sector in employment. This mode of economic growth may face two severe challenges in the near future. The first one is economical; that is, as manufacturing and agriculture become more and more productive and are being done by automation and robots, can the service sector still absorb the surplus labor force created? As increased productivity means fewer workers being employed and paid in the first and second industries, can the service sector grow further despite the ever-decreasing employment in those industries? Even when the remaining workers are paid more and the total wage incomes in the first and second industries are constant, the decreasing marginal propensity to consume implies that employee consumption in those industries would decrease. The second one is technological; service robots are becoming more and more capable of replacing human workers in the service sector. Where will the surplus labor force go when robots replace all the first-line (primary) service sector workers? A big test for human society is when the first-line (primary) workers in the service sector are nearly all replaced by robots. At that point, there will be no first-line workers in human society, and all people with a job will be managers and entrepreneurs. This epoch time-point

260

Chapter 5

can be called the zero-worker point, which arrives simultaneously as the zero-service worker point. Before that, society will meet a zeromanufacturing worker point when robots replace all first-line workers.

2. What is to Be Produced: A Critique of the Mainstream Economics View The non-satiation axiom alleges that people can consume an infinite amount of goods. Thus, the demand for goods will increase for a given population without limits. We have shown in Chapter 4 that this is incorrect. The production objective for a society is to generate products that satisfy the current and future needs of all members of society. The production to meet future needs is to create the capacity to produce goods and services in the future, the capital stock. How to allocate resources and production between current and future needs is an issue of intertemporal optimization. The utmost purpose of production for society is to meet the consumption needs of the members. However, the objectives of the owners of firms are primarily to meet their own physiological, psychological, sociological, and spiritual needs.

2.1. Finite Physical Goods for Non-Vanity Needs People consume physical goods because their consumption provides favorable stimuli to human sensory and cognitive modalities or facilitates such favorable stimuli. Foods, drinks, and clothes should be produced to maintain physiological function. Goods for fun, friendship, and curiosity are consumed for psychological and sociological needs. Tools, houses, home appliances, and public equipment must also be produced in their existing or improved forms to facilitate work and life. As we discussed in Chapter 4, goods to satisfy nearly every human sensory or cognitive modality, as well as tools to boost their satisfaction, have been invented, which is the reason why the proportion of services in the GDP has become larger and larger and why the growth of labor productivity has slowed down since the 1970s (David 1990). In the Robotic Age, new physical goods will be more likely to be improvements, combinations, or replacements of existing goods; the most frequent progress will be automated or intelligent versions of existing goods (i.e., characteristically novel goods). Since the range and quantities of goods society demands will remain the same, economic growth due to quality improvement will only create a few new jobs. In contrast, the First Industrial Revolution started by producing more existing goods cheaply with new

Production in the Robotic Age

261

technologies. The Second Industrial Revolution brought in many physical goods that did not exist before (i.e., functionally novel or mechanistically novel goods). Because the increase in demand outpaced the rise in labor productivity in non-farming sectors, both industrial revolutions greatly expanded the range and quantities of goods demanded and increased overall employment. The digital revolution produced new physical goods that mainly facilitate the acquisition, access, storage, processing, and exchange of information. They also promote the automation of existing physical products and their production. Although digital appliances per se are functionally or mechanistically novel goods that will create new jobs when they first appear, they will probably get people nowhere without goods invented during the first and second industrial revolutions in terms of creating jobs outside the digital sector. Online shops displaced brick-and-mortar ones and probably reduced overall employment rather than created more jobs. With the saturation of information-related physical goods and labor productivity growth due to automation, employment in the Robotic Age will unavoidably decrease.

2.2. Finite Services for Non-Vanity Needs The service sectors can be divided into two categories: to serve the producers and serve the consumers. Some firms conduct business in both types. Cargo transportation, wholesale, and retailing can be considered service sectors for producers, as are consulting, training, and financial services for firms. Consumer service sectors include public transport, entertainment, beauty and body services, restaurants, domestic help, and household financial services. The general trend of the division of labor because of economies of scale leads to externalization, professionalization, and specialization of many corporate and household responsibilities and activities. To externalize an activity, a firm or a household must have enough income from other activities to pay for the outsourcing. A key question regarding services in the Robotic Age is how far the externalization of corporate and household responsibilities can go. Many companies outsource their reception, information service, legal and regulation affairs, staff training, and logistics to business service firms. Companies of digital hardware tend to outsource manufacturing to contract manufacturers. Apple’s products, such as iPhone, are generally produced by contract manufacturers like Foxconn, the world’s largest technology manufacturer and service provider. There might be more functions that can

Chapter 5

262

be outsourced to specialized professional service providers. Preparing meals, making clothes, and doing daily chores used to be the responsibility of family members, especially homemakers. Nowadays, people usually buy clothes, have restaurant meals, or eat takeaways. Even when they cook, they often use semi-prepared food. Home appliances such as washing machines, dishwashers, and vacuum cleaners, especially automatic ones, have markedly reduced the workload of housework. Using the concept of home production (Gronau 1976), we may divide production into four types: social production for income, social production for spiritual needs (voluntary work), home production for saving costs, and home production for self-actualization (hobby or fun), as summarized in Table 5-2. Social production for income gradually moves into territories that previously belonged to home production for saving costs. In the Manual Age, feudal lords and wealthy landowners had servants to do their daily chores, which family members in ordinary households would do. In the Machine Age, especially toward the digital revolution, more and more activities previously done by family members are carried out by paid professionals, leading to the enormous growth of the service sectors or home appliances with much reduced human efforts. The services in the Robotic Age reach their upper limit when all home productions for saving costs (cooking, washing, cleaning, shopping, decorating, babysitting, planning, budgeting, making clothes, repairing, putting on makeup, etc.) are transformed into social production for services or performed by domestic robots. Since a person’s capacity to consume services is also finite, the variety of services will be exhausted sooner or later. Just like the categories of physical products, employment derived from service sectors will stop growing, and automation will reduce service sector employment. Table 5-2 Classification of individual production activity Subtype Social production

Paid job Voluntary work

No

Purpose Spiritual Saving needs money No Yes or No* No Yes

No

Yes

Earning income Yes

Selfactualization Yes or No* Yes or No*

Yes or Yes or No* No* Hobbies No No Yes Yes * Depending on personal circumstances, it can be yes for some people and no for others. Home production

Housework

Production in the Robotic Age

263

2.3. Finite Intangible and Information Goods for Non-Vanity Needs The digital revolution has dramatically reduced the costs of acquiring, storing, accessing, transferring, and processing information. Before the digital revolution, a good memory for social science studies and good mathematic skills for quantitative studies were essential. Some scholars were reputed for their mastery of information in their research fields. With search engines like Google, information can be obtained by clicking a button, which is much more than what the best human brain can memorize. Researchers who are weak in mathematics can readily perform quantitative analysis using software such as MATLAB, Wolfram Mathematica, EViews, Stata, and SPSS. In this sense, the digital revolution has reduced the impact of people’s intelligence differences on academic performance, just like firearms have mitigated the effects of people’s differences in physical strength on fighting. Table 5-3 Types of information goods Subtype Tangible Information Intangible

Software Information tools Hardware Virtual AI systems Intelligent systems

Intelligent Robots

Feature Physical objects carrying information Electronic files containing knowledge, stories, movies, etc. Computer programs help users perform specific tasks Instruments for transmitting or processing information Software that can complete an entire human job Machines that can complete an entire human job

Examples Books, magazines, photos, etc. Articles, videos, movies, etc., on the Internet or stored in memory devices Microsoft Office, electronic games, programming language, etc. Computers, telephones, routers, etc. AlphaGo, AlphaFold, Watson, ChatGPT Industrial robots have not reached the level of completing an entire human job

264

Chapter 5

Information goods can be classified into three types (Table 5-3): 1) information, including literature and knowledge in various media; 2) information tools for generating, processing, and transmitting information; and 3) intelligent systems that automate complete job systems. Type 3 products might eventually replace type 2 products, so what can be considered consumer products are type 1 information goods. Since people have a finite time endowment and physiological capacity constraints on how long they can continue to read, watch and listen, there are limits on the number of information goods people can consume. Moreover, as information goods after the digital revolution can be kept almost forever and provided at nearly zero costs, production of type 1 information goods to generate incomes will plateau and may decrease. By now, humanity has accumulated so much knowledge and literary works that nobody can read all or even a fraction of them in a lifetime. The only demand for a new production of type 1 information goods will be for information, fiction, and non-fiction about the contemporary world, nature, and society.

2.4. AI Systems and Robots AI and robots that automate complete non-physical job systems will be produced more and more. Their production will become the most important industrial sector. In the early stages of the Robotic Age, the production of these AI systems and robots will be a growing business, just like electric generators and motors, steam engines, internal combustion engines, and cars in their day. Unlike electricity and internal combustion engines, which led to more jobs in other sectors, AI systems and robots will eliminate jobs in other sectors. In the mature stage of the Robotic Age, the production of those AI systems and robots will plateau, and quality improvement will be the primary driver of growth. We have not got an overall labor productivity increase in the economy with current AI and robotics, which can be explained by the immaturity of these technologies. AI and robotics may need to go through two stages in their development: 1) effort-reducing technologies and 2) productivity-raising technologies. The effort-reducing technologies can improve workers’ conditions such that they can do the same job with less physical strength or mental power and attention. In contrast, productivity-raising technologies can increase output per unit of time. Most inventions in human history eventually became productivity-raising technology; for example, early farming instruments, such as digging sticks and hoes, and hunting tools, such as harpoons and arrows, significantly increased human productivity. If

Production in the Robotic Age

265

the product range is small and demand for them is soon satisfied, the productivity-raising technology will also reduce people’s effort. Productivity-raising technologies do not necessarily reduce workers’ efforts. The film Modern Times by Charlie Chaplin described how the fast assembly line made people work much harder. Most technologies are effort-reducing and productivity-raising, and some technologies could be effort-reducing in their infancy but productivity-raising when they mature. Information technology at institutional levels was more effort-reducing than productivity-raising during the 1980s and 1990s. The information technology’s lack of impact on productivity growth since the 1970s is a phenomenon of immature AI and robotics during the stage when they are effort-reducing technologies instead of productivity-raising technologies. From the 1990s, AI and robotics entered the transitional phase from an effort-reducing technology to a productivity-raising technology in the remaining blue-collar and traditional white-collar jobs. In the foreseeable future, the advances in AI and robotics will replace even more human workers in the traditional manufacturing sectors, thus increasing productivity. The target of such replacement is not the monotonically repeated operations that early robots have already replaced for production assembly lines. It is the supportive human workers who have not been replaced by robots yet and the professionals whose jobs have been considered beyond the capability of AI and robotics until a very distant future.

2.5. Goods and Services for Vanity Needs As discussed in Chapter 4, goods and services for vanity needs cannot be the primary driver for economic growth because vanity needs some form of exclusivity. A good becomes an item of vanity only when a small number of rich and famous people own it. In the Robotic Age, we might still have expensive luxury brands to satisfy the vanity of some people. There could be a class of human-made goods as luxuries for those willing to pay a high price, which may support some levels of human employment when most people prefer cheaper robot-made goods of better quality. There is another possibility, i.e., when every member of society lives a respectable life, society becomes much more equal than ever before, and using expensive goods will become a much less relevant factor in economic activities. People will be more concerned with their self-actualization and selftranscendence than how others will view them.

266

Chapter 5

3. Who Will Produce? In the Manual Age, it was obvious who the producer of a particular product was. In the Machine Age, because of the division of labor in producing a complex product, many workers and managers are involved in the production processes, so it is not obvious who should get more credit for making a product. Automation began to replace human workers at assembly lines in the early twentieth century. In the Robotic Age, the question is, who cannot be replaced by AI systems and intelligent robots?

3.1. Horizontal and Vertical Division of Labor The division of labor is the separation of tasks in any economic system so that participants may specialize. We may consider two types of division of labor: horizontal and vertical (or hierarchical). The horizontal division of labor includes the parallel and serial divisions of labor. The parallel division of labor means that people specialize in producing different (parts of) goods. In his Republic, Plato noted that a state would “need a farmer, a builder, and a weaver, and… a shoemaker and one or two others to provide for our bodily needs.” These workers are in a relationship of the parallel division of labor. The serial division of labor means that people specialize in working on one stage of the production process of a good. Workers on an assembly line are an example of a serial division of labor. William Petty showed the division of labor in Dutch shipyards, where several teams did the same tasks for successive ships (Petty 1751). The vertical or hierarchical division of labor is a more sophisticated organizational form of production. As more people produce a type of good, it becomes necessary to have someone to coordinate the activities of average workers and to have someone to maintain and service the machines instead of operating them. The coordinators are called forepersons, supervisors, or junior and middle managers. The supervisors do not actively participate in the production process, which can continue in the absence of the supervisor. The supervisors assign workload to the workers, solve problems that the workers cannot solve, and have the authority to call for help from inside or outside the firm. Similarly, technicians who maintain the machines do not normally operate the machines; the production process can carry on without machine maintenance technicians. For larger organizations, the senior managers are above the supervisors and the machine maintenance technicians. They are even more detached from the production process, which can generally continue for months without

Production in the Robotic Age

267

any input from the senior managers. The further away from the first-line production process, the higher the income somebody earns. Although senior managers are hardly needed in production processes, they are the most highly paid in the firms. Some economists may claim that the supervisors and senior managers earn the marginal products they generate in production, leading to higher incomes. Still, it is more likely that the higher salaries are incentives to discourage managers from doing things detrimental to the organizations or being negligent in performing their duties. Another reason for high salaries to senior managers is the price paid for the organization’s preference to appoint somebody with experience in a similar position. The extra money above the senior managers' marginal products can be considered premium paid for the “experienced” brand. The higher the position, the smaller the pool of candidates with prior experience; hence the higher the premium charged by the successful candidates. If organizations were more willing to appoint people without experience in a similar position but with good potential, the senior managers would be paid much less. Modern corporations tend to have horizontal and vertical labor divisions, let alone conglomerates with products in different industrial sectors. In addition to the production component, modern corporations have their marketing and sales, finance and accounting, human resources management, planning and strategy, and probably legal and regulatory elements. All these functional components can be analyzed in the same way to divide human members into first-line workers, supervisors, and senior managers in the vertical division of labor. These components are parts of the horizontal division of labor. In the manufacturing sector, electromechanical automation and robots have markedly reduced the number of first-line assembly line workers and significantly increased labor productivity in these industries.

3.2. Replacing First-line Human Workers with Robots The Machine Age is characterized by machines that replace human manual power and perform human operations with increased accuracy, efficiency, consistency, and endurance. Although the automation of certain operations has existed since ancient times, especially since the Industrial Revolution, most production processes still need constant human interventions and surveillance. The Robotic Age will be characterized by replacing human intervention, which represents input from human brain power, in most production processes. The Robotic Age will have different stages of development. The first stage will be the replacement of the current first-line workers by robots that perform the repetitive operations that need basic human intelligence of human workers. Our current robotic technologies are

268

Chapter 5

closing in at this stage. The second stage will automate most supervisory roles in the production and service processes. To reach this stage, AI and robotics need to have the capacity to produce robots capable of performing everyday life routines, excluding emotional and strategic behaviors. The challenge for reaching this stage is replicating human cognitive and motor systems rather than general human intelligence. The third stage might begin when AI matches general human intelligence without emotion. The application of robots in the industry is one phase of industrial automation. Therefore, industrial robotics is a sub-branch in industrial automation that aids various manufacturing processes. For now, such manufacturing processes include machining, welding, painting, assembling, material handling, and so on (Shell and Hall, 2000). Industrial automation replaces humans’ decision-making in the production process and manual command-response activities with mechanized equipment and logical programming commands. Most decision-making in the first-line workers’ operations was relatively simple. These include, for example, how a specific workpiece should be held in some position, how much material should be removed from a particular workpiece, when the machine should be switched on, when the next workpiece should be processed, and when the device should be switched off. With the rapid development of automation and robotic technologies, nearly all first-line workers in manufacturing will be replaced by robots. The current automated technology has made some factories run lights out. FANUC (Fuji Automatic NUmerical Control), the Japanese robotics company, has been operating a “lights-out” factory for making robots since 2001 (Ma 2013; Bogue 2014). Lights-out or lights-out manufacturing is a methodology (or philosophy) whereby factories require no human presence on-site. These factories run with the lights-out and are fully automated. Philips in the Netherlands uses lights-out manufacturing to produce electric razors, with 128 robots from Adept Technology (Bogue 2014). As we mentioned earlier, among the main factors that limit the progress in replacing first-line workers with robots, the technical one is whether robots can do the job equally well or better. The economic one is whether robots cost less than human workers to produce the same goods at the same quality. The current constraint on expanding lights-out factories is that technology is often not economical enough. Although firms may use robots to produce lower quality goods at lower production costs than those produced by human workers, this cannot be the trend of the future economy. Robots must do the same job equally or better to replace human workers.

Production in the Robotic Age

269

For many automated production processes, workers must set up tombstones (a pedestal-type fixture, tooling tower, tooling column, or fixture block) that hold parts to be manufactured. Moreover, even when human workers at assembly lines have all been replaced by robots, human workers may still be required for the following reasons: 1) the robots working at assembly lines need maintenance and may break down such that some human interventions are needed to fix the problems; 2) the robots working at assembly lines are usually mobile only on some defined platforms, the raw materials or intermediaries need to be fed on the assembly lines for the firstline robots to work on; 3) external supplies may still need to be handled by human workers and the finished products may still be transported to buyers by human workers or with substantial involvement of human workers. The following automation step is to develop robots that can unload supplies from delivery vehicles, load materials onto the tombstones or starting platforms for the assembly line robots to work on, unload finished products, and prepare to deliver products to buyers. These robots can be called logistic and support robots. A more demanding task for robots is to service machines (and probably robots) and fix breakdowns. If a logistic and support robot can also service machines and repair breakdowns, it is a multifunctional mobile robot. Such multifunctional robots can perform jobs in service sectors as well. Those conducting the most basic loading and unloading tasks can be called loading robots, which still need more advanced technologies than the assembly robots because they have to be mobile in a much larger area than assembly line robots and deal with various materials carried by different vehicles with different requirements. Although the loading robots may need to be more “intelligent” than the assembly line robots, the difference could be small. The supply-receiving decks could be standardized so that loading robots can handle supplies as readily as assembly line robots perform their manufacturing operations. Then the difference between the two types of robots is just the range of mobility. Loading products onto delivery vehicles can also be achieved by some standardized decks. The loading robots can be viewed as one type of service robot. The service industry is much less automated than the manufacturing industry, partly because the service operations are much less standardizable and customers are more diverse human beings than parts of goods to be worked on by robots. In recent years, service robots have become more functionally versatile, and some restaurants are using robots as waiting staff (Garcia-Haro et al., 2020; Hwang, Park, and Kim, 2020). The complete automation of service sectors needs mobile multifunctional robots.

270

Chapter 5

When robots replace the first-line human workers, the human supervisors need not be on-site except when they have to perform tasks handling physical objects there. The robots could be remotely controlled by human operators when necessary, especially when the intelligence of the robots has yet to be able to accomplish their duty without human supervision. The human supervisor can work far from the factory floor and monitor robots in factories in different cities. They may work from home or at several regional factories' regional human supervision and support centers (Fig.5-2). Several companies may share one human supervision and support center in an industrial cluster. Human supervision can be minimal when AI allows the loading robots to work independently.

Fig.5-2 Autonomous factories and the support center (human). One support center staffed by human workers can monitor and supervise several autonomous factories.

The automated guided vehicles or automatic guided vehicles (AGVs) currently available can fulfill the role of the mobility aspect of loading and multifunctional robots. An AGV is a mobile robot that follows markers or wires on the floor or uses vision, magnets, or lasers for navigation. They have been used in industrial applications to move materials around a manufacturing facility or warehouse, including transporting received materials to the warehouse and delivering them directly to production lines or from one process to another (Vis 2006; Azadeh, De Koster, and Roy 2019). They have been used to transport raw materials such as paper, steel, rubber, metal, and plastic. AGVs can store and stack rolls on the floor, in racking, and automatically load printing presses with paper rolls. They can move pallets from the palletizer to the warehouse/storage or the outbound shipping docks. They can also pick up pallets from conveyors, racking, or staging lanes and deliver them into the trailer in the specified loading pattern. Some automatic trailer-loading AGVs utilize natural targeting to view the

Production in the Robotic Age

271

trailer walls for navigation. They can be either wholly driverless or hybrid vehicles. Europe Container Terminals (ECT) pioneered using AGVs to move sea containers in the Netherlands at the Delta terminal in the Port of Rotterdam (Rodrigue and Notteboom 2021). Once using loading and other service robots becomes cheaper than hiring human workers for the same role, the manufacturing and service industries would return to their pre-Industrial Revolution form, where one ownerworker and their family members made up the main workforce of the family-owned workshop. The Industrial Revolution brought factories as the main form of industrial organization, with many workers under the same roof. Even professionals relying on and handling documents often work under the same roof. The wide application of robotic technologies and their fast progress have started to reverse the trend of industrial expansion with more workers working under the same roof. This reversion is achieved at a much higher level of technological development. The COVID-19 pandemic accelerated the adoption of remote working technologies. People who need not handle physical objects generally worked from home during 2020–2021 following the onset of the pandemic.

3.3. Supervisors and Robots An essential responsibility of supervisors is to monitor human workers’ efforts and assign jobs to them. With all first-line workers being replaced by robots, job assignments on the factory floor could be better accomplished by AI optimal planning systems, and human supervisors need not monitor robots’ efforts. The role of supervisors will change from assigning jobs and monitoring human workers’ efforts to 1) maintaining robots and assembly lines and fixing problems when capable; and 2) communicating with senior managers, suppliers, buyers, and maintenance services. In the early stage of the Robotic Age, human supervisors worked with loading robots to load and unload raw materials and finished products. Usually, the supervisors only need to monitor the irregularities in the robots and assembly lines. The next step of automation in production is to replace supervisors/maintenance workers with supervision robots, and firms will operate with only one or a few senior managers working in offices, as illustrated by Fig.5-3. If supervisors are replaced by robots that only need to monitor the assembly line and immobile robots and call for repair services when anything goes wrong, the technology for making such supervision robots will not be much more sophisticated than that for self-driving cars. Self-driving cars can respond to the environment without human intervention and transport

272

Chapter 5

passengers (or cargo) from one place to another. If the roles of supervisors and maintenance/service workers are combined, the ability to handle physical objects of different sizes, shapes, and textures is more challenging. For supervision robots, maintenance robots, or service robots (rather than self-service machines), more complicated physical maneuvering is needed to fulfill their responsibilities because they need to be mobile and able to perceive objects and events in their surroundings and respond with humanlevel physical dexterity. The supervision/maintenance robots will be the mobile multifunctional robots we mentioned earlier. The adoption of multifunctional supervision/maintenance robots should be not only technically feasible but also economically cost-effective.

Fig.5-3 Corporate management line charts at different stages of the Robotic Age. In the first stage, supervisors and senior managers are still human. In the second stage, supervisors are replaced by multifunctional robots. In the third stage, senior managers are also replaced by robots.

In service industries, serving human beings could be more challenging or less challenging than servicing machines and robots. The more complicated part is that when humans expect to be served, robots must perceive human behaviors and surrounding objects and respond to human behaviors and emotions. The less challenging part is self-service by human beings. Selfservice machines are another form of service robot, although they are usually immobile and primarily rely on customers to perform their job. For now, self-service machines need human workers for customer support and refilling stocks. When self-service machines (as well as shelves in brickand-mortar shops) can have their stocks filled by supervision/maintenance robots, and their breakdowns can be fixed by maintenance robots, automation of service industries will be genuinely achieved.

Production in the Robotic Age

273

3.4. Senior Managers, the Homeostasis of the Internal Environment, and Robots Senior managers are not directly involved in the production process, and their responsibilities need no sophisticated specialist knowledge or technical skills. The primary responsibilities of senior managers are setting strategy, making decisions that require the highest authority in the organization, supervising middle managers, appointing middle managers (supervisors), and representing the organization. As a firm’s strategy will not be set frequently, and middle managers are often relatively stable, the senior managers’ primary jobs are making choices because of their position of authority and representing the organization in society. Representing and projecting the organization’s image is probably the most important job for senior managers such as chief executive officers (CEOs). A high-profile CEO is a good publicity tool for a firm. One observation of leadership in organizations or states is that poor leadership often had little adverse effects on the overall performance of an organization unless the leaders had done some extremely bad or stupid damage to the organization in peacetime. Here, we may borrow the physiological concept of homeostasis, particularly the homeostasis of the internal environment formulated by Claude Bernard in 1878 (Cannon 1929). Because of the homeostasis of the internal environment, adverse impacts on an organism tend to have much less influence on the organism’s function. This is also true in organizations consisting of people, which also have this homeostasis of the internal environment. Most members of an organization would not want it to fail. They all have some power to correct the impact of bad decisions from their leader, which, together with their products and technology, form the homeostasis of the internal environment of an organization. Thus, members will mitigate the adverse impact of their leader’s wrong decision within their power and try to give feedback to the leader for amending the decision when they implement the leader’s decision. Therefore, if the senior managers do not make irrational decisions and the external environment does not change dramatically, the behaviors and competency of the senior managers do not significantly impact the overall performance of an organization because of this homeostasis of the internal environment. Except when the senior managers make irrational or extremely foolish decisions, or the external environment changes significantly, the performance of an organization is determined more by the internal and external environments than the senior manager’s ability.

274

Chapter 5

Senior managers make decisions based on choices available to the organizations. The decision-making situations generally belong to the following types: 1) There is a best option among several available paths, and the role of the senior manager is to complete the approval procedure. 2) Two or more competing options cannot be differentiated by recent experiences, expertise, or available information, so decision-making is simply a bet on luck or nature. 3) In rare situations, the correct path appears to be wrong while the wrong tracks seem correct, but the senior manager takes the right direction against objections from all sides. In the first scenario, the senior manager has no real intellectual contribution to decision-making. In the second scenario, the senior manager will take credit for making the right choice or get the blame for making the wrong choice. Still, they would usually be forgiven for bad decisions because of uncertain circumstances. The decision-making is similar to throwing a die. For the third scenario, although we can find many such examples in strategy textbooks and case studies, it is difficult to tell whether the manager is simply lucky, as in scenario 2, in making a correct bet, or they indeed had the superior insight to recognize the right path to the success. In most cases, a military commander who made a bold decision and succeeded in defeating a stronger enemy was more likely to be lucky enough to have an incompetent opponent who was too arrogant, overconfident, or simply stupid. From the above analysis, it is easy to see that the decision-making job of senior managers is a job of reasoning and logical inference, not a technical or dexterous one. From certain conditions, a conclusion can be reached with near certainty. When some information is missing for a definite conclusion, we may arrive at a probabilistic conclusion. Sometimes, we must throw a die to choose between two or more choices with the same or similar probabilistic outcome. Such reasoning and inferring capacity are more readily to be mimicked and surpassed by AI than manual maneuvering. The role of senior managers can be replaced more readily by AI expert systems in the twenty-first century. If we look carefully at the decisions and conditions senior managers face, we will find that the further the managers are away from the first-line production process, the less sophisticated the decision set becomes and the less complicated the action to complete the decision-making. An artificial manager system could be developed with

Production in the Robotic Age

275

current technology. Automatic machine trading carried out by computer systems has been operating in the stock market for years. When robots or AI management systems replace senior managers, firms become lights-out or humanless. Compared with senior managers whose role can be readily mimicked by machines, maintenance workers are more difficult to replace by machines because their responsibilities involve not only specialist knowledge (which is easy to replicate by machines) but also versatile physical maneuvering (which is difficult to replicate by machines at the moment). Robots might not readily replace the senior managers’ role as publicity stars for their firms. Still, in a world where big data provide information on quality goods and services, publicity stars’ impact on consumer goods will be marginal. We have shown that there are limits on the maximum non-vanity consumption per capita; replacing human workers with robots will significantly reduce the number of employed workers. Small and medium-sized enterprises may no longer need human workers and managers; just the owner and their family members will be sufficient to operate and manage a much larger business, similar to the pre-Industrial Revolution workshops.

3.5. Measuring Labor Productivity and Management Productivity With the replacement of human workers by robots in production processes, labor productivity will have an entirely different meaning from our usual understanding. The output will be attributed to managers who have not made any products. Since the aggregate demand will grow very slowly when the maximum consumption level is reached, the labor productivity of a country will also grow very slowly if measured against the working-age population. However, when measured against the number of people with a job, labor productivity will increase quickly when human-replacing AI and robot technology matures. During the process of replacing most human workers with robots, the labor productivity of a firm is still a useful measure. The overall labor productivity used to be measured by a firm’s average labor productivity ‫= ݕݐ݅ݒ݅ݐܿݑ݀݋ݎ݌ ݎ݋ܾ݈ܽ ݁݃ܽݎ݁ݒܣ‬

ܶ‫݀݁݀݀ܽ ݁ݑ݈ܽݒ ݈ܽݐ݋‬ ݊‫ݏ݁݁ݕ݋݈݌݉݁ ݂݋ ݎܾ݁݉ݑ‬

In the above equation, the employees include the firm’s managers and owner-managers if the owners also work full-time for their firm. In the

Chapter 5

276

Robotic Age, we need to measure the productivity of the remaining supervisors and senior managers, so the above formula will still work. In addition to the above-average labor productivity, we may also measure management productivity. Using our three-level vertical division of labor framework, we may measure management productivity by measuring the ratio between the total value added and the number of managers, including supervisors. ‫= ݕݐ݅ݒ݅ݐܿݑ݀݋ݎ݌ ݐ݊݁݉݁݃ܽ݊ܽܯ‬

ܶ‫݀݁݀݀ܽ ݁ݑ݈ܽݒ ݈ܽݐ݋‬ ݊‫ݏݎ݁݃ܽ݊ܽ݉ ݂݋ ݎܾ݁݉ݑ‬

Management productivity does not measure the value created by an average manager. If we assess the performance of a section using a similar measure, managers would be more motivated to increase the number of first-line workers under their supervision, no matter whether they can manage so many workers. “Empire building” has been an endemic problem for most organizations because managers are usually appraised by the output of their teams without considering the size of their teams. Even when the size of their teams is considered, having a large squad is often perversely perceived as a significant contribution rather than taking up resources that could be used more efficiently or more productively. Another reason for empirebuilding is that many managers enjoy being above their colleagues and having authority over their subordinates. Therefore, some managers may strongly oppose lights-out production. The employees-to-manager ratio (EMR) reflects how the first-line workers have been replaced by automation and AI systems: ‫= ܴܯܧ‬

݉ܽ݊ܽ݃݁݉݁݊‫ݕݐ݅ݒ݅ݐܿݑ݀݋ݎ݌ ݐ‬ ݊‫ݏ݁݁ݕ݋݈݌݉݁ ݂݋ ݎܾ݁݉ݑ‬ = ܽ‫ݕݐ݅ݒ݅ݐܿݑ݀݋ݎ݌ ݎ݋ܾ݈ܽ ݁݃ܽݎ݁ݒ‬ ݊‫ݏݎ݁݃ܽ݊ܽ݉ ݂݋ ݎܾ݁݉ݑ‬

A smaller value indicates that one manager manages a smaller number of employees. When there is no first-line worker in the production process, the EMR equals one. When robots and AI management systems entirely operate a firm, the management productivity and the average labor productivity will become infinitely large and lose their economic relevance.

3.6. The Evolving Relationship Between Humans and Robots An influencing view on robots and AI systems is that the future may lie in the cooperation between AI/robots and humans (Brynjolfsson and McAfee

Production in the Robotic Age

277

2014). People with such a view believe an association of AI/robots and humans will be more efficient than either working alone. Their collaboration will open new areas of economic activity and provide new demand for human workers. Our examination of production technology development shows that the general trend is decreased human involvement in the production process. In the Manual Age, people use tools to increase their productivity, especially the efficiency of their muscle power. During the Machine Age, people replaced their muscle power with machine power and introduced automation to reduce the involvement of their mental power in the production process. People can leave most production processes to robots and AI systems during the Robotic Age. Cooperation between robots/AI and humans works better than working alone could be a phenomenon due to the imperfection of robots/AI. As we have often observed in everyday life, the lower the quality of an appliance, the more human intervention and repairs are needed. The better performance of robot/AI-human cooperation could be simply a manifestation of the existing weakness in the robot/AI system, which can be corrected and compensated by humans. With technological progress removing more and more of such inadequacies, robot-robot cooperation could become more powerful than robot-human cooperation. There will be a long transition from robots performing simple, repetitive tasks for humans to robots performing complex tasks better than robot-human cooperation. Production processes will gradually evolve from human-dominated to robot-dominated in this extended transition. In some processes, human involvement might always be an advantage, whereas in most cases, robots and AI systems can excel without human involvement.

4. How Will Goods Be Produced? Developing intelligent robots and information technologies can fundamentally change the production process in many sectors. The successful changes in the production or business process are process innovations. A pure process innovation changes how a product, service, or business is done. In practice, many process innovations will lead to changes or improvements in the product or service. In theory, process innovations can be roughly classified into five categories according to their impact on production or business costs: 1) innovations that reduce fixed costs; 2) innovations that reduce marginal costs; 3) innovations that reduce marginal costs but increase the initial fixed costs; and 4) innovations that reduce both marginal costs and fixed costs; and 5) innovations that increase production flexibility and

278

Chapter 5

convertibility and reduce the marginal cost of producing an additional model. The computer per se is a great success story of innovation cutting costs, as supercomputers a few decades ago had much less computing power than smartphones nowadays. Computers and the Internet have led to many fixedcost-reducing innovations, especially for service sectors. The reduction in fixed costs lowers the barrier to new entrants. For example, starting a business to design and print leaflets could involve a substantial investment before personal computers (PCs) and printers were widely available at meager costs. Nowadays, many people can get such jobs done at home. These innovations reduce economies of scale and the minimum efficient scale (MES), the lowest point where the firm’s long-run average costs are minimized. Modern technologies have led to the invention of many cheaper machines capable of performing the same jobs as older equipment. Reduced fixed costs could increase competition in the marketplace and minimize monopolies.

4.1. Agriculture, Forestry, Husbandry, and Fishery Modern farming has been highly mechanized, with a dramatically reduced proportion of farmers and farm workers in the population. Grain production will likely be the first in agriculture to automate fully. Sowing and planting, watering, applying fertilizers, controlling weeds and pests, harvesting crops like wheat, barley, corn, and soybean, etc., will be carried out by robots under minimum human supervision or support and eventually be fully autonomous. GPS- and vision-based self-guided tractors and harvesters have already been available commercially. The progress in further automation depends more on whether new technology can make autonomous farming cost-effective than the current mechanized farming. Robots are being tested at rice paddies with an eye on the complete automation of rice production in Japan (Oshika 2023). Vegetable and fruit production, especially harvesting fruits, needs more human labor. Fruit-picking robots now cannot compete with human pickers, mainly because of three challenges: 1) fruits are sporadically distributed in three-dimensional space, which is more difficult for robots to locate them and grip them, in comparison with grain crops, which can be viewed as densely distributed on a two-dimensional plane; 2) unlike grain crops, fruit ripens at different times so that robots should be able to locate the ripe ones and leave the unripe ones undamaged, which requires more advanced visual recognition; 3) fruits are often more delicate and susceptible to bruise,

Production in the Robotic Age

279

which requires soft handling. These challenges are not insurmountable. Challenges 1) and 2) can be addressed by refining and improving some existing technologies in machine vision, recognition, and robotic arms. Addressing challenge 3) needs more advances in using soft materials to construct the vital functional parts of robots and developing the capacity to sense soft materials such as cloth, liquid, gel, paste, dough, meat, soft fruits, etc. Such technologies should be available in the not-too-distant future. Horticultural robots for tasks such as thinning, pruning, weeding, spraying, organizing potted plants, and monitoring have been developed. Complete automation for those tasks is less challenging than automation of fruit production and more challenging than farming grain crops. It can be readily achieved when full automation of fruit production is completed. Since people will have more leisure time with the wide application of robots and AI, they may want to take care of plants in their gardens and public places for personal enjoyment. This might make many horticultural tasks less worthwhile to automate and use robots. Carrying out a horticultural job feels more like exercising in a gym for gardening lovers. If too many people enjoy doing horticultural tasks, they may have to register for such opportunities. A possible future development for agriculture is the transformation of farmland into innovative farmland systems or intelligent farmland systems (IFS) connected with the Internet of Things (IoT). In an IFS, various sensors and field management devices are installed to monitor crops, vegetables, and fruit growth and conditions. They also water, apply fertilizers, control weeds and pests, analyze information on plant conditions and their local environment, and switch on the relevant devices to respond to a particular condition. For example, a sensor measures soil drought level; after that, the intelligent device in which it is installed turns on the watering automatically, without any active intervention by the human user. They are connected to the IoT. If desired, the user can control the IoT devices remotely via an app on a smartphone or other devices. Robotic seeders, planters, or harvesters can interact with these sensors and field management devices to carry out sowing, planting, or harvesting. Whether IFS will be implemented depends on whether they are cost-effective. Farming would still require some level of human involvement if they are not cost-effective. Suppose there is a technological breakthrough in making nutrients chemically in reaction chambers or biologically in fermentation vats with energy efficiency. In that case, food might be made in factories rather than farmlands. In a popular science book for children, it was imagined that carbohydrates would be made from water and carbon dioxide (CO2) in

280

Chapter 5

factories, and so would other nutrients; farming would no longer be needed (Li and Hua 1959). Chemically synthesized food from CO2 and water seems unlikely to be cost-effective with current scientific understanding. It is possible to grow microbes in fermentation vats or protein, fat, starch, and fiber-rich plants in fields and then transform them into different foods. However, they are unlikely to be both cost-effective and satisfy people’s tastes. Forestry could use robots and airborne drones for forest management, lumbering, and fire prevention (Torresan et al. 2017). Robots for forestry are likely to use similar technology as those for growing vegetables and fruit, but forestry tasks may be less challenging. An intelligent forestry system may also be implemented, with various equipment and sensors installed and connected to the IoT. A smart forestry system could complete all or most tasks without human involvement. Still, human users can control robots’ activities via mobile phone applications through the IoT. Robots could also be used in husbandry to carry out various tasks. Applications such as automatic milking and washing have already been used. Husbandry can be classified into two categories: 1) free-range husbandry, where livestock, such as cattle, can move around in a relatively large area, and 2) intensive husbandry, where animals are housed in a relatively narrow area. In free-range husbandry, robots can tend herds, feed them, and monitor their growth and health conditions. Replacing veterinary services with robots and AI systems would be challenging because animals usually would not cooperate with a human or robotic veterinarian and could not tell a veterinarian their main problems. In the future, robots may acquire the same observation ability as human veterinarians. Before that, there will still be human veterinarians. Robotic veterinarians may isolate and quarantine diseased animals, let them recover without treatment, and cull them if their conditions cause too much suffering. Intensive husbandry like poultry and pig farms have already been mechanized with automation in many production processes, but introducing robotics and AI systems can further reduce human participation in husbandry. Complete autonomous intensive poultry or pig farms will need to redesign their farm infrastructure to enable and empower the robots. Technologies for husbandry could also be used in fishery, which provides an essential part of the food supply. This includes fishing, i.e., catching aquatic animals (and plants) existing naturally in waters, and aquaculture, which farms aquatic animals and plants as food and industrial materials. Current technology can build an autonomous open-water fishing system,

Production in the Robotic Age

281

but it still needs to be cost-effective compared with fishers. Progress in the autonomous ship and fishing facilities may make it cost-effective. Aquaculture could adopt intelligent aquafarm systems incorporating sensors and farm management facilities connected to the IoT to monitor water quality and the conditions of farmed animals and plants and respond as conditions change. Humans can intervene through the IoT.

4.2. Mining Mining provides the raw materials for economic activities, and some mining sectors have been hazardous for miners. Mine equipment automation will continue in the direction of full automation. Sensors and positioning guidance devices could be installed in mining pits and connected to the IoT. Robotic mining vehicles such as excavators, bulldozers, and transporters could be coordinated or controlled by a central intelligent autonomous mining control system. Multifunctional robotic miners will be responsible for maintenance, removing obstacles, and fixing breakdowns. Besides fixed sensors and positioning guidance devices, sensors in robotic mining vehicles and multifunctional robotic miners provide information to the central intelligent autonomous mining control system. In the early stages of autonomous mining, human intervention may still be needed for fixing breakdowns when the problem-solving ability of robots is inferior to human experts. In addition to the routine operations discussed above, mines must set up their infrastructures before regular mining operations can start. Building a mine’s infrastructure in the Robotic Age is a task for the building industry, but it has different requirements from those of residential and office buildings or factories on the ground. The current arrangements of mines are still designed for human miners to work there, and mining automation is developed to fit with the existing mining infrastructure. When mines are designed to suit robotic miners, the arrangements could differ significantly from those for human miners. Consequently, mining automation could also be very different from that in mines primarily designed for human miners.

4.3. Manufacturing As we argued in Chapter 4, since the First Industrial Revolution, manufacturing gradually became the largest sector in the economy until the 1970s. Almost all categories of material goods that satisfy human desires have been invented; from then on, manufacturing mainly improves,

282

Chapter 5

segments, combines, and streamlines existing product categories rather than adds new product categories. The critical issues in such manufacturing are efficiently producing goods that satisfy different types of customers and clients. 4.3.1. Mass Production versus Individually Tailored Production In the 1960s and early 1970s, many homemakers in China still bought fabrics and made dresses for family members, such that a sewing machine was one of the key home appliances. For those households which could not make their dresses, they would go to tailor shops, which were widespread in Chinese cities and towns. Since the late 1970s, Chinese people have gradually moved to buy ready-made clothes from shops, and only a few households still have sewing machines. When the production scale is large, the manufacturers can produce an extensive range of differentiated products in size, style, materials, and color. Customers can choose one from those available. In the Robotic Age, AI and more advanced flexible manufacturing technologies may provide more tailor-made products for individual needs, stopping or reversing the trend of industrial concentration in manufacturing. 3D printing technology, which makes cost-effective tailor-made production of many consumer goods more probable, is such a technology. It is also known as additive manufacturing (AM), referring to processes used to synthesize a three-dimensional object by depositing successive layers of material under computer control to create an object (Berman 2012; Shahrubudin, Lee, and Ramlan 2019). Niche market producers might use 3D printers as their main instruments for tailor-made production. Consumers whose preferences could not be met by the diverse styles available from mass production can be served by the tailor-made manufacturing approaches, which can be better and cheaper because of the Internet, flexible manufacturing, and improved transport and delivery systems. Having a 3D printer in every household is probably not economical to print goods for everyday use. Likely, many consumer goods cannot be made by 3D printing at a comparable quality to goods made by traditional production technologies. The marginal cost-reducing innovations, usually associated with higher fixed costs, increase economies of scale and the MES. Since the Industrial Revolution led to more industrial concentration and market power of a few successful firms, if we view entertainment as one industry, photography, movies, radio, televised broadcasting, and audio-video recordings are all marginal cost-reducing technologies for live performances by artists. Artists

Production in the Robotic Age

283

would not have the incomes they currently enjoy without those technologies. Many potential artists, who would have made a living by serving the local people in an age without those technologies, must do non-artistic or nonentertainment jobs. The extreme form of marginal cost-reducing innovations is to reduce the marginal cost to zero. According to economics theory, goods should be priced at the marginal cost, so zero marginal cost means zero price. Intellectual products like software are zero marginal cost; law enforcement rather than market forces protects their non-zero prices. The near-zero marginal costs led to the phenomenon of “the winner takes it all” and the emergence of Microsoft (PC operating systems), Google (advertisements based on an Internet search engine), Facebook (advertisements based on a social networking service), and Alibaba (ecommerce platforms). Production by 3D printing still requires materials such as consumables and spare parts. If there is more than one 3D printer at a community production center, other printers could print a component to fix the out-of-order printer. However, the raw materials for printing have to be made somewhere else and transported to the venue of 3D printers, as illustrated by Fig.5-4. The production, storage, and transportation of 3D printing “ink” are more likely to be made at a massive scale. The cost to produce 3D printers by mass production is likely much lower than that of making them individually by 3D printing. As the consumables of the 3D printers need to be delivered occasionally, the finished goods that people in the community need could also be produced and delivered from mass-production assembly lines. Therefore, the future scenario will likely be 3D printing supplementing mass production rather than replacing it. Innovations in transportation and delivery systems will further reduce transportation costs and increase delivery efficiency, supporting mass production and concentration. With a sophisticated and efficient transport system, delivery of ready-made products to users is likely to be more costeffective than self-production by users. A simple analogy can be drawn from printing a book. Individuals can print a book at home with a computer and a printer. However, a book printed in this way tends to cost more, is less well-bound, and takes more time than books produced in factories. Before the advent of E-readers such as Kindle, it was better and cheaper to buy a factory-printed book than print the book by yourself using a PC and a printer. E-book downloads are a revolutionary transportation innovation for books, such that the contents of a book can be inexpensively shifted to the E-readers.

284

Chapter 5

Fig.5-4 Supply of 3D printer consumables. They are mass-produced and delivered to regional distribution centers for further delivery to community 3-printers.

4..2. Specialized Robots versus Multifunctional Robots The automotive industry currently uses the largest proportion of industrial robots. Production processes in the chemical, pharmaceutical, and life science industries are usually highly automated, and robotics can perform a wide range of tasks, such as dispensing, material handling, and packaging. Although robots are widely used in the automotive and chemical, pharmaceutical, and life science industries, they tend to be specialized robots to perform repetitive tasks. There are still many jobs that need human workers. For example, robots are often used in high-throughput screening for drug candidates in the pharmaceutical industry. Still, they only dispense liquid and transfer cell culture plates from one deck to another. Human workers have to prepare the solutions which need to be put into various containers for robots to dispense, place cell culture plates on decks for robots to transfer, solve problems when there is a breakdown, and remove the waste (used materials) from the automated screening system. Automating those supply and support tasks performed by humans requires the development of long-range mobile multifunctional robots, which handle supplies to assembly lines or other operation systems, solve problems that cause system breakdowns, and remove waste produced during operations. Using multifunctional robots to replace human workers depends on whether technological progress makes them a reality and whether using them is cost-

Production in the Robotic Age

285

effective compared to employing human workers. The costs depend on whether we can redesign factories to make them intelligent factory systems installed with sensors and guidance systems to support multifunctional robots performing their tasks. In the early stages, before humans can be fully replaced, humans may work with robots remotely in guiding multifunctional robots to solve breakdowns. When robots were first used for highthroughput screening, security staff on night shifts would be alerted when the automated screening system had a functional failure. Often security staff solved breakdowns at night with guidance from researchers/technicians over the telephone (video communication was not mature in the 1990s). Early multifunctional robots could play the role of security staff, performing their tasks with some help remotely from human workers. 4.3.3. Concentrated Production versus Distributed Production Intelligent factory systems might pertain only to mass production processes in terms of investment. They can be developed as one form of marginalcost-reducing and fixed-cost-increasing innovation. The significant increase in fixed costs can reduce labor and other variable costs per unit output. The increase in fixed costs implies further economies of scale, hence more concentrated production than distributed production. Here, concentrated production means that large firms are more efficient, so production should be concentrated in a few large or super-large firms. Distributed production means relatively more minor firms spread across different regions better serve society. Mass production tends to be concentrated, while individually tailored production tends to be distributed. Intelligent manufacturing could be developed as another type of marginalcost-reducing and fixed-cost-increasing innovation: to reduce the marginal cost of producing an additional brand or model in the same or similar product family instead of an additional unit of the same product. Reducing the marginal cost of production variability or flexibility leads to developing reconfigurable manufacturing systems (RMSs) and flexible manufacturing systems (FMSs). An RMS can quickly adjust its production capacity and functionality within a product family in response to sudden market or intrinsic system changes. This capacity for quick adjustments allows for rapid change in its structure and hardware and software components (Mehrabi, Ulsoy, and Koren 2000). An FMS possesses flexibility that will enable the system to react in case of changes (Buzacott and Yao 1986). Production flexibility generally falls into one of two categories: 1) machine flexibility, the ability to produce new product types or the same product in different orders of operations; 2) routing flexibility, the ability to use

286

Chapter 5

multiple machines to perform the same operation on a product and absorb large-scale changes. Intelligent factory systems could also incorporate these systems; then, different products might be produced by the same production line. The RMS or FMS can counterweight concentrated mass production to serve segmented or niche markets. With the development of 3D printing, many people think that in the future, all goods might be printed out by individuals who design them or by selecting and downloading a design available for free or at a price. 3D printing would be the ultimate RMS or FMS; the more versatile this technology, the more distributed and tailor-made manufacturing will become. Jeremy Rifkin argues in his book The Third Industrial Revolution that production in the future will be more distributed, with individuals becoming prosumers instead of producers or consumers (Rifkin 2011). Whether this will become a reality in the future depends on the relative costs between a mass-made commodity being delivered to an individual and that made by an individual using 3D printing. The most likely scenario will be a combination of the two approaches in the future: some goods are made by mass production and delivered to households. In contrast, individuals make others at home or in the local community “production centers” using 3D printing.

4.4. Energy and Water Energy, information, and physical materials are the three critical elements for all economic activities. Since all automation, robots, and AI systems need energy to drive them, energy cost may become the primary determinant in choosing between robot-dominated systems and humandominated systems. Because of global warming, traditional cheap fossil fuels are being replaced in favor of green energy, even though the proven global total reserves of fossil fuels have increased over the past 20 years (Dale 2021). The transition from fossil fuels to renewable and green energy is necessary for the sustainable development of humankind. The logistics of supplying fossil fuels and electricity can be readily automated and managed by AI systems. Renewable energy facilities are also working largely with little human intervention. With current technology, renewable energy still cannot replace fossil fuels entirely in the foreseeable future. Nuclear energy can replace fossil fuels, but there is strong opposition to nuclear energy because of safety issues. The primary fossil fuel demand is from power stations, heating installations, and transport vehicles. Heating installations usually use fossil fuels because it is cheaper than electricity,

Production in the Robotic Age

287

and transport vehicles use fossil fuels because electricity or hydrogen is difficult to carry with sufficient energy density. Electricity and hydrogen are secondary energy that is produced by consuming primary energy. Technology for the mobile storage of electricity and hydrogen, which can reduce pollution and CO2 emissions, is expected to mature soon. Hydropower, nuclear power, and coal-fired or gas-fired power stations are usually concentrated electricity producers. In the Machine Age and probably in the Robotic Age, concentrated production also implies high efficiency in pollution reduction investment. Solar energy and renewable energy will be exploited in a distributed way in every building, even in every household, as envisaged by Jeremy Rifkin (Rifkin 2011). With homeproduced electricity and heat from solar energy and renewable energy, energy-saving technology might reach a point where households can satisfy their own energy needs, including 3D printing at home or in the local community. Then many residents will become true prosumers. However, household energy sufficiency should not be a condition for the mature stage of the Robotic Age. Mass electricity production in power stations could coexist with the household production of solar and renewable energygenerated electricity. Water power has been an important power source since ancient times, and water is also essential for human life and the production of many goods. Hydropower stations and water supply can be managed automatically even by the current technology. However, maintenance is still mainly conducted by human workers. Like completely autonomous management of other sectors, the nearly utterly autonomous water resource management requires technological progress in multifunctional robots that can replace human service workers.

4.5. Building Many construction robots have been developed and tested (Bock 2015; Gambao and Balaguer 2002), and 3D printing technology may substantially transform the construction industry (Wu, Wang, and Wang 2016; Tay et al. 2017). For now, however, robots have not been widely used in construction. Countries with a booming construction industry in the past two decades, such as China, still rely heavily on human builders because building jobs are less standardized than factory assembly lines. Besides the residential, office, and commercial buildings, which are somewhat standardized, factory buildings, airports, ports, roads, bridges, railways, mines, etc., have their unique requirements and particular land conditions.

288

Chapter 5

Automation in the building industry requires progress in three directions: 1) multifunctional robots which can do a builder’s job at comparable costs and support single-task construction robots (STCRs); 2) building standardization to enable robot-oriented design, and 3) building materials that facilitate the application of robots and 3D printing in construction. Robot-oriented design should facilitate using construction robots in the construction process and installing service robots in the building when used. With mature technologies in these three directions, the building industry can be autonomous, like manufacturing in the Robotic Age, with minimum human intervention.

4.6. Commerce The service industry is currently the most important sector for employment, with the manufacturing industry hiring fewer and fewer workers due to automation and robotics. The service sector often involves face-to-face interaction, which has been challenging to automate. There are different levels of interaction in the service industry. In some shops, customers pay shop assistants who give them the goods, usually standardized commodities. In some situations, customers pick up the goods they feel have the right quality, then pay at the checkout. In food and beverage serving places where food and drinks are prepared and served on-site, the services include ordering by customers, preparing by kitchen staff, serving by waitpersons, and paying by customers. All these processes can be automated or carried out by robots in principle, and some have been automated to some extent. 4.6.1. Online and offline shopping Online shopping has made Amazon, Alibaba, and eBay among the most valuable worldwide. Alibaba and eBay are online shop platforms, while Amazon is an online shop and web service provider and a platform for other retail firms. Tencent provides a WeChat shop platform. The popularity of online shopping leads to huge volumes of deliveries by courier services, which stimulates the development of courier services and reduces their costs due to the economies of scale. The increase in labor productivity of the courier service and transport, in general, helps the development of ecommerce. In the future, the commerce landscape will likely have three components: online shops at Internet platforms, around-corner convenience shops/delivery collection points, and in-person boutiques in shopping malls. Consumers will buy all standardized and cheaper goods online, especially those that need not be tried. Courier services will deliver the purchased goods to the

Production in the Robotic Age

289

purchaser’s home or work address. Those that cannot be delivered because the purchasers are not home can be deposited at convenience shops/collection points. These convenience shops are where residents buy everyday goods in urgent need because they cannot wait for delivery from online shops. In the future, these shops might also operate like Amazon Go, where customers can be automatically logged in from their smartphones and pay for the goods by scanning them. The in-person boutique shopping malls will be more centrally located, and consumers can try the goods before they decide. These shopping malls will mainly deal with expensive goods that must be fitted by trying them on. With the development of virtual reality, such inperson shopping malls might also be in decline. 4.6.2. Automated Customer Service Larger firms in the service industry need a customer service section to deal with customer inquiries. Automation used by call center agents to handle customer inquiries is usually called agent-assisted automation, which includes two basic types: desktop automation and automated voice solutions. Desktop automation uses software programs to make it easier for the call center agent to work across multiple desktop tools. Automated voice solutions may recognize the voice messages of customers and provide customers with information in the form of pre-recorded audio files. The key benefit of agent-assisted automation is compliance and error-proofing. Automated voice solutions still need to be improved, and human service is often required. They will be fully autonomous in the Robotic Age when the systems become more intelligent.

4.7. Transport, Post, and Logistics Railways have largely been automated, and airliners rely heavily on autopilot during their flight, which can be further automated and improved. The current technology is sufficient to operate automated vehicles on tracks and roads if there is no error from (other) human road users. Automated highway systems (AHS) or smart roads will become integral to an intelligent transportation system in the Robotic Age. Cars with power steering and automatic speed controls can respond to information and instructions from the intelligent transportation system. Their intelligent autonomous driving system can also read passive road markings, communicate with each other, and organize themselves without the driver’s intervention. Fully autonomous cars, trucks, and lorries, in combination with an intelligent highway system, will allow closer vehicle spacing and

290

Chapter 5

higher speeds, enhance road safety by reducing the opportunity for driver error, and increase fuel economy. With the progress in communication technology, the number of letters handled by the post office has dropped dramatically. However, the volume of parcels delivered by various courier services has massively increased because of online shopping by customers. In China, many courier stations are set up in or near residential communities to handle and temporarily store courier parcels. Some courier services set up self-service lockers in communities for receivers to collect their parcels with passwords sent by text messages. Companies like Amazon have been experimenting with sending packets by drones or autonomous vehicles. In the Robotic Age, if online shopping is still needed at the current level, courier services will be fully automated. Multifunctional robots in storage houses will load parcels on autonomous delivery vehicles that transport parcels to the purchasers’ community’s self-collecting storage. Purchasers can collect their parcels from the self-collecting warehouse or by their home or community service robots. Logistic services for firms will also be fully automated, from the storehouses of suppliers to the storehouses or production lines of downstream producers, wholesalers, or retailers.

4.8. Accommodation and Catering Many aspects of hotel service have been automated; for example, online booking and payment are conducted without human intervention on the hotel side. The technology for robotic cleaners is mature, and robots have been used to deliver snacks and consumables to customers in hotel rooms. The service robots for fully autonomous accommodation services still need more progress in robotic technology for making multifunctional robots and machine sense and handling of soft materials, with costs comparable to human workers. Technology for robots to clean bathrooms and tabletops is relatively easy to realize. Making the bed and changing the sheet is more challenging than handling solid objects because soft materials need to be felt and tackled. These difficulties will be overcome, and multifunctional robots capable of handling soft materials could be developed and applied in the accommodation industry. Hotel buildings will also be designed to facilitate the performance of robots. The catering industry already uses robotic chefs and waitpersons (Lin et al. 2021; Gao et al. 2022), but there is still room to improve their performance. Chefs, waitpersons, cleaners, and kitchen assistants who wash and prepare food ingredients can all be replaced by robots. Robotic chefs need not be

Production in the Robotic Age

291

mobile, while waitpersons have to be mobile to serve patrons. Making cleaners and kitchen assistants fully autonomous is more challenging than making robotic chefs and waitpersons because they have to perform more variable jobs. Multifunctional robots would replace them. Restaurants and canteens should also have robot-oriented designs to facilitate the performance of robots.

4.9. Information and Communication The information and communication industries have been mainly automated in terms of providing regular information and communication services. What remains for human workers are sales, managerial, and maintenance workers, whose jobs require multifunctional robots and redesigning the whole system to facilitate the performance of robots. The positions of the sales, administrative, and maintenance workers in the information and communication industries are similar to those in other sectors. They will become autonomous with similar technologies and around a similar time.

4.10. Finance The main business activities of the finance sector do not involve handling physical objects; thus, they are among the easiest to be automated. The most difficult to automate is the everyday chores because the human ability to handle physical objects of different sizes, shapes, and textures has been perfected over millions of years. In contrast, computation and logical judgment need no manual operations or sensory functions, so they are much easier to automate. The recent developments in financial technology (Fintech) have automated many human operations in the finance industry (Gomber, Koch, and Siering 2017; Gai, Qiu, and Sun 2018; Goldstein, Jiang, and Karolyi 2019). It could all be automated in the near future. The concern that machines cannot have a big picture or the intuition of successful human traders is generally groundless. Unlike manufacturing assembly lines, finance does not need physical raw materials that require more sophisticated robotic systems. The automation of sales, managerial, and maintenance jobs in the finance industry will be similar to those in other sectors. In the Robotic Age, digital currencies replace notes, coins, and credit money. Based on blockchain technology, the cryptocurrency bitcoin has been widely traded as an investment or payment (Zohar 2015; Extance 2015). The People’s Bank of China (PBOC) has piloted the first central bank-

292

Chapter 5

issued digital currency in 2022. Digital currencies have many advantages, such as faster payments, no or low transaction fees, low costs to produce, more security, etc. Since a digital currency can be decentralized like bitcoins, it has been envisaged that firms can issue their digital currencies backed by their future products or services to raise capital (De Bono 1994; Birch 2020), which may change the finance industry fundamentally. Since transactions with digital currencies can be carried out between counterparties using digital wallets directly, people or firms need not deposit their money in banks. The role of retail banks will be severely undermined. At least three types of digital currency will be used in the future: 1) cryptocurrency based on cryptography, such as bitcoins, which have no issuers and are not supported by any assets; 2) digital fiat currency issued by central banks, such as the digital RMB issued by the PBOC; 3) digital currency issued by firms and supported by their products and services, or a basket of fiat currencies. The digital currency plan announced by Facebook in 2019 would be a digital currency supported by a basket of fiat money. Corresponding Internet or other communication infrastructures will support these digital currencies' coexistence and determine their exchange rates.

4.11. Real Estate The real estate industry is a project planning, management, and service industry for buildings for residences, offices, and other purposes. Some of these functions are similar to commerce; others are similar to service or finance. Automation is more readily achieved for those jobs responsible for payment in transactions. Like in other industries, sales, managerial, and maintenance jobs need more sophisticated robots.

4.12. Renting and Commercial Services These services can be roughly divided into two categories: 1) those that involve managing physical objects such as buildings, vehicles, grounds, roads, parks, etc., and 2) those that do not involve managing physical objects. Automating physical object managing services requires multifunctional robots and supporting AI systems to service the physical objects. Equipment used by the business per se also needs service by its multifunctional robots or external repairers. Automation of non-physical services is relatively simple because it requires only AI systems. Non-physical services are often called professional services. The term “professional” conveys the meaning of rigorous training and expertise.

Production in the Robotic Age

293

These professions include doctors, lawyers, teachers, accountants, tax advisers, financial advisers, management consultants, architects, journalists, etc. In the Chinese industrial classification, doctors belong to the health sector, and journalists belong to the culture sector. There are some common features in those professions: 1) they need a relatively high level of formal training to do the job; 2) they have special knowledge which is not readily accessible to ordinary people due to technical jargon, the requirement of foundational knowledge, cost of the information in traditional media, or physical inconvenience; 3) they provide solutions to problems met by ordinary people or advise people; 4) their solutions and ways of thinking are generally logic-based. These features make professionals more vulnerable to advanced levels of AI. AI systems will replace the experts and provide the same or better advice than average experts. IBM Watson and DeepMind’s AlphaGo have shown AI’s potential to provide professional services. Quill can produce automated articles in various areas, including sports, business, and politics. ChatGPT shows more potential for AI systems to replace human experts (Kirmani 2022; Gordijn and Have 2023; Stokel-Walker 2022). Google Bard has functions similar to ChatGPT and might have greater potential in updating its database and integrability. From the cases of Watson, AlphaGo, Quill, ChatGPT, and Bard, we can see that AI at present has been able to perform deep learning, retrieve data, and analyze data to provide a solution that is superior or comparable to those produced by human experts. With further developments in AI, the costs of such systems will drop enormously, and their function will be much more potent than Watson, AlphaGo, Quill, ChatGPT, and Bard. By then, there will be little room left for human experts.

4.13. Scientific Research and Technological Services The impact of AI and automation on science and technology can also be examined from two perspectives. One is the automation of physical activities in scientific research, which increases the productivity of scientists and technicians, implying that some jobs could be lost due to increased productivity. The other is the automation of mental activities by AI in the discovery or invention process, which is the mental part of the research. Automation has been extensively employed in laboratories, especially industrial research and development, to perform the repetitive operations of laboratory technicians or research students. Researchers and technicians must still prepare and load the experimental materials into the robotic experiment systems. With the development of multifunctional robots in the future to prepare experimental materials and load them into automatic

294

Chapter 5

systems, experiments will be fully automated, and there will be no or little human involvement in the physical activities of research except for those human scientists who enjoy doing them. Automation of mental activities by AI will replace researchers’ mental activities and produce research results and conclusions directly. Eureqa’s ability to develop several physical laws describing the pendulum's movement in only a few hours (Schmidt and Lipson 2009) has already demonstrated the great potential of computer algorithms in replacing human mental activities in research. DeepMind’s AlphaFold has further demonstrated the power of AI in research (Jumper et al. 2021). While thousands of structural biologists have elucidated the 3D spatial structure of about 190K proteins by experiments over many years, AlphaFold has done 200 million in one year. It seems inevitable that computer algorithms and robots will replace most scientists in their mental activities. An essential part of research is to write research papers for publication. As scholar-bureaucrats rely increasingly on publication (journal) metrics for evaluating research contribution, competition for top journal publication has become increasingly intense, and research journals are proliferating. Editors of (top) journals dealing with enormous submissions often rely on factors such as affiliation, status, influence, connections, topic familiarity, media value, writing style, etc., rather than scientific value to decide whether to send a submission for review or reject it right away (desk rejection), which is detrimental to scientific progress. This is understandable, however, given human editors' limited knowledge, speed, and capacity in processing various research papers. A priority in applying AI to scientific research is to develop AI editor systems that can objectively, speedily, comprehensively, and adventurously assess scientific contributions without indulging in minor issues such as style, wording, and referencing. As an interim objective, AI editor systems can focus on screening submissions for review and selecting reviewers. The long-term aim is to perform the jobs of both editors and reviewers, plus copy-editing for style, wording, grammar, typos, referencing, etc., for human authors. Competition for publication in top journals also affects their reviewers’ behaviors. Reviewers of leading (social science) journals, who are also researchers and, on other occasions, authors, tend to spend more time criticizing an article’s insufficient explanation of research motivation or inadequate recounting of previous studies than evaluating the research findings. Often reviewers reject research papers because they do not have the knowledge and ability to understand and assess the research, suppressing

Production in the Robotic Age

295

genuinely original and innovative works. Authors often have to go through several rounds of revision and resubmission to satisfy the reviewers. Thus, if a researcher does not have previous connections, publishing in top journals becomes more haphazard, depending on a person’s luck to get two more charitable reviewers. In the near future, as an interim stage before AI completely takes over research from human researchers, AI reviewer systems should be developed to give an objective and competent evaluation of a research paper. Concerning reviewers’ criticism on explaining research motivation or recounting previous studies, ChatGPT- and Google Bard-type AI systems can write and revise these parts of research papers so that researchers can focus on the important part of the research, the research findings. ChatGPT and Bard can answer questions and compose texts according to users’ requests, which could help people write literature reviews, develop research ideas, and prepare research papers (Dowling and Lucey 2023; Pavlik 2023). However, the information provided by ChatGPT is not all reliable, and sometimes it makes up evidence and conclusions (Dis et al. 2023), casting doubts on how much humans can rely on it. ChatGPT and Google Bard are large language models (LLM) that rely on statements in the training database to form autoregressive wording choices. They are not making logical inferences or induction, so they are good at reviewing literature and suggesting research ideas but not performing logical analysis for human researchers. To help logical analysis in research, we need AI based on logical inference, i.e., deduction and induction, besides AI based on LLM. It is imaginable that in the Robotic Age, general-purpose machine learning algorithms that can learn new skills by themselves will be developed such that there is no need to use specific programs such as AlphaGo, AlphaZero, or AlphaFold. The specific expertise of such generalpurpose learning algorithms is acquired when assigned to address issues in a particular field.

4.14. Residential Services and Repair Services Home automation or smart home (domotics) refers to the automation of household appliances and duties. Currently, robot cleaners are widely used as home service robots. In the future, home automation could be achieved via two channels: 1) intelligent home systems with sensors and switches connected to an AI control system (the central hub) which controls lighting, heating, ventilation, air conditioning, security, etc.; and 2) multifunctional robots which can perform different household tasks rather than a single task such as cleaning floors. The AI central hub is connected to the Internet.

296

Chapter 5

Home appliances such as washers/dryers, ovens, refrigerators/freezers will also be controlled by the central hub. Human owners can adjust the central hub if they are not happy with the parameters set by the AI systems. Home appliances, electricity, and the building will have problems and damage that must be fixed or repaired. In the mature stage of the Robotic Age, problems and damage could be fixed by multifunctional robots. Before reaching that level of automation, homeowners must call human repairers to correct such problems. Some problems may have to be fixed by human workers. If this is true, we can classify tasks involving physical objects into three types: 1) performed by single-task robots, 2) performed by multifunctional robots, and 3) performed by human workers. What can be performed only by human workers tends to be tasks that need manual dexterity rather than high-level intelligence.

4.15. Public Utilities and Environment Management Public utilities, such as the supply of electricity, gas, and water, are generally automated in their everyday operations. Automating their maintenance and repair jobs requires multifunctional robots like in other industries. Environment management similarly requires multifunctional robots to check, enforce regulations, and fix issues that damage the environment. Besides mobile multifunctional robots patrol and address environmental problems, fixed sensors, and equipment will also be installed and connected to the IoT to monitor and protect the environment. Like in all other industries, to replace all human workers, the autonomous system needs to be feasible technologically and cost-effective compared with using human workers. Multifunctional patrol robots can handle safety, public security, and environmental issues in cities and towns. With surveillance well distributed in all public places, big data analysis can warn of any emerging threat, and emergency response systems can respond promptly to accidents and natural disasters. Surveillance, sensors, and patrolling robots are all connected with the IoT, which links automatic response devices for various purposes to make a city safer and more secure. Such a city can be called a smart city.

4.16. Education The education system is among the longest-existing institutions in human society. It has become even more critical in modern times. With the new challenges created by technological progress, people look to education for

Production in the Robotic Age

297

solutions. Many economists do not think that the unemployment caused by new technologies is an issue of workers being replaced by machines and computer software. Instead, they consider it a skill mismatch problem between workers and employers. For those economists, education and retraining are crucial to solving unemployment. However, teachers may be in danger of being replaced by robots and computer software. Information and communications technology (ICT) has influenced education greatly, especially in delivering lectures in higher education. Although many university teachers still prefer the chalk and blackboard approach for providing courses, most teachers have shifted to prepared PowerPoint files and projectors. The chalk and blackboard method requires lecturers to be very familiar with the taught materials and able to write legibly and perhaps elegantly. Still, writing on the blackboard or whiteboard with colored pens takes time and leaves less time for explaining the content. With prepared files, lecturers have more time to introduce concepts and theories and need less time to memorize lecture contents. As students often have electronic files of their teacher’s lecture notes before the class, they need not take notes from the blackboard. This tends to have two different effects on students’ learning outcomes: 1) not taking notes gives students more time and attention to listen to their teacher and think actively; 2) students lose the benefits of focusing attention and learning through notetaking. AI may well take over the role of teachers when robots possess a course’s relevant materials/knowledge better than human teachers and can continue to learn. Lecturing might not be more complex than competing in the Jeopardy game at the highest level. Both teacher robots and virtual AI teacher systems could replace human teachers in the Robotic Age. It is imaginable that with cloud computing, robots will know more than human teachers about the subject they teach and can retrieve and process data. With proper training on performance, it can model the best practice of human teachers and may eventually replace human teachers. Massive Online Open Courses (MOOCs) and recorded video courses from the best teachers will remove the need to attend classroom lectures and, eventually, have lecture theaters and classrooms. This is readily achievable, as the Open University in the UK and similar universities in other countries have replaced classroom teaching with lectures on television since the 1960s. What is difficult to replace is the practical and experimental classes since virtual practices and experiments may have different effects and outcomes

298

Chapter 5

than the real ones. Local laboratory centers, run autonomously and owned publicly, could address the practice and experimentation issues. The education landscape in the future is illustrated in Fig.5-6. Publishers provide textbooks, reading materials, and recorded video lectures for selfstudy. In kindergarten, parents and human and/or robotic carers look after small children. Robotic and human teachers teach schoolchildren. University students learn via recorded video lectures or attend online courses and have practices and experiments in local laboratory centers. Online tutorial services use discipline-specific AI systems or human tutors to help students who need help make sufficient progress. The future examination could be organized by society and qualifications certified by society. Since most jobs are automated, and people can all have a decent life without a job, the demand for degrees and diplomas will largely disappear or become a competition like sports. People study a subject or a skill because they genuinely love doing that.

Fig.5-5 Application of ICT, AI, and robots at the early and mature stages of the Robotic Age. At the early stage, students use video lectures and AI systems to learn a subject by self-study and get help from human tutors. Students use ICT, AI, and robots at the mature stage.

The most vigorous opposition to this scenario will come from universities and teachers because it will lead to the near demise of universities and the teaching profession. However, most university students study a subject not because they enjoy what they study and are interested in its topic. Instead, they study a subject because they need a degree to find a well-paid job, and most courses they take in their university study have little use in their

Production in the Robotic Age

299

eventual jobs. The most important thing for students is the degree certificate from a top university rather than useful knowledge and skills acquired through university study. In his Wealth of Nations, Adam Smith strongly criticized colleges' monopoly on people’s credentials (Smith 2010). Teachers most likely will strongly oppose the wide application of autonomous teaching. They will also object to recognition of ability acquired by learning from AI systems. However, we must develop an international or national certifying mechanism to recognize knowledge and capacity gained through AI teaching systems. Investing in national or international AI certifying or accrediting systems will provide maximally equal opportunity to all people.

4.17. Health and Social Work Medical doctors used to rely heavily on their senses in diagnosing diseases. Besides asking about a patient’s complaints, case history, and other relevant information, a doctor uses their vision, hearing, smell, and touch to collect signs that help diagnose. Instruments such as stethoscopes were invented to facilitate observation. Later, laboratory tests were developed to gather the information that a doctor’s sensory organs cannot readily obtain; so were machines widely used for diagnosis in hospitals, such as X-rays, ultrasound, nuclear magnetic resonance imaging (MRI) machines, etc. Doctors now rely heavily on machines and laboratory tests to make diagnoses, much less than before on physical examinations. This makes automation for diagnosis the easier part of the automation of health care. The complete automation of laboratory and machine tests requires multifunctional robots to replace nurses and technicians who collect samples from patients, feed samples into machines, or guide patients to devices for scanning and measuring. Treatments currently conducted by doctors and nurses also require multifunctional robots to perform. In the future, robots might and probably will be more dexterous than human surgeons in performing operations and replace human surgeons entirely.

4.18. Culture, Sports, and Entertainment AI is at its best when dealing with logic-based knowledge. Comparatively, it is less capable of dealing with human emotions and mimicking manual dexterity. Culture is more emotional and manually involved, but there are many disciplines in which knowledge and logic are more important than emotions and manual dexterity. Information technology may influence culture in two aspects. One is that computers, game consoles, and the Internet may play the role of the medium both for electronic games and

300

Chapter 5

traditional forms of entertainment and culture. The other is that robots or computer programs can deal with and generate cultural products. With the advance of ICT, online shops, downloads, and file exchanges have eclipsed the traditional music industry. The music of superstars has become more accessible to ordinary people. Films, especially classical ones, are freely accessible to Internet users. Consumers with mobile Internet access can be entertained by various cultural products available on the Internet. Technology increases the reach of superstars and probably their incomes while leaving little room for local talent to develop and support themselves by providing services to local consumers. AI has begun to acquire the ability to produce cultural products. The Quill mentioned early can write various news reports and analyses. In July 2012, the London Symphony Orchestra performed a composition, Transits – Into an Abyss, composed entirely by an AI algorithm called Iambus running on a cluster of computers (Ball 2012). Researchers at the University of Malaga in Spain designed Iamus. It has already produced millions of unique compositions in the classical modernist style. We can expect more AI composers in the Robotic Age. Simon Colton, a professor of creative computing at the University of London, has built an AI program called “The Painting Fool” (Colton 2012). The software can identify emotions in photographs of people and then paint an abstract portrait that conveys their emotional state. It can also generate imaginary objects using techniques based on genetic programming. Dall-E2 and Midjourney can generate more realistic images from natural language descriptions at higher resolutions. They will evolve rapidly to have more robust functionality, higher integrability, and higher originality in their output. In the future, performing androids may take over jobs from actors and actresses in dramas and operas. Images in movies could be generated by computer programs such that human actors and actresses are no longer needed.

4.19. Public Administration and Social Securities Many government functions can be automated, especially those requiring no manual handling, for example, collecting taxes from businesses and individuals, paying social welfare benefits into personal accounts, evaluating benefits applications, and allocating funds to various public offices. Government functions, which handle physical objects and people, will require multifunctional robots. Government functions tend to be more

Production in the Robotic Age

301

knowledge- and rule-based activities, which can be more readily automated than artistic activities.

5. The Singularity The singularity point is when AI reaches the level of general human intelligence and surpasses it, as predicted by some AI researchers and futurologists. Some people think this moment will never come, and others dread that robots will take over the world after this moment. From the viewpoint of economics, a dramatic transformation of human society does not need robots with general intelligence surpassing that of human beings; that is, it does not need the arrival of the singularity moment. Specialized AI existing in each specialized human area and surpassing human intelligence in each technical area will be sufficient for a thorough transformation of human society. Surpassing human intelligence in each specialized area is just the continuation of the division of labor. The division of labor first produced two classes: the ruler-managers and the laborers. The ruler-managers acquired and maintained their status through their initial personal or familial quality and cultural or religious preaching. Then a third class evolved from the laborers, servants, or ruler-managers’ retinue. The laborers further diverged into farmers, artisans, and eventually, the workers of different professions in modern societies. The rulermanagers evolve into government officials, capitalists, and managers. In the evolution of the division of labor, human expertise in production becomes more and more specialized. Workers at each level need to know very little to perform their jobs routinely. For example, central bankers hold positions among the most important ones for a country’s economy, but what they need to know seems no more than a few simple rules: when the economy booms, the monetary policy should not be too loose; when the economy is in recession or growing too slow compared with the long-term trend, the monetary policy should not be too tight; when there is high inflation, the interest rate should be raised, and money supply should be reduced; when aggregate demand shrinks, and the economy lacks liquidity, the interest rate should be cut, and money supply should increase. The sophisticated models built by modern economists probably have minimal impact on central bankers’ interest rate decisions. While the division of labor makes human input (of time) in one unit of output smaller and smaller, individual human beings' consumption capacity has not changed much since the birth of Homo sapiens. There is no substantial specialization in people’s consumption capacity. However,

302

Chapter 5

people’s taste in consuming cultural and artistic products is fundamentally shaped by their education, upbringing, social status, religious belief, ideologies, and traditions. The change in people’s consumption over history is not a consequence of people’s consumption capacity; instead, it is a consequence of the ability of human beings to obtain the goods and services for their consumption. Therefore, although people have evolved with general intelligence, robots do not need superhuman general intelligence to completely replace human intelligence in production and perform better than human workers. What is required for AI is to fully characterize, imitate, and perfect the division of labor in the intelligence or human brain power for various human productive activities. The division of work in society includes two aspects, one is physical or manual, and the other is intellectual or mental. A profession's intellectual or mental aspect does not always require general intelligence. The workers at the early assembly lines did not need much intelligence or brain power other than cognition to perform their jobs. Even intelligent workers, university teachers, senior corporate managers, bankers, and financial market traders use specific intelligence in their routine jobs rather than general intelligence. From the above analysis, the singularity moment may come. Still, it is unnecessary to develop superhuman general intelligence to transform the production from a human intelligence-dependent process to a human intelligence-independent process. When humans are free from work, society must be reorganized regarding the distribution of goods and services, public decision-making, and environmental management. There are also the issues of what people do when they need not work, what motivates people, and what the meaning of life is when people need to do no work for a living. All those issues will arise even if AI never reaches the singularity. What will happen if the singularity does arrive? If general AI in robots surpasses human intelligence, robots will take over and care for the world. There is no reason to believe that super-intelligent robots will be evil. Human intervention in designing and developing intelligent robots must have mainly instilled good characteristics into robots. If human evolution is any guide for robot evolution, good will prevail. When good superintelligent robots manage the world, humanity will be looked after better than in the world managed by humans themselves. Humans to the robots will be somehow like family pets to humans. Humans might stop thinking about conquering or exploring the universe and developing new products for the masses. They will be more concerned with enjoying life and working on interesting things.

Production in the Robotic Age

303

6. Summary The progress in robotics and AI will reduce labor’s contribution to output. Single-task robots have already performed repetitive tasks. At the early stage of the Robotic Age, some human workers might still work on factory floors because loading parts to assembly lines, maintaining equipment, fixing machine breakdowns, and sending output to wholesalers need multifunctional robots rather than single-task robots. These remaining workers have the supervisor role, but they only supervise robots. There will be resistance, especially from senior managers, against using robots for senior managers’ jobs. At the mature stage of the Robotic Age, multifunctional robots will be responsible for loading raw materials and intermediary products to assembly lines, packing completed products, processing orders, and loading goods to transport vehicles. Multifunctional robots will also perform a large part of service jobs. AI systems will perform professional jobs.

References Azadeh, Kaveh, René De Koster, and Debjit Roy. 2019. “Robotized and automated warehouse systems: Review and recent developments.” Transportation Science 53 (4):917-945. Ball, Philip. 2012. “Computer science: Algorithmic rapture.” Nature 488 (7412):458-458. Becker, Gary S. 2009. Human capital: A theoretical and empirical analysis, with special reference to education. Chicago, IL: University of Chicago Press. Berman, Barry. 2012. “3D printing: The new industrial revolution.” Business Horizons 55 (2):155-162. Birch, David. 2020. Digital Currency Revolution. London: Centre for the Study of Financial Innovation. Bock, Thomas. 2015. “The future of construction automation: Technological disruption and the upcoming ubiquity of robotics.” Automation in Construction 59:113-121. Bogue, Robert. 2014. “What future for humans in assembly?” Assembly Automation 34 (4):305-309. Brynjolfsson, Erik, and Andrew McAfee. 2014. The second machine age: Work, progress, and prosperity in a time of brilliant technologies: WW Norton & Company. Brynjolfsson, Erik, and Andrew McAfee. 2015. “Will humans go the way of horses.” Foreign Affairs 94 (4):8-14.

304

Chapter 5

Buzacott, John A, and David D Yao. 1986. “Flexible manufacturing systems: a review of analytical models.” Management Science 32 (7):890-905. Cannon, Walter B. 1929. “Organization for physiological homeostasis.” Physiological Reviews 9 (3):399-431. Colton, Simon. 2012. “The painting fool: Stories from building an automated painter.” In Computers and Creativity, edited by Jon McCormack and Mark d’Inverno, 3-38. Berlin and Heidelberg: Springer. Dale, Spencer. 2021. BP statistical review of world energy. London: BP Plc David, Paul A. 1990. “The dynamo and the computer: an historical perspective on the modern productivity paradox.” American Economic Review 80 (2):355-361. De Bono, Edward. 1994. The IBM Dollar. London: Centre for the Study of Financial Innovation. Dis, Eva A. M. van, Johan Bollen, Robert van Rooij, Willem Zuidema, and Claudi L. Bockting. 2023. “ChatGPT: five priorities for research.” Nature 614:224-226. Dowling, Michael, and Brian Lucey. 2023. “ChatGPT for (finance) research: The Bananarama conjecture.” Finance Research Letters:103662. Extance, Andy. 2015. “Bitcoin and Beyond.” Nature 526 (7571):21. Gai, Keke, Meikang Qiu, and Xiaotong Sun. 2018. “A survey on FinTech.” Journal of Network and Computer Applications 103:262-273. Gambao, Ernesto, and Carlos Balaguer. 2002. “Robotics and automation in construction [Guest Editors].” IEEE Robotics & Automation Magazine 9 (1):4-6. Gao, Feng, Shuo Li, Yue Gao, Chenkun Qi, Qiyan Tian, and Guang-Zhong Yang. 2022. “Robots at the Beijing 2022 Winter Olympics.” Science Robotics 7 (65):eabq0785. Garcia-Haro, Juan Miguel, Edwin Daniel Oña, Juan Hernandez-Vicen, Santiago Martinez, and Carlos Balaguer. 2020. “Service robots in catering applications: A review and future challenges.” Electronics 10 (1):47. Goldstein, Itay, Wei Jiang, and G Andrew Karolyi. 2019. “To FinTech and beyond.” The Review of Financial Studies 32 (5):1647-1661. Gomber, Peter, Jascha-Alexander Koch, and Michael Siering. 2017. “Digital Finance and FinTech: current research and future research directions.” Journal of Business Economics 87 (5):537-580. Gordijn, Bert, and Henk ten Have. 2023. “ChatGPT: evolution or revolution?” Medicine, Health Care and Philosophy:1-2. Gronau, Reuben. 1976. “Leisure, home production, and work--the theory of the allocation of time revisited.” Journal of Political Economy 85 (6):1099-1123.

Production in the Robotic Age

305

Hwang, Jinsoo, Seulgi Park, and Insin Kim. 2020. “Understanding motivated consumer innovativeness in the context of a robotic restaurant: The moderating role of product knowledge.” Journal of Hospitality and Tourism Management 44:272-282. Jumper, John, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, and Anna Potapenko. 2021. “Highly accurate protein structure prediction with AlphaFold.” Nature 596 (7873):583-589. Kirmani, Ahmad R. 2022. “Artificial Intelligence-Enabled Science Poetry.” ACS Energy Letters 8:574-576. Li, Siguang, and Luogeng Hua. 1959. Scientists talk about twenty-first century. Beijing: Juvenile and Children's Books Publishing House. Lin, Ting-Yu, Kun-Ru Wu, You-Shuo Chen, Wei-Hau Huang, and Yi-Tuo Chen. 2021. “Takeout service automation with trained robots in the pandemic-transformed catering business.” IEEE Robotics and Automation Letters 6 (2):903-910. Ma, Qing-Ping. 2013. “Inter-and intra-industry wage inequality and potential impact of automation in the economy.” In Wages and Employment: Economics, Structure and Gender Differences, edited by A. Mukherjee, 97-126. New York: Nova Science Publisher Mehrabi, Mostafa G, A Galip Ulsoy, and Yoram Koren. 2000. “Reconfigurable manufacturing systems: Key to future manufacturing.” Journal of Intelligent Manufacturing 11 (4):403-419. Mincer, Jacob. 1958. “Investment in human capital and personal income distribution.” Journal of Political Economy 66 (4):281-302. Oshika, Yasuaki. 2023. “Robotics tested at rice paddies with eye on full automation.” The Asashi Shimbun, Last Modified May 10, 2023, accessed May 25, 2023. https://www.asahi.com/ajw/articles/14895684. Pavlik, John V. 2023. “Collaborating With ChatGPT: Considering the Implications of Generative Artificial Intelligence for Journalism and Media Education.” Journalism & Mass Communication Educator:10776958221149577. Petty, William. 1751. Political Arithmetic. Glasgow: Robert and Andrew Foulis. Piketty, Thomas. 2014. “Capital in the twenty-first century.” In Capital in the twenty-first century. Cambridge, MA: Harvard University Press. Rifkin, Jeremy. 2011. The third industrial revolution: how lateral power is transforming energy, the economy, and the world. New York: Macmillan. Rodrigue, Jean-Paul, and Theo Notteboom. 2021. “Automation in Container Port Systems and Management.” TR News (334):20-26.

306

Chapter 5

Schmidt, Michael, and Hod Lipson. 2009. “Distilling free-form natural laws from experimental data.” Science 324 (5923):81-85. Shahrubudin, Nurhalida, Te Chuan Lee, and Rhaizan Ramlan. 2019. “An overview on 3D printing technology: Technological, materials, and applications.” Procedia Manufacturing 35:1286-1296. Shell, Richard L., and Ernest L. Hall. 2000. Handbook of industrial automation. Boca Raton: CRC press. Smith, Adam. 2010. The Wealth of Nations: An inquiry into the nature and causes of the Wealth of Nations. Petersfield: Harriman House Limited. Stokel-Walker, Chris. 2022. AI bot ChatGPT writes smart essays-should academics worry? Nature. doi:https://doi.org/10.1038/d41586-02204397-7. Tay, Yi Wei Daniel, Biranchi Panda, Suvash Chandra Paul, Nisar Ahamed Noor Mohamed, Ming Jen Tan, and Kah Fai Leong. 2017. “3D printing trends in building and construction industry: a review.” Virtual and Physical Prototyping 12 (3):261-276. Torresan, Chiara, Andrea Berton, Federico Carotenuto, Salvatore Filippo Di Gennaro, Beniamino Gioli, Alessandro Matese, Franco Miglietta, Carolina Vagnoli, Alessandro Zaldei, and Luke Wallace. 2017. “Forestry applications of UAVs in Europe: A review.” International Journal of Remote Sensing 38 (8-10):2427-2447. Vis, Iris FA. 2006. “Survey of research in the design and control of automated guided vehicle systems.” European Journal of Operational Research 170 (3):677-709. Walker, Tom. 2007. “Why economists dislike a lump of labor.” Review of Social Economy 65 (3):279-291. Wu, Peng, Jun Wang, and Xiangyu Wang. 2016. “A critical review of the use of 3D printing in the construction industry.” Automation in Construction 68:21-31. Zohar, Aviv. 2015. “Bitcoin: under the hood.” Communications of the ACM 58 (9):104-113.

CHAPTER 6 HUMAN RESOURCES, NATURAL RESOURCES, AND POLLUTION

The Robotic Age is an age of affluence, which inevitably requires more resources than an age of want. Usually, more physical and intangible goods imply more production and waste. Humans and human brains have been refined for millions of years; they are extremely energy efficient and might be material efficient. Computers and robots are much poorer performers than human and human brains in energy efficiency. To sustain an advanced Robotic Age society, there must be sustainable energy sources and physical materials. Many mineral resources are depletable, and it would be very costly to find substitutes when a natural resource is depleted. The positive side is that in the Robotic Age, the material needs of society have largely been met, so the demand for materials and energy has stabilized. Recycling materials and renewable energy might satisfy all the new demands for materials and energy. With artificial intelligence (AI) systems and robots replacing human workers, the labor supply becomes abundant and less critical to production, enabling most people to develop their interests and skills to fulfill their purpose in life. This chapter will examine how human and natural resources will be used in the Robotic Age and how pollution will be dealt with. It will also explore attention span and employment as resources for firms.

1. Human Resources in the Robotic Age There are two opposing views on the future employment prospect of human workers. On one side, many people still believe that robots and AI systems are only good at performing routine or repetitive manual and cognitive tasks but not at non-routine tasks, especially those requiring creativity (Autor, Levy, and Murnane 2003). Conversely, some people, including Stephen Hawking, Bill Gates, and Elon Musk, think AI could become too smart to benefit humanity (Sainato 2015). Some researchers are worried by the prospect of a jobless society, so they have proposed various schemes which might keep people in jobs (Susskind 2020). As shown in the preceding

308

Chapter 6

chapters, robots and AI systems will replace human workers in almost all jobs in the Robotic Age, of which people have dreamed since ancient times. However, complete automation of production and service leads to another question, what will humans do in the Robotic Age besides consumption and leisure? Since humanity creates robots and AI to serve humans, humans should still be the ultimate lawmakers of society and the top overseer of robots before AI systems are allowed to be more intelligent than humans. If humans are still the lords of robots, human artistic creativity and taste will ultimately determine what creative robots will do if they eventually come into existence.

1.1. Demand for Human Talents and Technological Progress Since 1977 China has made finding and cultivating talented people a priority. There are two interesting phenomena. More and more people, mainly researchers and engineers, are paid millions of Chinese yuan as talent at various government levels, far above what their average peers earn for similar jobs (Qiu 2009). The other is that people often deplore that intellectuals no longer have the strength of character or dignity (Fenggu, 仾 僘) that traditional Chinese scholars are thought to have. In China, there are many stories of how scholars in the past kept their pride in the face of powerful monarchs and warlords or how monarchs and senior officers begged scholars to serve their regimes. Mencius best summarizes this dignity: “He cannot be led astray by riches and honor, moved by poverty and privation, or deflected by power or force” (Bloom and Philip 2009). Why do Chinese scholars no longer have the dignity that characterized their predecessors in ancient times or even during the Republic of China period? This can be explained by the changing supply and demand of talent as society progresses. On the supply side, a few individuals spontaneously obtained professional skills because of their innate physical or mental superiority. People with superior physical or mental strength often had high self-pride and strong character and tended to be less obedient. Later, professional skills were acquired through training, but excelling required higher levels of innate physical or mental strength. The rise of machines made physical strength unimportant in society except for sports and street brawls, but mental strength was still highly regarded. Before printing made books a cheap consumer commodity, learning knowledge and researching social and natural issues required a good memory and a solid analytical mind, which most people do not have. Thus, the few people reputed for their

Human Resources, Natural Resources, and Pollution

309

academic or practical achievements became treasured by sensible rulers because some routine tasks had to be completed by those few people. When universal education and professional training have standardized routine procedures, people with average or below-average levels of intelligence can follow the procedure and complete tasks satisfactorily. Procedure standardization of normal operations and universal education made it challenging for the rulers to separate highly intelligent people from those with average levels of intelligence, and it also made the few more intelligent people more dispensable to even the sensible rulers and employers. Therefore, talented people struggle to stand out from people with average intelligence and lose the bargaining power to deal with rulers or employers. Progress in AI will further erase the impact of differing intelligence on performance in most professions except for competitions barring machine assistance. Technological progress increases the talent pool of people who can perform a job competently but makes it more difficult for truly outstanding people to stand out and be appreciated by society, especially the government and business leaders. Table 6-1 summarizes the abilities valued in different historical periods. Table 6-1 Ability required in different periods Subperiod Paleolithic Neolithic

Bronze Manual Age

Appreciated Quality Physical strength Physical strength Problem-solving, innovation Physical strength Work skills Knowledge learning and creation Organizational ability Military ability Physical strength Work skills

Iron

Knowledge learning and creation Organizational ability

Acquisition means Innate Innate Innate, selfdevelopment Innate Experience Innate, selfdevelopment Innate, selfdevelopment Innate, selfdevelopment, family training Innate Experience, family training Innate, selfdevelopment, public and private schools Innate, selfdevelopment

Chapter 6

310

Military ability Work skills First Industrial Revolution

Analytical, mathematical ability or literary ability Organizational ability Work skills

Machine Age

Second Industrial Revolution

Analytical, mathematical ability or literary ability Organizational ability Work experiences Organizational ability

Information Revolution

Robotic Age

Communication and networking skills responsibility and independent thinking

Innate, selfdevelopment, family training Experience, apprenticeship Innate, selfdevelopment, university education Innate, selfdevelopment Experience, apprenticeship Innate, selfdevelopment, university education Innate, selfdevelopment, college training Opportunities and luck Innate, selfdevelopment, university training Innate, selfdevelopment, professional training, university training School and university training, selfdevelopment

On the demand side, leaders and managers tend to prefer subordinates who can do their jobs satisfactorily and behave obediently and obsequiously (Beu and Buckley 2004; Einarsen, Aasland, and Skogstad 2007). When choosing someone for a position, they need to balance a person’s competence and obedience and make a trade-off between them. When competent candidates are rare, a sensible leader, employer, or monarch has to tolerate their disobedience when they cannot find competent and obedient candidates. Leaders need to compete for talented people, and they compete by showing humility toward talented people and giving them wealth and high status. Talented people can afford to have their dignity intact and show the strength of their characters because they know the leader or monarch depends on them for success. With universal education and standardization of procedures and operational protocols, people with average or belowaverage intelligence can perform these jobs. When qualified candidates are

Human Resources, Natural Resources, and Pollution

311

abundant, some would be obsequious, so a leader can choose obedient and competent candidates. This has two impacts on the behaviors and careers of talented people: 1) it induces more people to behave obediently and obsequiously to curry favor from the leader, and 2) it excludes principled as well as talented people from important or senior positions, so they miss the opportunities to gain experience and improve their competence through doing an important job. From the above analysis, we can understand that there are people with real talent and strength of character. Still, nowadays, they will not be successful in their careers and become famous, so the general public would never hear of them. Networking and making superiors happy became the most useful talent after the Second Industrial Revolution. In early hominids, the most important talent was probably physical strength, but collaboration between individuals for dominance in the herd might have occurred, as shown in chimpanzees (Wrangham and Glowacki 2012; Willhoite 1976; De Waal and Waal 2007). The few people with extraordinary physical strength or with superior intelligence to find solutions to their herd’s problems would become leaders. Offspring inheriting their genes and social conditions were likelier to become the next generation of leaders. Gradually aristocratic families formed and became the ruling class. The ordinary members of the herd or tribe generally would submit willingly or habitually to their authority or leadership. The invention of weapons and hunting and farming tools reduces the impact of differences in physical strength among people in society. By improving their skill in using weapons or devices, people with less power might become a match for people with extraordinary physical strength in fighting, hunting, or farming. On the one hand, this increases overall labor productivity; conversely, it gives the ruling class more choices in assigning tasks. Before the invention of machines using nonbiological power, people’s physical strength still matters. Compared with activities relying on physical strength, intellectual activities were more difficult for ordinary people to master before the invention of printing or even before steam engine-powered printing. With hardly any tools to facilitate mental activities, only a few people with excellent memories and outstanding analytical abilities could become scholars or officials in charge of culturerelated activities. Except for some fatuous and cruel rulers, monarchs and leaders would show great respect to those scholars or sages and compete with each other to hire them. The appearance of books reduced the importance of a good memory for academic work, and educational institutions lessened the importance of

312

Chapter 6

analytical abilities in problem-solving by training students with standard protocols, which increased the talent pool for performing intellectual activities. The more the supply of competent people in academic work, the more choices the rulers or employers had, and the more difficult for the scholars to have their characters and maintain their dignity in front of the rulers or employers. The invention of printing made books more widely available and further reduced the importance of a good memory in spreading general knowledge and technologies. However, a good memory was still vital for doing scholastic jobs, and an excellent analytical ability was also essential for solving problems encountered in engineering projects and academic work. Looking for information in books costs time and could be less convenient than accessing knowledge stored in one’s brain, especially for inspiration or afflatus to emerge. Therefore, scholars with superior intelligence could still show the strength of their character and maintain their dignity before rational rulers or leaders. The progress in information and communications technology (ICT) makes good memory and analytical ability much less important, perhaps no longer critical for scholarly or engineering jobs. Relevant information can be readily accessed through the Internet and e-books; computer software can perform sophisticated statistical analysis much better than the best mathematical minds without machine help. With the expansion of higher education, thousands and thousands of postgraduates with Ph.D. degrees or other doctorate degrees are churned out each year globally. Many of them would not be able to complete their research projects in the days without analytical software. Because of the oversupply of Ph.D. graduates, it is not easy in many countries, including China, for many Ph.D. holders to find a job commensurate with their training and expertise (Patton 2012; Mervis 2016). Since supply far outstrips demand in terms of appropriate academic assignments, scholars cannot afford to have the strength of character in front of government leaders and employers. Because of the oversupply, although government and business leaders have constantly preached the importance of human talents to economic growth, talented employees, especially those in higher education and scientific research jobs, tend to be treated poorly regarding academic freedom and career opportunities than before. Many universities have introduced corporate performance evaluation systems and currently conduct annual performance appraisals of academics with short-term KPIs (key performance indicators), managing academics like industrial workers.

Human Resources, Natural Resources, and Pollution

313

1.2. The Curse of Success: How More Scholars Can Hinder Scientific Progress The mass production of university graduates, including Master’s and Ph.D. degree holders, enabled by ICT profoundly impacts the job market and scholastic institutions. First, networking and communicating skills become more important than intelligence/intellect in career success (Nabi 2003). Second, scholars become more and more like business people and conduct research as a business to advance personal income and status rather than pursue their curiosity in nature and society. Third, scholars are managed as factory workers, evaluated on their short-term output measured by quantitative metrics (Braun et al. 2010; Lane 2010), and fired at the end of the contract when metric targets are not met; university managers no longer bother to understand a scholar’s research contents and efforts. Fourth, the academic world is partitioned into networked circles, and networking becomes a priority for academics and a route to career success. Academic conferences are the venue of networking, particularly for social sciences. Fifth, academic stars are more likely to be those good at publicizing themselves, obtaining funding to hire researchers to work for them, and networking to form a circle of vested interests. Sixth, it becomes more difficult to challenge the mainstream view. New ideas that challenge established views are usually barred by peer-reviewed mainstream journals, Internet forums, and preprint websites sponsored by academic societies or mainstream researchers. The increased academic publication output incentivized by short-term metrics slows scientific progress (Chu and Evans 2021), and papers and patents become less disruptive over time (Park, Leahey, and Funk 2023). There are many mechanisms through which the increased outputs may slow down scientific progress. Let us assume that a few people can think critically and make research breakthroughs in the population. Most researchers lack critical thinking and research to expand applications of current theories or work out some details of the fringes of existing ideas. They would follow the mainstream propositions and believe the mainstream view correctly understands the world. Then, the following are possible reasons for more researchers and more publications to slow down scientific progress. First, the output by the majority will consume resources that more talented researchers might use for genuine breakthrough research. This may be explained by borrowing biochemical mechanisms of competitive inhibition and partial agonists with low intrinsic activity (Ariens 1954). Competitive

314

Chapter 6

inhibition is the phenomenon in which one chemical substance inhibits the effect of another by competing with it for binding or bonding to receptors or enzymes; the inhibitory effect depends on the relative concentration (and affinity to the receptor) between the two chemicals. If a chemical has the full effect when binding to the receptor, it is a full agonist. If a chemical has no effect when binding to the receptor, it is a pure antagonist. If a chemical has some effect when binding to the receptor, it is a partial agonist, which means it has lower intrinsic activity than a full agonist. People with good analytical ability and independent thinking are like full agonists, and people lacking analytical ability and independent thinking are like partial agonists. The more scholars, the stronger the inhibition to the activity of people with good analytical ability and independent thinking. Second, the majority’s output will consume the scientific community’s attention and bury authentic breakthrough findings in information overloads or pollution. This may be explained by product inhibition in biochemistry. Product inhibition is a type of enzyme inhibition where the product of an enzyme reaction inhibits its production (Cleland 1963). If two similar chemicals (substrates) A and B are catalyzed by an enzyme, chemical A leads to the desired product A, and chemical B gives rise to a defective product B. If there is too much chemical B, then more product B will reduce the production of product A (and product B per se). Similarly, the more mediocre papers published by researchers lacking good analytical ability and independent thinking, the more difficult for innovative articles to get published. Third, people with low analytical power and no independent thinking tend to reject new ideas because they lack critical thinking and cannot appreciate real innovations. This resembles the phenomenon that partial agonists have low intrinsic activity, so their binding to the receptor will prevent the binding of the full agonists, leading to lower effects (Hoyer and Boddeke 1993). A similar example of competitive inhibition but beneficial to human health is that nonpathogenic bacteria can contain pathogenic ones by outcompeting them for local resources. Pathogenic bacteria can cause diseases when the distribution of nonpathogenic bacteria is disrupted. This phenomenon is called dysbacteriosis or dysbiosis (Petersen and Round 2014). Comparing creativity with pathogenicity and scientific breakthroughs with diseases (somehow inappropriately), more researchers lacking creativity will hinder those with creativity in making scientific breakthroughs. Fourth, since the majority of the scientific community, who tend to follow the authority opinion, is the judge of new theories in science, it becomes

Human Resources, Natural Resources, and Pollution

315

more difficult for a new correct idea to be accepted by the scientific community because now its proponent has to convince more members who cannot appreciate the new theory. This can also be understood in analogy with competitive inhibition, the outcome of which is determined by the relative concentration between two chemicals. The more the antagonist or partial agonist, the stronger the inhibition to the effects of the agonist. When science was a concern for a few people with outstanding critical thinking and analytical power, a new idea only needed to win over them to form the majority opinion. In the Robotic Age, when AI systems and research robots can perform most tasks previously conducted by researchers, the law of the negation of negation may prevail, and scientific research returns to its previous curiosity-driven nature. People who conduct research as an income-earning job rather than for curiosity can enjoy their leisure with a guaranteed basic income that supports a decent life and personal dignity. Researchers driven by curiosity can collaborate with AI systems and research robots to satisfy their desire to solve myths in nature or society. Networking ability, vital in current research career success, especially in the social science research community, will no longer matter.

1.3. Human Talents in the Robotic Age Whether human talents are still needed in the Robotic Age depends on whether robots replacing them can outperform their jobs well. Many people think that the areas in which robots and AI are unlikely to outperform humans are creativity, artistic taste, and insight into complex situations. If these qualities are unique to humans and cannot be replicated by AI, human workers will continue to hold key positions in industrial organizations. However, the development of AI may erode human superiority in these areas and make AI more capable than human brains. Moreover, most jobs in the current economy do not need creativity, artistic taste, and insight into complex situations. We considered creativity in research and development in the early chapters. In inventions and innovations, afflatus and inspiration are often cited as evidence that human thinking is unique and machines cannot replace humans in those areas. According to this view, an important discovery or invention has some mysterious or illogical factors that cannot be attributed to logical inference or induction. For example, Isaac Newton discovered gravity after watching an apple falling from a tree (Keesing 1998); the German chemist Friedrich August Kekulé discovered the ring shape of the

316

Chapter 6

benzene molecule after having a reverie or daydream of a snake seizing its own tail (Rothenberg 1995). Although we are still unsure about how afflatus or satori arises, we should be confident that its beneficial outcomes are consequences of logical reasoning. Afflatus or satori in scientific research occurs in humans not because humans have superior memory accessing and information processing ability but because they only have insufficient memory accessing and information processing ability. This insufficiency prevents them from combining relevant premises to reach the conclusion or result they seek. An occasion or event that triggers the association between premises and conclusion becomes afflatus or satori. For AI with much larger readily accessible memory, much higher memory-accessing ability, and much higher information processing capacity, afflatus or satori in humans will not be needed in arriving at the correct solution. This is like Go chess; for a complex situation, there should be an optimal strategy. However, human players usually struggle to work out these optimal steps and need the afflatus or satori. But for AI players like AlphaGo (Silver et al. 2016), AlphaGo Zero (Silver et al. 2017), and AlphaZero (Silver et al. 2018; Kissinger 2018), their powerful capacity can readily lead to the optimal strategy without a need for afflatus or satori. Humans might have an advantage over AI regarding artistic tastes and the ability to appreciate and create art, but these might not be as impregnable as many believe. Intelligent machines become better at assessing people’s emotions by examining facial expressions, postures, and body movements with deep learning than human psychological counselors. They may also be able to pick up artworks that satisfy human artistic taste. Since AI composers or painters will create music and paintings for human appreciation, which lacks logical underpinning, AI systems probably could only mimic existing successful artworks. They would not be able to create successful revolutionary artworks, as changes in human preference are often illogical and are likely resistant to logical analysis. However, we should not push this view too far because revolutionary artworks are often the remainders among many failed attempts; even the works of the great painter Vincent van Gogh were not appreciated during his lifetime. If we allow AI systems to be radical revolutionaries in arts and fail in their endeavors, they might eventually create art like abstract and cubist paintings (Hong, Peng, and Williams 2021). Many people doubt the ability of AI systems to have a strategic insight into complex situations. Some consider AI good only at the detailed and precise

Human Resources, Natural Resources, and Pollution

317

analysis of costs and benefits and weak at the big picture and overview. Those people have probably overemphasized the mystical side of human strategic insight. Before the match between AlphaGo and 18-time world champion Lee Sedol (Lim 2017), commentators in China generally thought that AlphaGo would lack the strategic overview of human players and excel in local calculations. The actual games showed that AlphaGo won because of its strategic overview rather than local calculations. The only game it lost was probably due to an unexpected program bug. By introspection, we can readily find that we decide by comparing costs with benefits based on the likelihood of success, which is further constrained by weaknesses in our human nature. Except for weaknesses in human nature, human strategic insight can be replicated by AI with more powerful information accessing and processing capacity. One thing that might be difficult for AI to outperform humans is the gambler psyche. In history, especially military history, many outstanding commanders bet against all odds, took the riskiest and most failure-prone approaches, and succeeded in achieving their objectives. For example, would an AI commander launch the landing at Inchon, as General Douglas MacArthur did in 1950 (Heinl 1998)? Would an AI commander launch an attack on a larger army like General Han Xin at Jingxingkou in 205 BCE? AI commanders who make decisions based on the highest probability of success might not obtain such achievements. However, we should not attach too much significance to these military successes in a wide range of decision-making processes because these successes might be a few among much more failed attempts or exceptions due to the mistakes of their rivals. General Han Xin’s success probably depended more on the bookish stupidity of the Zhao Kingdom Commander, Confucian scholar Chen Yu than his tactics (Sima 1993). Had Chen Yu used General Li Zuoju’s plan, China’s history might have been different. Many of the greatest commanders in history might be the luckiest because their unintelligent or arrogant rivals made them successful. It might be possible to add a gambling trait into AI decision-making mechanisms, such that the AI system would take a measured risk against a higher probability choice from time to time. From the above analysis, we may conclude that while AI systems and robots can take over most of the daily routine jobs from human workers, there might still be some jobs to which humans can add unique contributions. Innovative artworks and risk-taking decision-making might need human contributions the most. However, such unique contributions from human artists, researchers, and visionaries should not be exaggerated. The illogical and emotional components in human artworks, risk-taking decisions, and

318

Chapter 6

inventions might be mimicked by randomly inserting unconventional risktaking factors into AI.

1.4. Human Beings as Members of the Community Robots are developed to replace humans in workplaces, not to replace humans in the human community. Human beings as members of a community will always be needed. With the progress of AI, intelligent robots can read human emotions and minds and respond accordingly. Will humans accept such intelligent robots as equal members of their society rather than treating them as smart tools? Will humans treat robots differently from treating fellow humans? Before robots acquire general intelligence comparable to humans, it is unlikely that people will treat robots as fellow human beings, even if robots and AI systems have performed all human jobs. At that stage of development, humans as social animals must have fellow humans in a community to be comfortable, discussing and debating issues, planning and initiating new projects which might be carried out entirely by robots, and appreciating and enjoying the arts or music together. Being a community member is an essential human resource in the Robotic Age. An excellent human listener and an applauder to other humans are much more valuable than a robotic listener. In the early stage of the Robotic Age, fellow humans are needed to form a community; even in more advanced stages, robots could be as intelligent and “human” as humans. Also, in the early stage of the Robotic Age, robots might be unable to perform some specific jobs in the community, and humans have to carry out such community jobs. Therefore, a responsible and active community member is one of the community’s most important assets, contributing to common happiness. As robots become more intelligent during the Robotic Age, how to deal with robots and coordinate the relationship between humans and robots becomes a key issue in the community. If robots have motivation, ego, and intelligence comparable to human general intelligence, humans may have to accept robots as members of the human community. The question may become whether humans want to let robots have motivation and ego if AI can do so and whether humans can prevent robots from acquiring motivation and ego as well as superhuman general intelligence if AI can let robots have those capacities. When this becomes an issue, humans need to be vigilant for any possible signs of robots becoming self-motivated and acquiring an ego and act to prevent any humans or robots from instilling

Human Resources, Natural Resources, and Pollution

319

self-motivation and ego into robots. This could be a profession for humans in the Robotic Age.

1.5. Humans as Legislators Humans are still the masters of society during the Robotic Age, and robots are intelligent machines working for humans. If this is true, legislators are the most important human jobs in the Robotic Age. Humans need to make laws and regulations to govern the behavior of humans and robots and the relationship between robots and humans as well as between humans. In modern democracies, the most developed countries except Switzerland are indirect representative democracies in which citizens elect members of parliament, the head of state, or the head of government to make decisions and manage a country on their behalf. But members of parliament and heads of government can make decisions against the opinion of the majority who elected them. In a direct democracy like Switzerland, important decisions must be made by referenda (Wagschal 1997; Papadopoulos 2001). In the Robotic Age, because of the convenience provided by ICT, a system of direct democracy is more likely to prevail in developed countries. If, in the Robotic Age, direct democracy is more prevalent, all adult humans are legislators who should study and investigate law-related issues and participate in the legislative processes. Robots can enforce the law and help draft and compile statutes and regulations. If humans want to stay as the leaders of society, they must be the legislators; in a direct democracy, all humans should be legislators. To maximally use AI and robotics for the benefit of humanity, humans must try to make the best laws to govern the behavior of humans and robots and the relationship between humans and robots. In the advanced stage of the Robotic Age, whether robots should have a human or superhuman level of general intelligence and whether human-level- or superhuman level-intelligent robots should have the same legislation rights as humans will become critical factors in determining the future of society and humanity. Table 6-2 summarizes robots’ different intelligence levels and their future roles.

Chapter 6

320

Table 6-2 Different intelligence levels of robots Intelligence Assembly/firstline robots

Single-function service robots

Multifunction robots

Robots with creativity

Robots with emotional intelligence

Robots with general intelligence

Mobility

Function

Ability to complete assigned operations Ability to complete assigned operations

immobile or in very constrained ranges Mobile in designated areas

First line workers

Ability to perform different tasks Ability to perform different tasks and to invent and discover new things Ability to perform different tasks and to interact with humans emotionally Comparable to humans with motivation and selfenhancing learning ability

Mobile

Performing repetitive tasks in factories, etc. Conducting simple operations that need to move around Performing maintenance and service jobs Performing inventive and innovative jobs

Mobile

Collaborating with humans in tasks involving emotions

Colleagues, partners, and family members.

Mobile

Performing all human jobs

humans

Mobile

Replace

First-line service workers Maintenance workers and supervisors Scientists and engineers

One possibility is that humans will put a brake on the level that the robotic general intelligence can reach, including motivation and ego/selfconsciousness. In this sense, robots will always be intelligent machines under the control of humans because of a conscious choice by humans. Another possibility is that human-level or superhuman-level robotic general

Human Resources, Natural Resources, and Pollution

321

intelligence is unachievable by human research and development. If this is the case, questions like whether robots should have a human or superhuman level of general intelligence and whether human-level- or superhumanlevel-intelligent robots should have the same legislation rights as humans become irrelevant. The third possibility is that humans, by informed decisions or mistakes, let robots have human or superhuman levels of general intelligence. It will probably be unstoppable that human-level- or superhuman-level-intelligent robots have the same legislative rights as humans. In the third scenario, humans will relinquish their status as the master of society and the Earth, as robots evolve much faster than humans in becoming more intelligent.

2. Natural Resources in the Robotic Age Although the most straightforward production function in economics needs only capital and labor or even only labor, natural resources are indispensable in the real world. In the Robotic Age, production still needs natural resources. In Chapter 1, we noted that our ancestors were hunter-gatherers who collected natural foods for physiological needs. If humans were still relying on natural foods, the Earth would not have been able to support over eight billion people. The primary natural resources needed in the Robotic Age will not differ much from those required for the Machine Age: farming land for producing foods, water for industry and life, minerals and other raw materials for making physical products, materials that carry energy to power all production and consumption activities, and biological materials.

2.1. Natural Resources for Producing Foods The current agricultural technologies require farming land to produce food, and animal husbandry relies heavily on farming to provide animal fodder. At present, developed countries are concerned with the overproduction of agricultural products, so they need to take measures to discourage production. Many residents in developed countries switch to organic farming products because they are worried about the problems caused by chemicals and genetic modification used by modern farming. Generally speaking, it is unlikely that traditional organic farming could supply the current world population (Connor 2008), and modern farming is a great success story for humanity. Suppose the total fertility rate in the least developed countries, when they reach the middle-income level, drops to the level of the current middle-income countries. In that case, the world population will be stabilized at a level at which the current farming technologies will be

322

Chapter 6

sufficient to meet all the demand for food in terms of nutrition and energy. Food will not cause a severe resource issue as long as the world population stays at a resource-sustainable level. Suppose the world population becomes unsustainable due to unexpected factors. In that case, future generations will have the wisdom to legislate on the fertility rate allowed. Population is a factor whose increase can be limited by legislation and law enforcement. From what happened in the developed and middle-income countries, legislation is unlikely needed to limit world population growth. The onechild policy implemented by the Chinese government since 1980 so effectively slowed down the population growth in China that the policy had to be stopped by 2015 (Whyte, Feng, and Cai 2015). China’s population decreased by about 810,000 in 2022. The starvation poor people face in the least developed countries is partly caused by poor natural conditions but mainly caused by poor management and political instability. Without civil war, an efficient administration would relieve many of the least developed countries from starvation with the current farming technologies. An inefficient and corrupt government is often the primary cause of slow economic growth. Political stability and protection of property rights are the first steps for ensuring economic growth. For some small failed states with no change or negative growth for an extended period, the United Nations might negotiate or impose a mandated international administration to establish social stability and help those countries progress into a thriving modern economy. This could be started with one failed small state, for which one large developed country can be given a strong mandate by the United Nations, like the United Nations Trust Territories. The Internet of Things (IoT) may further increase agricultural productivity and the output per unit of farming land. With sensors embedded in soils, irrigation systems, and facilities for applying fertilizers and other chemicals (pesticides and growth agents), water, fertilizers, and pesticides can be used more efficiently and timely to save resources, reduce pollution, and increase yields. The IoT, automation, and robotics in farming will sharply reduce the number of human workers in agriculture and might lead to non-human smart farms. Although the IoT and robotics may significantly increase agricultural production, the existing technologies are sufficient to feed the global population well. One possible deviation in food production from farming, husbandry, and fishing is to produce food ingredients from more basic organic materials acquired by growing fungi or other suitable microbes in fermentation tanks.

Human Resources, Natural Resources, and Pollution

323

Quorn is the brand name for a commercially marketed mycoprotein made from Fusarium venenatum, a fungus, in a fermentation vat. Mycoprotein has been used as a meat analog (Derbyshire and Ayoob 2019). Producing food materials in fermentation vats might become comparable to farming regarding energy efficiency and saving land resources. Fungi produced in fermenters can then be used as raw materials alone or in combination with other raw materials (produced by farming or raising animals) to produce artificial food ingredients. Various meat alternatives are often made from beans, nuts, grains, and mycoprotein for vegetarians and vegans. In the future, more food ingredients might be prepared with raw materials produced in fermenters. There might be future disruptive breakthroughs in food production, such as chemically produced foods. For example, science fiction has long imagined starch directly synthesized from water and carbon dioxide in chemical reaction chambers instead of plants through photosynthesis (Cai et al. 2021). From the current understanding of the relevant science and technologies, with limited fossil fuel reserves on the Earth, unless nuclear energy becomes the primary energy source for synthesizing food, chemical reactions to produce food will not be energy efficient. Fossil fuels can be viewed as solar energy preserved by ancient plants through photosynthesis and consumed by us now. Agriculture is also transforming solar energy into food for human and animal consumption. With the present preference for renewable energy sources, modern agricultural technologies are the more energyefficient choice for food production.

2.2. Water Water is an essential resource for all human societies on Earth. The water supply depends mainly on natural conditions; freshwater is more abundant in some areas and less so in others. For the dry regions, if they have additional resources and industries to accumulate wealth, such as Saudi Arabia and the United Arab Emirates, it is possible to produce fresh water from the desalination of seawater (Younos and Tulou 2005). In the future, the cost of seawater desalination might be sharply reduced to such a level that a middle or even low-income country could mainly rely on desalination for its water supply. Pollution has been a major threat to freshwater resources for many countries with good water sources. In China, some rivers and lakes are so heavily polluted by industrial wastes that they cannot be used as drinking water sources (Bao et al. 2012). It can be expected that with increased income and

324

Chapter 6

improved living standards, people will be more conscious of pollution and associated health risks. Society will have stricter legislation on pollution, the IoT will help the effective implementation of the law, and the economy will operate with clean technologies. A significant problem with water resources is an uneven distribution among different regions and between different seasons. Often there is too much water in an area which causes severe flooding, while other parts are in severe drought. Or in one season, there is no rain such that even drinking water cannot be adequately supplied, whereas, in another season, too much causes a flood that leads to heavy losses of property, even human life. In the future, technologies might be developed to intervene in the quantity of rain and its regional distribution. Alternatively, more reservoirs could be built in the regions with more rainfall and canals to transport water to the dry areas. As there is a vast amount of seawater and the desalination technology has more room to be improved, water resources should not be a major issue in the Robotic Age. Although glaciers might be disappearing, the natural water cycle will still exist to supply new fresh water. Water is not a depleting resource as long as we do not pollute water sources and there are no astronomical catastrophes.

2.3. Energy The Robotic Age will need energy, probably much more than the present Machine Age, as discussed in the preceding chapter, because all the robots, AI systems, the IoT, and so on need energy to operate. The most likely scenario is that more energy is required, but energy efficiency is also much higher, i.e., the output per unit of energy consumed is much higher. In 2021, the total energy consumption in the world was 595.15 exajoules, among which oil, natural gas, coal, hydroelectricity, nuclear energy, and renewable energy are 184.21 (30.95%), 145.35 (24.42%), 160.10 (26.90%), 25.31 (4.25%), 40.26 (6.76%) and 39.91 (6.71%) exajoules respectively (Dale 2022). The present dominant energy sources, petroleum, natural gas, and coal, are depletable. Known oil reserves are estimated at around 244.4 billion tons (1,732.4 trillion barrels) at the end of 2020 (Dale 2021), which can only supply about 58 years if current production (4,221.4 million tons in 2021) remains stable. The major proven resources (in cubic meters) of recoverable natural gas in the world are 188.1 trillion cubic meters at the end of 2020 (Dale 2021), which can last for 46.6 years at the current level of production (4,036.9 billion cubic meters) (Dale 2022). The 1,074,108 million tons of recoverable coal reserves can be consumed for 139 years at the current production level (Dale 2021).

Human Resources, Natural Resources, and Pollution

325

As we can see from the above data, petroleum, gas, and coal may not last hundreds of years to supply the energy needed during the Robotic Age, which will likely stay for thousands or millions, if not billions, of years. There are more coal reserves than petroleum, but coal has been considered to cause global warming and has gradually been replaced by natural gas as fuel in developed countries. Reducing the use of coal to slow down global warming may leave more coal reserves for future use. Clean coal technologies developed in the future may play a more critical role in providing energy for the early Robotic Age. Water power can be considered a sustainable energy source, the Industrial Revolution’s first significant nonhuman/animal power source. Hydroelectricity has provided a part of the energy need for a long time, but water power is unlikely to satisfy the total energy need in the Robotic Age. Nuclear energy could be the main energy source for the Robotic Age. In theory, nuclear fuels (mainly uranium) are also depletable. Although uranium is a fairly common element (approximately as common as tin or germanium) in the Earth’s crust (Meinrath, Schneider, and Meinrath 2003), it can be economically extracted only where it is present in high concentrations. It is estimated that for 130 USD/kg, the current economically recoverable resources of uranium in the world are 6,147,800 tons of uranium metal (tU), which is enough to last for 100 years, with world production at 54,224 tU in 2019. The OECD also estimated that at the higher cost of USD 260/kg, the total identified resources amounted to 8,070,400 tU (Chmielewski 2008; Grancea, Mihalasky, and Fairclough 2020). Like every other natural metal resource, every tenfold increase in the cost per kilogram of uranium leads to a three-hundredfold increase in available lower-quality ores that would become economical. If other energy sources are depleted, and renewable energy technology is still insufficient to satisfy the energy need in the Robotic Age, more uranium sources will become economical. Advanced breeder reactors could be designed to utilize recycled or depleted uranium and all actinides efficiently. In that case, there is 160,000 years’ worth of uranium in total conventional resources with phosphate ore at 60100 US$/kg of uranium. A breeder reactor generates more fissile material than it consumes (Weinberg 1960). Thermal breeder reactors use neutrons in the thermal spectrum to irradiate thorium-232 to produce uranium-233 as fissile fuel. Fast breeder reactors use neutrons in the fast spectrum to irradiate uranium-238 (99.3% of natural uranium) to produce plutonium239 as fissile fuel. Current light water reactors use only the very rare uranium-235 isotope (0.7% of natural uranium). Nuclear reprocessing can

326

Chapter 6

make used uranium waste reusable. It has been estimated that there are up to five billion years of uranium-238 for use in these power plants. Breeder technology has been used in several reactors, such as the Phénix in France and the BN-600 reactor in Russia. Still, the high cost of reprocessing fuel safely makes it economically less justifiable for the time being. This consideration of costs will change if other energy sources are depleted. If controlled nuclear fusion using hydrogen isotopes is realized, nuclear power can be used for an infinite time. On December 13, 2022, US Energy Secretary Jennifer Granholm announced that US scientists had made a breakthrough in nuclear fusion research; the National Ignition Facility at the Lawrence Livermore National Laboratory in California successfully produced a nuclear fusion reaction resulting in a net energy gain for the first time. In many developed countries, people are opposed to using nuclear energy because of safety concerns over nuclear waste and radioactive contamination, and they prefer a wide usage of renewable energy. Renewable energy could soon play a significant role in the world’s energy consumption with the continual progress in various forms of sustainable and green energy technologies. For now, renewable energy cannot compete with fossil fuels or nuclear energy without government subsidies, and renewable energy technologies still cannot produce enough power to drive most large manufacturing factories, which are generally energy thirsty. No matter whether it can replace nuclear energy in the long run or not, renewable energy could play a key role in supplying domestic energy needs. It is easy to calculate how much solar energy a household can extract from the equivalent surface area it owns, from which it should be clear that households living in high-rise buildings could not satisfy all their energy needs from domestic solar energy devices. Renewable energy sources such as solar, wind, tidal, and bioenergy generally come from sun radiation (tidal energy is mainly from the lunar and solar gravity and the Earth’s rotation). The energy contained in fossil fuels also comes from the sun’s radiation. The energy contained in fossil fuels has been accumulated for millions of years from solar radiation. Hence, the energy density of fossil fuels is much higher than that of renewable energy, which utilizes only the current solar energy. With more progress in sustainable and energy-saving technologies, the cost of producing renewable energy might be reduced to a level comparable to that of fossil fuels. Then the objective of 100% renewable energy might be achievable. Renewable energy production is more distributed, making high-energy density applications less convenient. In the Robotic Age, if renewable

Human Resources, Natural Resources, and Pollution

327

energy technologies cannot provide enough energy for society and fossil fuels are exhausted, nuclear power will become the only choice for meeting our energy needs.

2.4. Minerals and Building Materials Most minerals are not abundant enough for infinite exploitation. Many ores for producing metals and non-metal materials are essential to modern technologies and consumer goods. More such mineral resources will be needed in the Robotic Age, so a sustainable supply of those minerals or the extracted materials is essential. Although over-mining can exhaust mineral resources, recycling metals and non-metals in various products is possible. The fewer resources for a particular metal or non-metal substance, the higher the price and the higher the motivation to extract the substance from scrapped products. There is a difference between mineral resources and exhaustible energy resources. When oil, natural gas, and coal are used, the energy they contain is released. The second law of thermodynamics does not allow the retraction of released energy without consuming more energy. The second law of thermodynamics states that the total entropy will either increase or remain the same for an isolated system in any cyclic process. In an isolated system, neither energy nor matter can enter or leave; entropy can be considered a measure of the amount of energy unavailable to do work. Therefore, although there is a law of conservation of energy, used energy (in fuels) cannot be recycled. However, the law of conservation of matter states that for any system closed to all transfers of matter and energy, the system’s mass must remain constant over time. Therefore, the chemical elements will not increase or decrease without a nuclear reaction. They can be extracted from scrapped goods and used again. Recycling of used materials will increase, and eventually, society might use nearly 100% of recycled materials without the need for mining anymore. The rare materials in the scrapped goods will probably not be recovered 100%, and some mining will be needed to fill the gap. It is also possible to produce rare elements by nuclear reaction if they have been completely depleted and cannot be substituted by other elements. The supply of minerals in the Robotic Age will be met by the following means: 1) resources will be more efficiently used, such that the same amount of materials can generate more outputs; 2) technological progress will make it possible to substitute most of the essential materials with other materials which are more abundant; 3) materials in the scrapped goods are

Chapter 6

328

almost 100% recycled; 4) outer space mining that extracts minerals from the moon, other planets and their satellites, and planetoids when it is technologically and economically feasible (Dallas et al. 2020). Recycling is the most viable solution to the depletion of mineral resources in the future. The future supplies of natural resources are summarized in Table 6-3. Recycling will address not only the supply of minerals but also the pollution issues. Waste discharges have been the leading cause of environmental pollution during the early stages of industrialization. Processing industrial and domestic wastes will protect the environment and provide useful and important resources. In the Robotic Age, recycling technologies will play a key role in the economy because they ensure production sustainability and the continued supply of essential minerals. As society progresses into a more advanced stage, the facilities, machines, and home appliances become more and more sophisticated. They require more and more different materials and contain increasingly more different chemical elements. The mines of these elements can be exhausted, but society has to carry on, which makes it necessary to recycle and reuse materials in the scrapped goods. Table 6-3 Natural resource solutions in the Robotic Age

Energy Water

Exhaustible No

Recycle No

No

Yes

Future supplies Solar, wind, geothermal, tidal, nuclear powers, etc. Preservation, purification, and desalination Recycling, out-space mining Recycling Recycling, substitutes

Rare metals Yes Yes Other metals Yes Yes Non-metal Yes Yes* minerals Synthetic Yes Yes Recycling materials from petroleum Building materials Yes Yes Recycling * Some non-metals are not recyclable, such as sulfur and phosphorus

Many widely available materials, such as stones, sand, and clays, could be in short supply if humans want to preserve natural landscapes. Hopefully, new building technologies that do not require sand and stones in a nonrecyclable manner will arise, and houses could be built with more recyclable materials. As sands and rocks are more abundant than rare minerals, human society could find the right balance between the needs for building and the sustained supply of natural building materials.

Human Resources, Natural Resources, and Pollution

329

3. Pollution Issues in the Robotic Age Industrialization, especially in its early stages, usually brought severe pollution and carbon dioxide accumulation into the atmosphere. In the Robotic Age, abundant material wealth could lead to more production waste and cause environmental pollution if the pollution issue is not managed correctly. More production and more consumption usually mean more waste and more pollution. Still, more strict control on pollution and encouragement for recycling could reduce production wastes and the impact of pollution. Since people’s demand for goods and services plateaus, pollution will be under control in the Robotic Age.

3.1. Sources of Pollution Generally speaking, most substances released by human activities into the environment are likely to cause some adverse effects on ourselves and become a source of pollution (Table 6-4). Production and service processes and domestic waste from households could cause pollution. Natural processes such as wildfires and volcanic activities can also release environmental pollutants. Household waste includes sewage, food, and debris from preparing food and cooking. The dramatic increase in Internet shopping and food delivery has significantly increased household waste in the form of packing materials. Industrial pollution is probably more ubiquitous; open-field mining produces both waste and pollution; extracting metals from ores releases pollutants into the atmosphere and produces solid and sometimes liquid wastes; the chemical industry produces wastewater and chemicals; thermal power stations generate greenhouse gases; nuclear power stations produce radioactive wastes; manufacturing industry has various solid wastes; fertilizers and agricultural chemicals such as pesticides including fungicides and herbicides, hormones and other plant growth agents might also pollute the environment. Pollution can be divided into air, water, and soil pollution. Noise pollution seems to be a particular category that might be classified as air pollution. 3.1.1. Air Pollution Air pollution occurs when harmful or excessive substances are introduced from various sources into the atmosphere. Chemically, these substances include carbon dioxide (CO2), sulfur oxides (SOx), nitrogen oxides (NOx), carbon monoxide (CO), volatile organic compounds (VOCs such as methane, benzene, toluene, xylene, and 1,3-butadiene), ammonia, free radicals, toxic metals such as lead in compounds and mercury, chlorofluorocarbons (CFCs),

Chapter 6

330

biological molecules, and radioactive materials, etc. Their physical features include gas and particulates (dust and aerosol). Their sources can be human activities and natural processes. Table 6-4 Classification of pollution

Air pollution

Pollutants

Sources

Solutions

substances of harm or excessive quantity, such as sulfur oxides, nitrogen oxides, etc. Chemicals, trash, nutrients, heat, etc.

Industries, households, volcanic activity, vegetation, etc.

Limit discharges, and develop lesspolluting technologies and products

Industries, households, soil pollution, etc.

Hydrocarbons, agrochemicals, solvents, heavy metals, etc.

Agriculture, industries, households, etc.

Sound

Machines, vehicles, audio systems, etc.

Purify waste water, recycle used chemicals, limit discharges, and develop lesspolluting technologies and products Reduce trash, recycle trashed materials, and develop lesspolluting farming technology. Control the production of noise and develop low-noise products

Water pollution

Soil pollution

Noise pollution

Human activities causing air pollution are mostly related to the burning of multiple types of fuel by fossil fuel power stations, factories, waste incinerators, furnaces, and other types of fuel-burning heating devices, motor vehicles, marine vessels, aircraft, controlled burn practices in agriculture and forest management, as well as traditional biomass burning in developing and developed countries. Fumes from paint, hair spray,

Human Resources, Natural Resources, and Pollution

331

varnish, aerosol sprays, and other solvents can also be substantial pollutants, especially in enclosed spaces. Natural causes of air pollution include dust from large areas of land with little or no vegetation, methane emitted by the digestion of food by animals, radon gas from radioactive decay, smoke and carbon monoxide from wildfires, VOCs emitted by vegetation (such as black gum, poplar, oak, and willow) (Guenther, Zimmerman, and Wildermuth 1994), and pollutants (sulfur, chlorine and ash particulates) produced by volcanic activity (Devine et al. 1984). 3.1.2. Water Pollution When contaminants are introduced into the natural environment, water pollution contaminates water bodies (rivers, lakes, oceans, aquifers, and underground). It can be grouped into surface water pollution and underground water pollution. Surface water pollution, including ocean pollution (also called marine pollution), is caused by the discharge of chemicals and trash into surface water. Nutrient pollution is due to the release of too many nutrients into surface water, which causes excessive algae growth. Water pollution may come from point sources that have one identifiable cause of the pollution, such as a storm drain, wastewater treatment plant, or stream. It may also come from non-point, diffuse sources, such as agricultural runoff. Various chemicals, pathogens, and physical parameters, such as elevated temperature, can cause water pollution. High water temperature decreases oxygen levels, killing fish, altering food chain composition, reducing species biodiversity, and fostering invasion by new thermophilic species. A common cause of thermal pollution is discharged by power plants and industrial manufacturers of water used as a coolant. 3.1.3. Soil Pollution Soil pollution or soil contamination is the presence of xenobiotic (humanmade) chemicals or other alterations in the natural soil environment. Industrial activity, agricultural chemicals, or improper waste disposal usually cause it. Air pollutants may also fall to the ground and become absorbed by the soil. The common types of soil contaminants include 1) petroleum hydrocarbons; 2) agrochemicals such as pesticides, herbicides, and fertilizers; 3) polynuclear or polycyclic aromatic hydrocarbons such as naphthalene and benzopyrene; 4) solvents such as halogenated non-polar aromatics, aliphatics, and heterocyclics; 5) lead, and other heavy metals.

332

Chapter 6

The level of contamination is highly correlated with the degree of industrialization and the intensity of chemical substance use. Soil contamination can cause health problems via various channels. Direct contact with the contaminated soil and vapors from the contaminants can cause health problems. Crops grown on contaminated land may absorb toxic pollutants, which remain in the food chain and make the food unhealthy. Soil pollution could cause secondary contamination of water supplies within and underlying the soil. Mapping contaminated soil sites is time-consuming and expensive, and even more so are the cleanups. Developed countries have taken steps to map and measure contaminated land and clean up or contain them. The extent of contaminated land is best known in North America and Western Europe. Developing countries often repeat the mistakes of developed countries during their industrialization process. For example, China’s soil contamination has severely worsened during its rapid economic development since the late 1970s. The severity of soil contamination in many developing countries is not well defined. 3.1.4. Noise Pollution Noise pollution is also known as environmental noise or sound pollution. The World Health Organization (WHO) defines noise above 65 decibels (dB) as noise pollution. Noise pollution not only affects people’s concentration and mood but also impairs people’s health. Noise becomes harmful when it exceeds 75 dB and is painful above 120 dB. Noise, as a social issue, causes disharmony and conflicts among neighbors. As a health issue, it is associated with several health conditions, such as cardiovascular disorders, hypertension, high-stress levels, tinnitus, hearing loss, sleep disturbances, etc. Noise sources are usually machines, transport vehicles, and propagation systems. Loud sounds from home audio systems, musical instruments, especially a piano, or even television may affect a neighbor’s quality of life. As noise does not produce physical waste, the main solution is to set up and enforce proper regulations to control noise production, especially those produced by factories and transport vehicles, including airplanes. Domestic noise that causes conflicts in the neighborhood is often more challenging to resolve because different people’s tolerance levels differ. Society needs to promote civilized behavior and concern for other people to resolve noise issues in the neighborhood and make relevant regulations.

Human Resources, Natural Resources, and Pollution

333

3.2. Waste Processing Since developed countries will most likely enter the Robotic Age before developing countries and currently have well-established regulatory mechanisms to reduce pollution and improve the environment, pollution in the Robotic Age will be an issue under adequate control. Waste processing and pollution prevention will be an important part of the economy. Recycling rare chemical elements and materials can be economically viable, and some waste processing facilities would be supported by public funding. Technological progress in manufacturing could significantly reduce raw materials and energy used in production and, consequently, waste. For example, additive manufacturing with 3D printing will produce much less waste than traditional reductive manufacturing with lathe and milling machines. The Robotic Age will be an age of affluence in which people are more concerned with the quality of life than personal wealth. A clean and healthy environment is necessary for a high quality of living. Industrial and domestic wastes must be properly processed before discharge to ensure a clean and healthy environment. Therefore, waste processing and material recycling will be among the most important industrial sectors. Waste processing includes wastewater, chemical liquids, kitchen waste, biological waste, scrap equipment, and disused home appliances. Waste processing and material recycling in the Robotic Age will reach such a level that 1) wastewater be purified to be used for irrigation without decreasing the soil quality; 2) heavy metals, rare metals, and chemicals that pollute the environment will be removed, and extracted for reuse; 3) waste solids used for landfill will not decrease the soil quality; 4) gaseous wastes will be processed to reduce the hazardous pollutants as closely to the naturally occurring concentration as possible. Because people are more concerned with the quality of living, an equilibrium in economic activities between waste generation in production and consumption processes and waste reduction will be reached to maintain a clean and healthy environment. On the one hand, there is production to supply all the goods for the population’s satisfaction. On the other hand, there is the service to clean up all the waste generated by production, service, and consumption in society. It is impossible to clean up all the waste, but the residuals of pollutants after the waste treatment will be reduced to a level that nature can handle. The Robotic Age will also be a recycling age where all industrial and domestic wastes will be recycled to reuse rare substances and remove hazardous pollutants. People living in modern society often

334

Chapter 6

complain about polluted air. Clean air will return when all the waste is treated to reduce pollution. In modern society, waste falls into three categories: 1) side products, cutoffs, and leftover raw materials from production; 2) scrapped goods; and 3) domestic biological waste from humans, animals, and plants. Regulations can enforce the processing and recycling of industrial waste while recycling materials contained in scrapped goods and processing domestic biological waste will require public funding through general or hypothecated taxation. When natural resources such as minerals and petroleum are nearly exhausted during the advanced stage of the Robotic Age, the high prices of ores will make recycling rare chemical elements and other materials from scrapped goods economically viable. Processing biological wastes to generate biogas and organic fertilizers may also become a profitable or breakeven business. With home appliances becoming intelligent and automatic and connected with the IoT, processing domestic biological waste will be an integral part of a smart residential building system.

3.3. Pollution Control In economics, excess pollution has been explained as a consequence of marginal private costs for maximal private revenue being lower than overall marginal costs, the negative externality of production. People ignore the adverse effects of pollution during industrialization because their basic physiological and psychological needs have not been met. When these basic needs are met, they are more willing to pay higher prices for goods produced by clean manufacturing technologies to have clean air and water. In the Robotic Age, everybody will have a guaranteed income to live a respectable life. Therefore, people will support much stricter regulations on industrial wastes and pollutants. They will also be happy with public subsidies for recycling their scrapped goods and processing their domestic biological waste. In the Robotic Age, the measures for pollution control will be similar to those in the Machine Age, except stricter. Regulations for stringent emission and waste discharge criteria, tradable emission and waste quotas, and subsidies for recycling and waste processing facilities can still be used during the Robotic Age. Public subsidies can be financed by general taxation or hypothecated taxation. Research and development for new technologies to prevent and clean up pollution will be supported by public funding. Entrepreneurs developing such technologies for profit should be encouraged and partially funded by general grants from the government.

Human Resources, Natural Resources, and Pollution

335

Environmental protection is an endeavor benefiting every member of society.

4. Attention as a Resource The success of Google and Facebook makes investigating how they generate revenues and profits worthwhile. The most successful Internet companies usually belong to one of two types: 1) firms generating revenues by posting advertisements on the Internet for other firms, such as Google, Facebook, and Twitter, and 2) firms generating revenues by displaying goods on the Internet for other firms, such as Alibaba and eBay. Amazon may also be considered the second type, although it primarily displays goods owned by itself for sale. Their success seems to be built on their ability to attract and hold people’s attention, which suggests that people’s attention is a vital resource.

4.1. Firms Posting Advertisements Two of the most successful Internet firms, Google and Facebook, primarily rely on advertisement income as their primary sources of profits. In this sense, they are not much different from their colleagues in traditional media. The price of traditional medium advertising depends on the range of exposure, time (timing, duration, frequency, and display period), position, and content (length of text and size of graphs). For example, since a national television program has national exposure, its prime-time advertisements will charge more than an advertisement on a local television program broadcasted during its non-prime time. The success of traditional advertisers depends on four factors: 1) programs that attract the attention of the audience or places that people have to pass by; 2) a flexible pricing scheme that suits the different needs of as many clients as possible; 3) an enormous capacity to handle advertisements; and 4) low marginal costs. It is difficult for a traditional advertiser to be good at all four aspects. Google and Facebook appear to have all four factors. They have the most popular search engine and social networking site, respectively, attracting the largest number of website visitors. They price their advertisements according to the number of clicks and viewing time, which enables small enterprises to advertise their products. It is easy and cheap for them to invest in the capacity to handle a large volume of advertisements. The marginal costs of their operation are meager. Of all the four factors, the ability to attract most people’s attention is the first step to success. Therefore,

336

Chapter 6

people’s attention nowadays and in the Robotic Age is one of the most important resources for business success. Google can afford to give its employees high pay and let them decide what to do, which is not because high pay and an employee’s decision rights on their job can make Google profitable. It is because Google’s advertisement service can generate enormous profits without much ongoing effort from its employees. The advertisement profits of Internet service firms like Google and Facebook come from three sources: 1) take the market share as well as profits from traditional advertisers; 2) increase the volume of the advertisement market to share the increased profits of firms that previously advertised little or did not advertise at all; and 3) increase profits margin because of low operational and marginal costs. The first causes the decline of traditional advertisers. The second increases the profits of the firms that previously attracted little attention from consumers and other firms while decreasing the profits of the firms that attracted all or most attention from consumers and other firms. Its overall effect is to erase the profits gained from information advantage or attention advantage. The third leads to the concentration or consolidation of the advertisement market.

4.2. Firms Displaying Goods In the preceding section, we have pointed out that people’s attention is essential for firms displaying online advertisements. It is also important for producers, wholesalers, and retailers to sell their goods. Before a customer or firm decides to purchase goods, the goods must attract the customer or its purchasing manager’s attention. Then the customer or purchasing manager compares a few brands’ costs and functions to make a purchase decision. Shop owners set up their shops in high streets and large shopping malls to attract shoppers’ attention. When it started, eBay was an auction website for people to auction their used goods, but soon some users found that it could be used as an online shopping mall to set up online shops because of the attention it attracted. Alibaba was initially a copycat of eBay to serve business users. It has attracted more attention and users since Taobao was established as an online shopping platform for people and firms to set up online shops to serve consumers. The online display of goods provides more functional virtual shop shelves and advertisement billboards because of its ready access. The more attention an online platform like Taobao and Amazon attracts, the more likely it is that individuals and firms would choose it to set up their online shops, i.e., the network effect. This leads to the concentration or

Human Resources, Natural Resources, and Pollution

337

consolidation of the platform market. The platforms enable producers, wholesalers, and retailers to reach customers they could not get previously and increase their revenues and probably their profits. Firms previously with better access to the market and customers now face more competition, and their profits would decrease. The overall effect of the online platforms displaying goods for other firms is to erase the profits gained from information advantage or attention advantage.

4.3. Live Commerce, Internet Celebrity Commerce, and SelfMedia An interesting business phenomenon in China is the rise of live commerce on the Internet in recent years, which means Internet celebrities sell goods through live streaming. The relationship between live Internet commerce and TV shopping channels is similar to that between Internet advertisers and TV advertisers. The high incomes of Internet celebrities in live commerce attracted public attention in China when in 2021, one of the live commerce Internet celebrities received a tax evasion bill of 1.341billion RMB for 2019–2020, which far exceeded the bill of 0.881 billion RMB received by one of the most famous actresses in China, Fan Bingbing, in 2018 for tax evasion over many years. Internet celebrities usually use their influence on their fans to market goods on behalf of producers or retailers and charge commission for goods sold from their accounts. Their fans’ attention is the resource an Internet celebrity controls, and firms pay a piece rate to use this resource, which aligns the celebrity’s interests with those of the firms. Many less influential live streamers may sell goods they bought from wholesalers. Many Internet celebrities use their public WeChat accounts, microblogs, or blogs to market and sell goods. Self-media or we-media has also become an important means for many people to make a living or earn extra income in addition to their providers’ earnings from a full-time job (Yuan 2019). When advertisements are inserted in their videos and viewed by netizens, YouTube shares advert incomes with account holders. Some self-media people earn millions of dollars by working on their presentations in their spare time. Others make self-media a full-time job and often can get a fairly reasonable income from it. Li Ziqi, a self-media celebrity, reportedly earns several hundred million RMB annually by presenting traditional Chinese cooking and recipes. YouTube and self-media journalists/entertainers are in a symbiotic relationship; YouTube attracts the attention (time) of many netizens, so self-media people set up their bases on YouTube to help them attract more people.

338

Chapter 6

Successful self-media people can also bring more netizens to YouTube and keep them there. Attention (time) appears to become one of the most critical resources nowadays and may become even more critical in the Robotic Age. In the Paleolithic Age, physical strength was probably the most important resource at the disposal of human hands. Both physical and mental strengths were among the most important resources in the Neolithic Age and probably the following Bronze and Iron Ages. In the Machine Age, physical strength was still needed but no longer a main factor; capital and mental strength were the most critical resources. In the Robotic Age, all product markets are saturated. Therefore, consumers’ and firms’ attention (time) becomes the most vital resource. Internet celebrities can use control of their fans’ attention to make a living or even an enormous wealth.

5. Jobs as a Resource in the Robotic Age In the Robotic Age, most jobs that human workers currently perform will be carried out by robots. Then the question of what human workers will do arises. There are many different forms of entertainment for humans to enjoy themselves. However, will all humans be satisfied by pure entertainment? One likely scenario is that many humans would like to do some work they enjoy. Such work can be challenging to themselves and beneficial to others. Since jobs for human workers become scarce resources, humans might compete for the few remaining human jobs.

5.1. The Disappearing Human Jobs At the beginning of the Machine Age (the Industrial Revolution), workers were worried about losing their jobs because machines resulted in high productivity. Some of them were violently opposed to the use of machines. The concerns over machines did not materialize because more jobs were created with economic development than jobs lost due to the use of machines. Three factors at play prevent the loss of employment during the Machine Age. First, the demand for existing goods and services was far from satisfied at the beginning of the Machine Age. The high productivity with machines must first meet the increased demand for existing goods and services rather than cutting jobs. The first factor is relevant probably only during the beginning of the First Industrial Revolution.

Human Resources, Natural Resources, and Pollution

339

Second, new goods and services have been continuously invented in large numbers. The high demand for these goods and services leads to more labor demand, surpassing the lost labor demand due to more efficient machines. Since the 1970s, the expansion of the service sector has created more jobs than those lost due to increased productivity in the manufacturing sector. Third, socio-political reforms increased labor protection, abolished child labor, and decreased workers’ working hours, somewhat decreasing the labor supply. The working hours of workers were reduced substantially around the beginning of the twentieth century following the rise of the Second Industrial Revolution. The decline in working hours since 1870 seems to have ceased in the postwar period, and working hours have remained approximately constant since 1960 (Cooley, Hansen, and Prescott 1995) (in the United States and Canada). In many European countries working hours have decreased slower since 1970 than previously (OECD 2006). Therefore, both the second and the third factors play a role, and in the second half of the twentieth century, the creation of new jobs surpassing the loss of existing positions may be a critical factor. However, at the same time, unemployment seems to become a chronic problem in developed countries. The fourth reason that machines in the Machine Age have not reduced human workers’ jobs is that human workers are needed to operate the machines. The latter three factors act together to increase employment for human workers. Reduction in working hours plays a role, but it may not have a significant effect, as the decline in working hours has slowed and stopped almost entirely in the US since the 1960s. Novel goods and services are the leading cause of increased jobs for human workers. Robots in the Robotic Age will take over all the repetitive factory jobs in the early days and nearly all positions at the mature stage because human workers will not be needed to operate machines. Concerning the job situation in the Robotic Age, the optimistic view considers that history will repeat itself such that more than enough jobs will be created to compensate for the loss of employment due to using robots. The pessimistic view on human workers’ job situation is that robots will replace human workers. From our early analysis, the situation is truly different this time. When the machines (robots) no longer need human workers’ control, the new jobs that new goods and services might create will not come because robots can operate in the Robotic Age. Therefore, on the one hand, progress in research and development leads to many more new products and services; on the other hand, the demand for those new

340

Chapter 6

goods and services will largely be met by robotic workers’ output. Robotic workers can produce goods better and more efficiently than human workers. The Robotic Age will see the end of routine jobs for human beings. What remains for human workers is the most innovative type of work. It can be imagined that robots will become innovative and inventive in the advanced stage of the Robotic Age so that we will have robotic scientists, inventors, innovators, entrepreneurs, etc. Some human beings may still have the brain power to compete with robots and work hard to discover, invent and innovate, but most will probably give up competing with robots and enjoy the benefits of robots. The progress in machine learning indicates the possibility and eventuality of inventive and innovative robots. When robots can produce better goods than human workers, and their costs are lower, the right approach for businesses is to use robots instead of hiring human workers. To compete with robots, human workers must have a wage rate commensurate with their lower performance. However, reducing the wage rate to compete with robots implies that human workers will have a lower wage rate when robots have higher productivity in terms of output per unit currency/spending. If hiring human workers is lower than investing in robots in terms of output per unit currency, the employers would instead employ low-wage human workers. If we leave everything to market forces, humans replacing technologies with decreasing costs will depress wage rates for human workers and increase employers’ profits. The reserve wage rate for human workers will be the minimum social benefit for the unemployed. At the limit, all human workers will be employed at the minimum social benefits. Nobody will work at that wage rate if there is no difference between wages earned in a job and social benefits for the unemployed.

5.2. Work as a Scarce Resource When human-replacing technologies are widely applied, jobs for humans, except freelancing, will become scarce. Many people would enjoy living on the income provided by the social welfare system and pursue personal interests such as reading, writing, painting, entertaining, jogging, hiking, playing sports, etc. However, some may prefer being part of the organized production process or administration. The more successful a person feels about their career, the more likely they enjoy working and want to climb the next career ladder in the same field. People who would retire for the same income generally would not find their job fulfilling or feel they have a successful career.

Human Resources, Natural Resources, and Pollution

341

From a sociological point of view, jobs are critical personal resources as well as organizational resources. As a unique resource, a job gives its holder an income and the corresponding social status as well as public or corporate resources under the control of the job. In China, most government officials are reluctant to retire even though they are paid the same salary level when retired because they will lose social status and resource control. As an organizational resource, firms and public institutions can use jobs (associated status) to attract people to work for them. Jobs have different values as a resource, depending on their associated status and power. People are willing to work unpaid for top jobs. Famous institutions may often pay lower salaries to people with the same or better qualifications than less prominent institutions. When work becomes scarce, people may find work a pleasant experience and vie for an opportunity to work. People pay to exercise in gyms, swim in swimming pools, and climb mountains. It is quite possible that people have to pay to work in the Robotic Age because working gives people utility. If the opportunities to work cannot be bought or auctioned, they would be allocated by lottery or other allocation mechanisms. When work becomes a reverse service, consumption, the opportunity to work becomes a service to be consumed. Work would be critical in a consumer’s optimal bundle of goods and services. The scarcer the work becomes, the more valuable to consumers the work will be. People with average intelligence might be unable to do the remaining jobs that robots cannot perform. There will be a selection and competition procedure for people to acquire the opportunity to work. The more advanced the humanreplacing technologies become the fewer jobs that still can be performed by human workers remain. Entrepreneurs might create jobs for people to “consume,” essentially role-playing or acting for self-satisfaction. In the extreme of humans replacing technologies, it is imaginable that all jobs have been taken over by robots more cheaply and efficiently. Robots will run human society, and humans will become pure consumers who do not participate in the production process except for self-entertaining and mutual entertainment. What can limit human consumption is time.

6. Summary In the Robotic Age, robots and AI systems will mostly replace human workers, but people can still work to pursue their interests and may collaborate with robots. The most important job for humans in the Robotic Age is to play their role as legislators and to accompany each other in the

342

Chapter 6

community. Depletable mineral resources will be recycled to sustain the economy and might be exploited in outer space if economically and technologically feasible. Nuclear energy and renewable energy will support energy needs, and electricity and hydrogen energy will be used by transport vehicles. Pollution will be addressed through legislation and technological progress. People’s attention has become one of the essential resources in the Robotic Age. Producers and retailers will compete for the attention of people. People may compete for jobs which will become much scarcer because working satisfies them.

References Ariens, Everhardus Jacobus. 1954. “Affinity and intrinsic activity in the theory of competitive inhibition. 1. Problems and theory.” Archives Internationales de Pharmacodynamie et de Thérapie 99:32-49. Autor, David H, Frank Levy, and Richard J Murnane. 2003. “The skill content of recent technological change: An empirical exploration.” The Quarterly Journal of Economics 118 (4):1279-1333. Bao, Lian-Jun, Keith A Maruya, Shane A Snyder, and Eddy Y Zeng. 2012. “China’s water pollution by persistent organic pollutants.” Environmental Pollution 163:100-108. Beu, Danielle S, and M Ronald Buckley. 2004. “This is war: How the politically astute achieve crimes of obedience through the use of moral disengagement.” The Leadership Quarterly 15 (4):551-568. Bloom, Irene, and J. Ivanhoe Philip. 2009. Mencius. New York: Columbia University Press. Braun, Tibor, Carl T Bergstrom, Bruno S Frey, Margit Osterloh, Jevin D West, David Pendlebury, and Jennifer Rohn. 2010. “How to improve the use of metrics.” Nature 465 (17):870-872. Cai, Tao, Hongbing Sun, Jing Qiao, Leilei Zhu, Fan Zhang, Jie Zhang, Zijing Tang, Xinlei Wei, Jiangang Yang, and Qianqian Yuan. 2021. “Cell-free chemoenzymatic starch synthesis from carbon dioxide.” Science 373 (6562):1523-1527. Chmielewski, Andrzej G. 2008. “Nuclear fissile fuels worldwide reserves.” Nukleonika 53 (suppl. 2):11--14. Chu, Johan SG, and James A Evans. 2021. “Slowed canonical progress in large fields of science.” Proceedings of the National Academy of Sciences 118 (41):e2021636118. Cleland, WW. 1963. “The kinetics of enzyme-catalyzed reactions with two or more substrates or products: II. Inhibition: Nomenclature and theory.”

Human Resources, Natural Resources, and Pollution

343

Biochimica et Biophysica Acta (BBA)-Specialized Section on Enzymological Subjects 67:173-187. Connor, DJ. 2008. “Organic agriculture cannot feed the world.” Field Crops Research 106 (2):187-190. Cooley, Thomas F, Gary D Hansen, and Edward C Prescott. 1995. “Equilibrium business cycles with idle resources and variable capacity utilization.” Economic Theory 6 (1):35-49. Dale, Spencer. 2021. BP statistical review of world energy. London: BP Plc Dale, Spencer. 2022. BP statistical review of world energy. London: BP Plc. Dallas, JA, S Raval, JP Alvarez Gaitan, S Saydam, and AG Dempster. 2020. “Mining beyond earth for sustainable development: Will humanity benefit from resource extraction in outer space?” Acta Astronautica 167:181-188. De Waal, Frans, and Frans BM Waal. 2007. Chimpanzee politics: Power and sex among apes. Baltimore: John Hopkins University Press. Derbyshire, Emma, and Keith-Thomas Ayoob. 2019. “Mycoprotein: Nutritional and health properties.” Nutrition Today 54 (1):7-15. Devine, JD, H Sigurdsson, AN Davis, and S Self. 1984. “Estimates of sulfur and chlorine yield to the atmosphere from volcanic eruptions and potential climatic effects.” Journal of Geophysical Research: Solid Earth 89 (B7):6309-6325. Einarsen, Ståle, Merethe Schanke Aasland, and Anders Skogstad. 2007. “Destructive leadership behaviour: A definition and conceptual model.” The Leadership Quarterly 18 (3):207-216. Grancea, Luminita, Mark Mihalasky, and Martin Fairclough. 2020. Uranium 2020: Resources, Production and Demand. A Joint Report by the OECD Nuclear Energy Agency and the International Atomic Energy Agency. Pairs: OECD. Guenther, Alex, Patrick Zimmerman, and Mary Wildermuth. 1994. “Natural volatile organic compound emission rate estimates for US woodland landscapes.” Atmospheric Environment 28 (6):1197-1210. Heinl, Robert D. 1998. “From Our May 1967 Issue...: The Inchon Landing: A Case Study in Amphibious Planning.” Naval War College Review 51 (2):117-134. Hong, Joo Wha, Qiyao Peng, and Dmitri Williams. 2021. “Are you ready for artificial Mozart and Skrillex? An experiment testing expectancy violation theory and AI music.” New Media & Society 23 (7):1920-1935. Hoyer, Daniel, and Hendrikus WGM Boddeke. 1993. “Partial agonists, full agonists, antagonists: dilemmas of definition.” Trends in Pharmacological Sciences 14 (7):270-275.

344

Chapter 6

Keesing, Richard G. 1998. “The history of Newton’s apple tree.” Contemporary Physics 39 (5):377-391. Kissinger, Henry. 2018. “How the enlightenment ends.” The Atlantic (6). Lane, Julia. 2010. “Let’s make science metrics more scientific.” Nature 464 (7288):488-489. Lim, Wonhyuk. 2017. “The fourth industrial revolution and its challenges.” Global Asia 12 (2):42-46. Meinrath, A, P Schneider, and G Meinrath. 2003. “Uranium ores and depleted uranium in the environment, with a reference to uranium in the biosphere from the Erzgebirge/Sachsen, Germany.” Journal of Environmental Radioactivity 64 (2-3):175-193. Mervis, Jeffrey. 2016. “‘Employment crisis’ for new Ph. Ds is an illusion.” Science 352 (6288):880. Nabi, Ghulam R. 2003. “Situational characteristics and subjective career success: The mediating role of careerဨenhancing strategies.” International Journal of Manpower 24 (6):653-671. OECD. 2006. OECD Employment Outlook 2006: Boosting Jobs and Incomes: Organisation for Economic Co-operation and Development, Paris, France. Papadopoulos, Yannis. 2001. “How does direct democracy matter? The impact of referendum votes on politics and policyဨmaking.” West European Politics 24 (2):35-58. Park, Michael, Erin Leahey, and Russell J Funk. 2023. “Papers and patents are becoming less disruptive over time.” Nature 613 (7942):138-144. Patton, Stacey. 2012. “The Ph. D. Now Comes with Food Stamps.” Chronicle of Higher Education, May 6, 2012. Petersen, Charisse, and June L Round. 2014. “Defining dysbiosis and its influence on host immunity and disease.” Cellular Microbiology 16 (7):1024-1033. Qiu, Jane. 2009. “China targets top talent from overseas.” Nature 457 (7229):522. Rothenberg, Albert. 1995. “Creative cognitive processes in Kekule’s discovery of the structure of the benzene molecule.” The American Journal of Psychology 108 (3):419-438. Sainato, Michael. 2015. “Stephen Hawking, Elon Musk, and Bill Gates Warn About Artificial Intelligence.” The Observer, accessed August 19, 2020. https://observer.com/2015/08/stephen-hawking-elon-musk-andbill-gates-warn-about-artificial-intelligence/. Silver, David, Aja Huang, Chris J Maddison, Arthur Guez, Laurent Sifre, George Van Den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Veda Panneershelvam, and Marc Lanctot. 2016. “Mastering the game of

Human Resources, Natural Resources, and Pollution

345

Go with deep neural networks and tree search.” Nature 529 (7587):484489. Silver, David, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, and Thore Graepel. 2018. “A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.” Science 362 (6419):1140-1144. Silver, David, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai, and Adrian Bolton. 2017. “Mastering the game of go without human knowledge.” Nature 550 (7676):354-359. Sima, Qian. 1993. “The Biography of the Marquis of Huaiyin (Han Xin).” In Records of the grand historian: Han dynasty, 163-184. New York: Columbia University Press. Susskind, Daniel. 2020. A world without work: Technology, automation and how we should respond. New York: Metropolitan Books. Wagschal, Uwe. 1997. “Direct democracy and public policy-making.” Journal of Public Policy 17 (2):223-245. Weinberg, Alvin M. 1960. “Breeder reactors.” Scientific American 202 (1):82-98. Whyte, Martin King, Wang Feng, and Yong Cai. 2015. “Challenging myths about China’s one-child policy.” The China Journal (74):144-159. Willhoite, Fred H. 1976. “Primates and political authority: A biobehavioral perspective.” American Political Science Review 70 (4):1110-1126. Wrangham, Richard W, and Luke Glowacki. 2012. “Intergroup aggression in chimpanzees and war in nomadic hunter-gatherers.” Human Nature 23 (1):5-29. Younos, Tamim, and Kimberly E Tulou. 2005. “Overview of desalination techniques.” Journal of Contemporary Water Research & Education 132 (1):3-10. Yuan, Yuan. 2019. “Behind the Bright Lights.” Beijing Review (48):32-33.

PART III WEALTH DISTRIBUTION, POLITICS, AND PREPARATIONS FOR THE FUTURE

CHAPTER 7 INCOME AND WEALTH DISTRIBUTION IN THE ROBOTIC AGE

An important economic question is how an economy’s output or income should be distributed in the Robotic Age. Orthodox economics tells us that at equilibrium, income is distributed according to the contribution of production factors: labor receives wages at a rate that is equal to the marginal product of labor; capital receives profits at a rate that is equal to the marginal product of capital; land owners receive rent at a rate that is equal to the marginal product of the land. In the Robotic Age, machines will replace labor at a lower cost, diminishing and eventually eliminating the share received by labor. When labor is increasingly less necessary in the production process, should workers or non-capitalists receive less and less income in the absolute term or relative to GDP? Besides technological progress, international trade and globalization also affect income and wealth distribution within developed and developing countries. They also impact the income and wealth distribution between developed and developing countries. As production moves to developing countries, the real income of blue-collar, working-class households has decreased in the US, while the income and wealth of the superrich households have been increasing fast. The income gap between the poor and the rich returns to the previous high (Piketty 2014). Will this trend of income and wealth concentration continue in the Robotic Age? This chapter will examine how technology, international trade, and globalization influence and change income and wealth distribution in the Robotic Age. As intellectual property rights play an increasingly important role in the modern economy, it will also investigate how intellectual properties affect wealth distribution in the Robotic Age. In the real world, there is no pure economics; all macroeconomic policy decisions are ultimately political. Therefore, the secondary or final income distribution outcome will be determined by redistributing income and wealth through the tax and social security system, which is more of a political decision.

Income and Wealth Distribution in the Robotic Age

349

1. Motivation, Productivity, and Equality As discussed in Chapter 4, people desire higher social status and control more resources than their peers. This desire motivates them to work hard and effectively, be innovative and enterprising, and create wealth. Therefore, a difference in outcomes is the most important driver of productivity growth. However, many people would like equality in outcomes when their outcome is below the average. If goods are abundant, most people would also like to share what they have with those who do not have (Phillips 2004; Van Lange 1999). Generally speaking, if we stipulate equality in outcome payoffs, few people would try to produce the best outcome. The reason is that most people can be altruistic to the less fortunate, but they are reluctant to be taken advantage of by those who are lazy. For economic growth, differences in income must exist and reflect differences in efforts to motivate people to work for more gains. Motivated people make more effort and therefore have higher productivity (Pfeffer 1995). Thus, emulation promotes economic growth. Almost all successful CEOs and senior managers after the 1970s, except in a few fast-growing sectors, became successful because they dared to create income differences among employees. They mercilessly fired employees thought to be underperforming. Celebrity CEOs’ successes are invariably associated with a decrease in the number of employees in their existing business, for example, General Electric (O'Boyle 2011). There is another channel through which differences in income and wealth promote economic growth. Differences in income and wealth are conducive to capital accumulation because more affluent people proportionally consume less than poorer people, i.e., rich people have a higher marginal propensity to save. When the output is low but still above the minimal subsistence level, unequal distribution increases the gross saving rate of society because those people with more wealth have a higher propensity to save. An increase in gross savings increases the capital stock and future output, leading to higher labor productivity. Economic growth will be slower if the output is distributed equally among the population. In the early days of human history, uneven distribution of wealth was achieved by force. Later it was mainly achieved by the exchange at the market or law enforcement; people who control the scarce resources could extract higher shares of the output. During its reform years, China’s policy of “let some people get richer first” (Naughton 1993) caught the spirit of this mechanism of inequality-promoted growth. A large part of state-owned assets that nominally belong to the whole people was transferred to managers of state-owned enterprises (SOEs) and private entrepreneurs

350

Chapter 7

through ownership reform, management buyouts without competing bids, and suspicious transactions involving briberies and kickbacks (Smyth 2000; Garnaut, Song, and Yao 2006). China’s rapid growth has been accompanied by growing income and wealth inequality. Motivation can also derive from doing what a person enjoys doing. Scientific research used to be an activity akin to a hobby for people of wealth or with another job for a living. They spent their money and time conducting research without economic gains because of their curiosity about nature or human society. In solving their curiosity, they also increased human knowledge of nature or society and sometimes developed new technologies and tools that increased the productivity and wealth of society. Although scientific research has become a profession for people to make a living, and some of the research community is more interested in material gains and personal fame, many researchers are still driven more by their curiosity than material gains and fame. In other professions, many people are also more interested in perfecting their skills and causes than in pure material gains. Business people who have made billions of dollars, such as Warren Buffet, Bill Gates, Jeff Bezos, and Elon Musk, might belong in this category. For those people, “labor has become not only a means of life but life's prime want” (Marx 1972). Inequality of outcome with equality of opportunity is necessary for motivating members of society and promoting productivity. However, some people with below-average wealth desire equality of outcome rather than equality of opportunity or permission (Ma 2023). People with wealth above the average often sympathize with such desires. Here, equality of permission means giving people the same rights to engage in certain activities. When people with below-average wealth feel there is neither equality in the outcome nor equality in opportunity, they might try to change the situation by breaking the social conventions or the law, which can lead to a violent revolution. A violent revolution, especially one lasting for years, could destroy capital stock accumulated over generations and cause economic retrogression. Therefore, a social welfare system is necessary to avoid the scenario where a substantial proportion of the population feels maltreated and oppressed by society. A proper system should have 1) a sufficient difference or inequality in the outcome to motivate people to make more effort; 2) equality in the opportunities or permission to let people feel the unequal outcomes are fair and justified; 3) people with the worst outcomes are supported by society to have a decent life. The failure of the communist movement is partly caused by the efforts of communist partyled governments to equalize outcomes rather than opportunities, which led

Income and Wealth Distribution in the Robotic Age

351

to economic stagnation, and partly caused by the communist elites to treat themselves with privileges as “red aristocracy.” Equality of opportunity or permission will naturally lead to unequal outcomes because of differences among people, as Marx analyzed in Critique of the Gotha Programme (Marx 1972). However, everyone is more likely to be better off in absolute terms due to increased productivity when people are motivated by inequality in outcomes. This is a tradeoff between equality in income and income per se when people are motivated by income inequality based on equality of opportunity or permission. When the cake is much larger, everyone can get a larger piece than when it is small. With more output, society has more resources to help the unluckiest. Is it impossible to reduce income inequality? The answer is no. The feasible approach for reducing inequality is through technological progress. The relationship between technological progress and equality is a complicated one. Technological progress reduces the impacts of inequality in people’s innate abilities, which in effect makes them less scarce and capital stock and natural resources relatively scarcer. The reduced scarcity of people’s innate abilities increases the share of capital and natural resource owners in the total output. However, other factors can mitigate or even reverse the inequality-increasing effect of technological progress. One is political; with more and more people’s incomes below the average, as wealth becomes more and more concentrated in fewer and fewer people, people and parties advocating more social welfare to reduce income inequality will dominate the government and increase social welfare. One is cultural; equality in the outcome has been an ideal for many nations since ancient times and has become more prevalent in modern times; as Confucius commented, “Inequality rather than want is the cause of trouble” (Waley 2012). The third is technological; productivity increase generates more resources to improve the social welfare system so that people with more income/wealth than the average will be more willing to give an increased share of the wealth created to help low-income people. The productivity increase caused by technological progress will compensate for the productivity loss caused by the demotivation of equality in outcomes.

2. Technology, Employment, and Wealth Distribution The level of technological development determines a large part of the division of labor and the level of employment in the economy. The level of employment in different sectors is crucial in deciding income distribution in society. Although the Industrial Revolution increased labor productivity

352

Chapter 7

and there was concern that machines would take over jobs from human workers, employment in society actually increased. With the progress in artificial intelligence (AI) and robotics, many people worry about the possibility of massive technological unemployment. Others think technological progress will create more jobs, just like in the previous industrial revolutions (Ford 2015; Brynjolfsson and McAfee 2014). It is essential to have an in-depth understanding of the impact of different technologies on employment. As discussed in the preceding chapters, advances in automation, robotics, information and communications technology (ICT), and transportation will cause the following economic changes. First, the number of human workers employed in the economy decreases. Second, wage growth of human workers stagnates, and wages may slowly fall, first in blue-collar workers and then in white-collar workers. Only a tiny minority of workers who hold critical positions in their firms may see their wages growing fast, but their higher salaries may not offset the effects of job loss on the aggregate wages. Third, the number of middle-class workers (the professionals) declines, with many being out of a job. Fourth, firms serving local or regional markets disappear. Fifth, the wealth is concentrated in a small minority of society. These changes would be the outcome of technological progress if society lets market forces decide the future of human society. They are worth examining in detail.

2.1. General Employment-Enhancing Technology During the First Industrial Revolution, machines did take jobs from some experienced workers because, with machines, less skilled workers could do the jobs faster, cheaper, and usually better. The concerns about job losses led to the Luddites destroying machines (O’Rourke, Rahman, and Taylor 2013). However, history shows that machines have led to more jobs following the Industrial Revolution. We can attribute this phenomenon to the innovation of employment-enhancing technologies. An employmentenhancing technological innovation is an innovation that increases employment in a sector or the economy. Employment increases because the increase in demand for the product caused by a lower price due to raised productivity exceeds the rise in labor productivity driven by innovation. If innovation in one sector only increases employment in its industry, it is sector employment-enhancing technological innovation. Technological progress in one sector may have differential effects on different sectors of the economy. We can examine the impact of one technological stride on the overall economy. If it increases the overall demand for labor in the economy,

Income and Wealth Distribution in the Robotic Age

353

we can call it a general employment-enhancing technology. If it decreases the overall demand for labor in the economy, we call it a general employment-reducing technology. When general economic employment increases, the absolute amount of labor income and the share of labor in the national income will increase. When new technology raises labor productivity and reduces production costs, the lower price of the goods will increase their demand. If the increase in demand for goods exceeds the rise in labor productivity, more workers are needed to meet the increased demand. The Industrial Revolution increased employment because the price elasticity of most goods during the Industrial Revolution was much larger than one. The increase in demand and supply also led to a rise in employment. Therefore, the impact on the total employment of reduction in the labor for producing one unit of output caused by technological progress is more than compensated for by the increase in the total demand caused by the decrease in the price and the appearance of new sectors related to the technological progress. The breakthrough employment-enhancing innovations either increase demand for labor by increasing demand for the existing products more than labor productivity or create a new product that brings in new demand for labor. Existing products can have such an employment-enhancing effect because their output is far from meeting the potential need for the products. For example, steel was produced over two thousand years ago, but the demand was only met with technologies developed in the late nineteenth and early twentieth centuries. The textile industry before the Industrial Revolution was a similar story. In most countries before the Industrial Revolution, fabrics were expensive, coarse in quality and texture, and few in variety. Hence, the demand for clothes in society was far from being met. Therefore, the saturation level of the textile market was very low even after the Industrial Revolution. In Britain, woolen and flax fabrics were materials for making clothes. The Industrial Revolution greatly enhanced labor productivity and reduced fabric costs. Cotton from India and North America provided more choices for consumers at lower prices. The massive increase in demand for textiles increased employment and labor income. When the market for a particular type of consumption goods has been saturated, innovations to increase productivity for producing those goods become employment-reducing technological innovations. At the beginning of the Industrial Revolution, the increased production capacity of the textile industry was to add to the current output rather than to replace the older products; the ability to produce the much larger additional output was a big

354

Chapter 7

increase in society’s wealth. The technological development and expansion in the textile industry's production rapidly transformed the economy's landscape, and the industry nearly began to meet society's clothing needs. From then on, new developments in the textile industry only marginally increased the wealth of human society because better-quality fabrics only replaced poor-quality ones. The total consumption of clothing has a limit for individuals, which depends on 1) how frequently people can change their clothes without finding them burdensome; and 2) how much storage room they have for clothes. By the late twentieth century, the textile industry may add little new wealth to society. Once the market for existing consumption goods has been saturated, the economy’s further growth depends primarily on whether a functionally novel or mechanistically novel product can be invented. The Industrial Revolution brought many new products and new sectors into the economy. Those products were either severely under-supplied or non-existent at all. The emergence of new goods or increased supply of existing goods constituted new wealth in society. The Second Industrial Revolution started many more new industry sectors and mechanized agriculture. The mechanization of agriculture increased the output to some extent, but it hardly increased the total income of farmers and farm workers. The direct effect of the industrialization of agriculture was the loss of agricultural employment. Fortunately, the new industrial sectors could absorb all the workers who could no longer find jobs in agriculture. As the new industrial sectors were fast expanding to meet the market demand for the new products, workers in those industries were paid higher wages than they would get from farming work. The higher profits from a new technology would enable employers to pay higher wages to attract workers and keep the competent ones. The First and Second Industrial Revolutions brought in many new products that satisfied the unmet needs of society or induced new demands. The wealth in society increased dramatically because the capacity to produce goods for unmet demand increased. When the expansion of new products increases the demand for labor faster than the growth of society's overall labor productivity, workers' wages and their share of the national income also increase. Once the demand for a new category of products is roughly met, an increase in labor productivity due to technological advances will no longer increase social wealth. An accumulation of goods will increase social wealth only when they can be used to improve the capacity of society to produce more valuable goods. Overstocking of goods will reduce the value of existing goods. For example, reducing market supply by stocking the

Income and Wealth Distribution in the Robotic Age

355

goods had been the tool for De Beers to control the diamond market (Bergenstock and Maskulka 2001). However, a stockpile far beyond the need to stabilize the supply and to continue to grow without being used as capital stock does not increase social wealth. Cars and airplanes are examples of mechanistically novel products, and airplanes are also functionally novel to a large extent. Telephones, television sets, computers, washing machines, and dishwashers are also functionally novel products, the production of which increases employment in their sectors at least. These functionally novel products usually increase overall economic employment to some extent. Therefore, labor income will increase, but its share in the economy might not increase. Since most human consumption needs will be categorically or functionally met in the Robotic Age, there are few opportunities for inventing functionally or mechanistically novel products. New technologies tend to be general employment-reducing innovations, which at most increase employment only in the sector that produces the new goods.

2.2. Employment-Reducing Technology An employment-reducing technological innovation reduces employment in a sector or the economy because the increase in demand for the product caused by a lower price due to raised productivity is lower than the increase in labor productivity driven by the innovation. Usually, such innovations happen during a stage when the goods produced by the new technology satisfy the consumption needs of society. For example, the mechanization of agriculture markedly reduced farming workers. The machines developed for agriculture are consequences of advances in machine-making, transport vehicles, and innovations in adapting these advances for agriculture. The current agricultural output was close to that for meeting society’s needs, so a drop in the price would not increase the demand sufficiently to compensate for the decrease in labor demand caused by the increased productivity. When general economic employment decreases, the absolute amount of labor income or the share of the national income will decrease. By the 1970s, the demand for physical goods was almost saturated in developed countries. New products tend to be improved versions of existing products, enhance our non-physical consumptions, or replace our mental functions. In the late 1970s, personal computers (PCs) became office gadgets, making the job of secretaries, accountants, cashiers, and many others much easier. There had been a productivity puzzle (Griliches 1988; Cole and Zuckerman 1984) in the 1980s, which became another puzzle

356

Chapter 7

regarding the broad application of information technology (IT) later (Gordon 2000). The puzzle may have a simple explanation, i.e., most of the applications are in the processes that do not directly impact the final output or are not mature enough to increase labor productivity markedly at the technological level of the 1980s-1990s. By the late 1970s and early 1980s, traditional manufacturing sectors arising from the Second Industrial Revolution had been reasonably well automated. The emerging PCs, Internet, and mobile communication technologies before 2000 really could not improve much over the productivity level of traditional manufacturing. More advanced robotic and automation technologies are needed to increase their productivity further. The ICT at that time was not sophisticated enough to raise the productivity of the service industry because the technology was insufficient to replace human operations in service. The transient or short-term increase around 2000 in productivity growth is probably mainly an increase in the ICT industry itself. Starting from the early 1980s, the industry matured by the late 1990s with improved manufacturing/service technology. The “millennium bug” scare triggered a wave of intensive investment in IT equipment and software by Western firms and governments to avoid a scenario in which non-2K compliant computers crashed and caused havoc globally (Best 2003). The intensive investment in IT around 2000 released the potential for productivity growth in the ICT industry. The raised productivity growth for the whole economy reflected the change in the ICT industry, the utilization rate of production capacity, and the proportional weight of the ICT industry in the economy. Once the ICT industry reached a highly automated production and service stage and more stable investment levels in ICT, productivity growth fell back to the long run trend from the 1970s. The productivity elasticity of employment in a sector depends on the maximum consumption capacity of its products in the economy. For many new industries and other undersupplied sectors, productivity growth reduces prices and increases demand more than productivity growth; hence employment increases and the absolute amount of labor income also increases. For industries with demand largely met, productivity growth reduces employment, and the total real labor income decreases. The maximum consumption capacity is determined by individual temporal, spatial, physiological, psychological, and sociological constraints. In addition, consumers may be constrained by law to have a small space, for example, a prohibitive land use tax that prevents rich people from using too much land.

Income and Wealth Distribution in the Robotic Age

357

Another type of employment-reducing technology consists of technologies that reduce human input to production processes, such as automation, robotics, and AI. With the progress in these technologies, more and more human workers will be replaced by robots and AI systems. When the capacity to consume is gradually saturated, new technologies will reduce more and more employment. This technological unemployment cannot be solved; human society has to live with technological unemployment. The extreme scenario will be that almost no human worker is needed in the economy. Then if there is no redistribution of income, capital and land owners will receive all the national incomes. When the capacity to consume is saturated, besides employment-reducing technologies, there will still be quality improvement technological innovations, which will not increase overall consumption in quantities. As agriculture is the first industry in human society, its output has nearly met society’s needs from its very beginning. By the Industrial Revolution, new farming methods and crops from the new world had brought more outputs and varieties of agricultural products. These changes improved farmers’ earning ability but only marginally, as the capacity of society to consume additional foods is relatively limited. Better-quality foods would replace low-quality foods, and people would consume more only when their purchasing has increased. This means that food prices decrease or people earn more money from other sectors of the economy. The physical volume of people’s stomachs and the ability of the digestive system put a limit on society’s consumption capacity of agricultural products such as food. In recent times, the agriculture and food industry in developed countries has grown to such a capacity that overeating and over-intake of energy have become the primary concern. The food industry can still provide new products that supposedly are healthier or give customers new experiences. Agriculture as a source of food has little potential for generating more wealth. It might be argued that humans are satisfied by consumption and ownership of goods or wealth. Individuals may be happier owning vast sums of money, big houses, expensive jewelry, and designer clothes. People are motivated to hold goods beyond their consumption needs in a society where goods are undersupplied. This ownership desire might be constrained by space entitlement. How effectively can an individual use or control an ample space to satisfy their consumption and sense of ownership? Which will they enjoy more, a house of 200 square meters or 20,000 square meters? For many people, living in a huge house is no more enjoyable than living in a reasonably sized one. An individual usually has an effective personal space,

358

Chapter 7

and a larger area beyond this effective space may not give them a sense of ownership. The perception of adequate personal space will constrain how much space an individual wants.

2.3. Return of Capital with Decreasing Labor Demand As we have analyzed earlier, the time entitlement for individuals is finite, so will be their non-vanity consumption because of the time constraint and physiological constraints. At a particular stage, production for consumption will plateau in quantities; new products are simply better substitutes for existing goods and services. It is likely that, well before reaching that stage, the increase in labor productivity due to automation and robots has led to fewer and fewer human workers in employment. Since at equilibrium, the marginal products of capital should be equal to the marginal product of labor, and the wage should be equal to the marginal product of labor, as capital stocks use larger and larger shares of the funds for production, workers will get less and less in wage income. Fewer and fewer human workers imply that a larger and larger share of the national income goes to the owners of capital. Wealth distribution depends on the shares of national incomes between labor and capital. Traditional businesses need more workers because of very low-level automation. The Cobb-Douglas production function came from their initial finding that labor got two-thirds of the national income and capital got one-third (Douglas 1976). With automation and robots, the proportion of labor in the overall production factors has been substantially reduced for the traditional manufacturing industry. This trend of replacing human workers with robots will continue. In an extreme situation, an owner runs a factory with all workers being robots, so the only labor input is the salary paid to the owner. All the rest of their factory income is capital earnings. The rise of Internet firms represents another trend in reduced employee numbers in firms with high market valuations. The number of workers employed by those successful Internet firms amounts to a small fraction of that by traditional manufacturing firms of comparable valuations. Their capital stocks tend to be much smaller than their machine-age counterparts of similar market capitalization. These Internet firms can be classified into three types: advertisement services, platform services, and Internet shops. Internet firms may take over market shares from traditional brick-andmortar shops and advertisement agencies with fewer employees to achieve the same revenue levels. For example, Google, Facebook, and YouTube

Income and Wealth Distribution in the Robotic Age

359

have replaced traditional advertising businesses by providing more efficient advertisements and taking over market shares from conventional media, especially newspapers and magazines. With the decline of traditional firms, Internet firms rise with a net loss of economic employment. Internet shops also operate with fewer employees because of their automated transactions. The competition from Internet shops will also stimulate the automation of traditional shops to lower their operational costs by reducing the number of employees. The reason for the increasing share of capital owners in social wealth and income is simple. The increase in labor productivity is usually and perhaps always associated with the introduction of new technology and new equipment in the production process. Introducing new technology or equipment increases the weight of capital relative to labor in terms of funds used. The optimization of capital to labor ratio (hence maximization of profits) requires that the marginal product of capital (= return of capital) equals that of labor (= wage rate). As they have the same return, the share of capital in income distribution will increase with the labor productivity when the demand for the product is almost satisfied. The improvement in communication technologies and transportation intensifies competition to the extent that many producers cannot resist the overwhelming force of the leading firms in the sector. The “winner takes all” so that near monopoly or oligopoly becomes the outcome of low-cost communication and transportation. With the concentration of industry to only a few large firms, automation and application of robots become more feasible, which leads to fewer human workers employed than with many smaller firms. The fewer human workers who are needed, the larger the share that goes to the owners of capital. The number of capital owners who can live on the return to their capital will also decrease.

2.4. Productivity Growth and Social Polarization There is an inherent contradiction between aggregate demand in the economy and cheaper human-replacing technology. Human-replacing technologies decrease the wage rates of human workers, reducing the aggregate demand from poorer classes of society. If the population stabilizes, the aggregate demand from the poorer classes will stabilize or decrease with the progress of human-replacing technologies. Then the wealthy classes must consume or invest all the additional outputs from economic growth for future growth. Since there are limits to how much the wealthy classes can consume, consumption will not be able to clear the

360

Chapter 7

market unless they resort to meaningless consumption. Investment in new capital stocks will increase output growth, a higher social saving rate, and dynamic inefficiency. In an extreme scenario, society accumulates so much capital stock that most goods are produced to replenish depreciated capital stocks. The third possibility is creating treasures that can last, such as human-made diamonds and other gemstones. However, if there are too many treasures, their value will decrease. The more the wealthy classes accumulate human-made treasures, the less valuable the treasures would become. Therefore, enriching a small minority of the population at the cost of the majority will eventually devalue the riches held by the small minority. To avoid the abovementioned phenomenon, most of the population should have a corresponding increase in their income level and standard of living. As the human-replacing technologies progress and the limits of consumption loom larger and larger, the social benefits for the unemployed should increase to ensure a steady increase in the aggregate demand from the poorer classes. Reducing working hours and increasing workers’ income would increase their consumption demand. However, globalization and exports from developing countries dampen the possibility of reducing working hours and raising wage rates in developed countries. Manufacturing has been shifted to developing countries. In developed countries, asset price increases exist because the wealthy classes can benefit from globalized manufacturing and invest their earnings in domestic assets. What remains in developed countries are those sectors that produce non-tradable goods and services or create and design new products to be manufactured abroad. Because new products are produced abroad, the foreign workers making those goods are paid much lower wage rates than those in developed countries; in contrast, those elite employees designing and developing the new products are paid much more than if those products were manufactured in developed countries. Productivity growth increases social polarization if there is no political intervention.

3. Globalization and Wealth Distribution Most modern mainstream economists are supporters of free trade and opponents of protectionism. Globalization is the natural consequence of international trade following revolutionary progress in transport and communications in the twentieth century. Although economists continue to preach free trade, protectionist measures at the national level have remained the same and may rise in the twenty-first century (Witt 2019; Tomlinson 2012). International trade and capital flow not only increase overall outputs

Income and Wealth Distribution in the Robotic Age

361

in the world but also change wealth and income distribution in countries participating in globalization from that of a closed economy.

3.1. The Law of Comparative Advantage and Its Limitations Mainstream economists think that what governs international trade is the law of comparative advantage, which was proposed first by British economist David Ricardo (Ruffin 2002). If country A has a lower cost for producing a good than country B, then country A has an absolute advantage over country B in producing that good. Suppose the ratio between the costs of making goods 1 and 2 in country A is lower than in country B. In that case, country A has a comparative advantage in making good 1, while country B has a comparative advantage in good 2. The law of comparative advantage states that countries should specialize in producing goods with comparative advantages, then both countries will benefit from international trade. In principle, even if a country has absolute advantages in making every good, it is still beneficial to specialize in producing goods with comparative advantages. Eli Heckscher and Bertil Ohlin built on Ricardo’s theory and developed the Heckscher-Ohlin model (H-O model) as a general equilibrium mathematical model of international trade. It predicts that countries export products that use their abundant and cheap factors of production and import products that use the countries' scarce factors (Ohlin 1935). It is easy to prove that the law of comparative advantage is correct for two individuals. Economics textbooks usually give examples such as Alice makes good one at four hours apiece and good two at one hour apiece, while Betty makes good one at five hours apiece and good two at one and a half hours apiece. Although Alice is better than Betty at making both goods, it will benefit Alice and Betty if Alice specializes in making good two and Betty specializes in making good one. In the original example, David Ricardo used two countries, the UK and Portugal, and two goods, wine and cloth. Still, his oversimplified model was not much different from a twopeople model. There are severe problems in extending the inter-individual analysis to countries. First, the difference in potential skill sets between countries of a similar size is much smaller than between individuals. Here potential skill sets mean those that can be acquired after adequate learning and training. When new products invented in one country bring high profits in international trade, other countries will try to learn the latest production technology rather than continue to specialize in producing low-value-added traditional products.

362

Chapter 7

The law of comparative advantage works only in a static world, especially in terms of endowments of production factors in countries involved in international trade, but not in a dynamic world where people are keen to learn new technologies that make more profits. The law of comparative advantage is more pertinent to accounting for the division of labor in an economy requiring diverse and sophisticated intellect and dexterity. Second, comparing countries with and without specializations is usually based on overall outputs or revenues rather than utility analysis. In economics, decision-making is generally based on total expected utility. Although international trade could increase a country's GDP or gross national product (GNP), there is a high probability that the aggregate societal utility decreases because polarized income distribution is further exacerbated by international trade. Capital owners or their proxies decide where to buy goods from and whether to invest in foreign countries, while workers can only take what has been on offer or not. The savings from consuming cheaper imports made in country A by factories owned by firms of country B may not fully compensate for the loss or decrease in incomes received by workers due to the decline of their industries in country B. Since there is no consensus on the correct form of the societal utility function, it is not easy to estimate the impact of international trade on the total societal utility. Economists generally regard the utility as ordinal, and attempts to sum up individual utilities are usually dismissed because it requires the utility to be cardinal. However, it is possible to count the number and percentage of individuals whose utility has been adversely affected by international trade. An increase in aggregate utility globally may accompany a decrease nationally for developed countries due to income polarization. Third, the law of comparative advantage requires some unrealistic conditions. It assumes international immobility of capital and labor, balanced trade, full utilization of capital and labor, and the non-existence of technological change and productivity growth resulting from international trade. The only gain from international trade is a static increase in outputs due to specialization in particular goods or sectors, a reminiscence of the division of labor for individuals in production. When there are movements of capital and labor, imbalanced trade, and underutilized capital and labor, the outcome predicted by the law of comparative advantage will be much less certain. When a country with an absolute advantage in many sectors is determined to hold an increasing foreign exchange reserve, the comparative advantage theory may fail because the rising foreign exchange reserve implies that some goods are not exported for buying goods abroad. Paul Samuelson has shown that technological progress by one country, such as

Income and Wealth Distribution in the Robotic Age

363

China in its previous comparatively disadvantaged sector, will cause a net loss for its trading partner, compared with their results before China’s technological progress (Samuelson 2004). Therefore, international trade does not always benefit both trading partners as asserted by the law of comparative advantage; instead, it is subject to many conditions to have the outcome asserted by mainstream economics. Fourth, since a country cannot act like one person for which the law of comparative advantage is correct, if the price ratio between two goods reflects the relative productivity in producing the goods, individual international traders will not import goods with comparative advantage and higher prices than the same domestic goods. As illustrated by Fig.7-1, when goods one and two in country A are cheaper than in country B, an international trader can export good one, which has a comparative advantage for country A, to make a profit. Although country B has a comparative advantage for good two, importers in country A will lose by importing good two from country B. When individuals or firms in one country pursue their interests rather than act strategically as one entity, what is correct for the relationship between two individuals might not be correct for the relationship between two countries.

Fig.7-1 Good one is priced at $6 and $12 in countries A and B, respectively; Good two is priced at $9 and $15 in countries A and B, respectively. A trader in country A makes a profit of $120 by exporting 20 units of good one to country B but will make a net loss of $96 by using the revenue to import good two from county B. For simplicity, transaction and transport costs are not considered here, and domestic prices are converted into dollars.

364

Chapter 7

Fifth, the law of comparative advantage has not considered the finiteness of consumption, especially the finiteness of consumption of products in one sector. As shown in Chapter 4, the individual (and consequently aggregate) maximum capacity of non-vanity consumption is finite due to time, space, and physiological constraints. With finite aggregate consumption capacity in both countries, especially when the demand for goods produced by traditional sectors has been satisfied, the country specializing in traditional products will be locked into a perpetual poor state. For example, when trading with industrialized countries, an African country that currently produces wheat and corn and has no comparative advantage in making cars and heavy machines should specialize in producing wheat and corn. However, since food production in developed countries has been satisfied a long time ago, any further increase in farming productivity and output in this African country will only reduce its real income because over-supply will reduce market prices of its products without increasing demand. Moreover, the farmers in industrialized countries are usually much more productive because they have cheaper capital and more advanced industrial technologies to help increase farming productivity. Therefore, if a country wants to become more affluent, it should not specialize in sectors that have roughly satisfied its needs; it must venture into new industries that still enjoy a highly unmet demand in other countries.

3.2. Rising Protectionism and Its Causes With the election of Donald Trump as president of the United States in 2016, protectionism has been rising worldwide, and the trade war between China and the United States erupted. Although the law of comparative advantage supports trade between the two countries, President Trump thought the United States had suffered from trading with China and many American allies. Economists widely criticized his trade policy. According to mainstream economists, there is no need to retaliate for protectionist measures by other countries. Paul Samuelson compared tariffs raised by other countries to digging holes in their roads, and a country should not dig holes on its road simply because other countries did so (Samuelson 1976). Tariffs in the real world have been attributed mainly to small vocal groups of citizens who suffered more seriously personally from international trade exerting more powerful political influence than most of the population who enjoy a small benefit per capita. However, the aggregate benefits of the entire population are much higher than the aggregate costs suffered by those small groups. Interestingly, although economists almost unanimously

Income and Wealth Distribution in the Robotic Age

365

support free trade, no sovereign country lets foreign goods freely enter its market. Although economists such as Paul Krugman acknowledge that dynamic scale economies justify protectionism because temporary protection of industries enables them to gain experience (the infant industry argument) (Krugman 1987, 1983), they hold that the law of comparative advantage is generally true. As mentioned earlier, Paul Samuelson has given a scenario that if China increases its productivity in its previous comparatively disadvantaged sector, the United States will have a net loss in international trade compared with the outcome before the productivity increase (Samuelson 2004). Mainstream economists fail to appreciate the implication of finite maximum consumption capacity, especially in consuming products from one sector. For sectors whose products have met consumer demand, international trade will cause substantial unemployment and no meaningful increase in employment in other sectors for the importing country. While developing countries learn new technologies to take market shares of traditional manufactured goods in developed countries, developed countries usually cannot innovate fast enough to create sufficient jobs for those unemployed due to international trade. By the 1970s, industrial sectors that emerged from the Second Industrial Revolution had almost reached their peak in the developed countries. The trade union movements also reached the peak of their power. Most workingclass families could afford decent food and clothes, accommodation, television sets, washing machines, dishwashers, vacuum cleaners, and family cars. Their houses or apartments had telephones, electricity, gas, tap water, central heating (in cold regions), and flush toilets. When consumer products have reached such a saturation level, it is difficult to imagine how many more physical products people can still consume in the developed world. The goods produced in developed countries can be sold to people in developing countries, but those people can rarely afford to buy goods produced by developed countries because of their low incomes. From very early on, developing countries with rich natural resources, such as minerals, petroleum, and crops that do not grow well in developed countries, become suppliers to developed countries. Exports of those natural resources are insufficient to raise living standards in those countries to the level of developed countries except in oil-exporting countries with small populations. Therefore, developed countries cannot increase their national incomes substantially by exporting their physical products to developing countries for two reasons: 1) the high labor costs make their products less competitive than those produced by less developed countries, and 2) the developing

366

Chapter 7

countries could not afford to buy large amounts of goods produced by the developed countries. The converse is often true because moving production to and importing products from developing countries is more profitable than producing goods at home due to much lower labor costs in developing countries. When physical products are almost saturated for consumption in developed countries, two pressures make blue-collar workers unable to have their wages increased: 1) technological progress continuously increases labor productivity, while the physical goods have roughly saturated the market, which means that fewer and fewer blue-collar workers are needed for the manufacturing industry; 2) workers in developing countries can more cheaply produce tradable physical goods because the communication and transport costs have been dramatically reduced. With these two factors, the price level of consumer goods in developed countries is kept low because the incomes of blue-collar families are suppressed by surplus labor supply and cheap imports from developing countries. As multinationals move their production from developed countries to developing countries by foreign direct investments (FDIs), developing countries can produce and export goods previously made in developed countries. The saturation of consumer physical goods implies that the more affluent classes would not buy more of those goods with their fast-increasing incomes and wealth. Because of international trade, the developing countries active in globalization are better off. FDIs bring in the necessary capital and technical know-how for industrialization. Unlike the blue-collar workers in developed countries, those in developing countries improved their living standards during globalization. However, the lion’s share of the new wealth created by the economic growth went to the elites. The overall effects of globalization benefit both developed and developing countries, but most people in developed countries probably lose out while the elites reap all the benefits. In developing countries, the working and elite classes benefit from globalization, with the elites getting much more than the working classes. From the perspective of developed countries, international trade and globalization have three aspects that could be beneficial: 1) developing countries are markets for products and services of developed countries, especially those high-value-added products; 2) developing countries are places where multinationals from developed countries can invest for higher returns than those in domestic economies, for example setting up factories to produce goods for exporting back to their domestic markets; and 3)

Income and Wealth Distribution in the Robotic Age

367

developing countries are suppliers of cheap products for developed countries. At the firm level, as long as the positive labor cost difference is larger in the long run trend than the negative cost difference in logistics, hiring experts, and political risks, the multinationals will move production to the low labor cost regions. Generally speaking, high-value-added products tend to be less labor-intensive, more expertise-intensive, and intellectual-proprietary. The high productivity for producing high-valueadded goods implies that only a few human workers will be employed to produce them. Thus, national incomes generated from the production will mainly go to the capital owners, senior executives, and a few workers with technical expertise. The growth of high-value-added products contributes to the increasing share of capital in the national incomes and of the elite highsalary employees in the total salaries. For those products whose labor cost difference is much larger than the negative cost difference in logistics etc., moving production to low labor cost regions can increase the firm’s profitability but not the incomes of working-class households in the home country. For those firms, the labor incomes from their output will stay in low-wage countries, while the home countries receive profits if they are not re-invested in the low-wage countries. When production has reached the level where demand for material goods has been largely met, countries with more advanced production technologies may produce high-value-added goods sufficient to exchange for more traditionally manufactured goods. In this way, blue-collar workers in developed countries would be worse off because the high-value goods need only a small number of human workers, so that most blue-collar workers would be unemployed. Without changes in the social security system, more and more of the population will be worse off, and fewer people will get richer and richer. Therefore, the comparative advantage cannot ensure that blue-collar workers have a constant share of national incomes or even an increase in real income. The political pressure from blue-collar workers, trade unions, and firm owners of the affected traditional manufacturers will empower protectionism in developed countries. The production and exports by domestic and foreign-owned firms in lowwage developing countries will increase the income of blue-collar workers and capital owners. Generally speaking, in a country with a large reserve labor pool, the return of capital in the economy will be larger than in countries with more capital. The increase in blue-collar workers’ incomes caused by international trade in developing countries is partly due to the loss of jobs by blue-collar workers in developed countries on the surface. The deeper cause is the comparative advantage of developed countries in

368

Chapter 7

producing high-value goods. When the labor productivity of high-valueadded goods is much higher than that of low-value-added goods, developed countries will obtain the low-value-added goods via international trade. The developing countries (at least some of them) will try to catch up with the developed countries, which means that the developing countries would also like to produce high-value-added goods. The increase in the production of high-value-added goods will reduce the price premium charged by the developed country’s manufacturers. Eventually, the developed countries may have to compete with previously developing countries in producing high-value-added goods. Developing countries would also enforce protectionist measures to protect their infant industries against competition from developed countries. The success of export-led growth by developing countries such as China, especially when tied to strong control measures on imports and business into their countries, has taken large shares of the traditional manufacturing sectors in developed countries (Amiti and Freund 2010; Buckley 2009). Moreover, China uses the same approach to catch up with developed countries in high-tech sectors, undercutting firms in developed countries with cheap costs, government subsidies, and tax rebates. The success of China in catching up and taking over has alerted developed countries, especially the Trump Administration in the United States. Trump, many American opinion leaders, and those not benefiting from globalization wanted to stop China from taking more ground in economic activities, which underlies the rising protectionism in the United States. Globalization promotes economic development in developing countries, which causes an increased demand for raw materials such as iron ore, copper, crude oil and grains, etc. Demand from emerging economies raises the prices of commodities, which will be transmitted to the price level of developed economies. With the economic growth of developing and newly industrialized countries, the high saving rates may lead to more national savings than they could efficiently invest in their economies (Ma 2020). The extra savings often result from inequality in those countries because those countries may have more pro-capital political and legal systems, weak labor protection legislations, and weak trade union activities. The entrepreneurs and capitalists in emerging economies get a larger share of national incomes and invest in both the domestic economy and foreign countries. Capital flows into developed countries for international investment by emerging economies, which further increases asset prices, exacerbates the positive feedback of asset market boom, and increases inequality in terms of wealth

Income and Wealth Distribution in the Robotic Age

369

distribution because low-income households have fewer financial and real estate assets. Since the early 1980s, the dominant thinking in developed countries, especially the US and the UK, has been free market, deregulation, and globalization. However, blue-collar workers have seen their real wage income decrease from the 1970s peak (Krugman and Lawrence 1993). Since the 1990s, even middle-class households have decreased real incomes while the rich get richer. National wealth becomes more and more concentrated in the hand of the superrich. The prevalent view of low-income people is to blame globalization and trade imbalance for losing their jobs. As pointed out earlier, job loss in developed countries has more to do with technological progress in the long run, so protectionists cannot solve the unemployment caused by technological progress. Implementing the protectionist policy will hinder the optimal allocation of resources and, consequently, global economic growth, but with more people losing their jobs, protectionism will gain more strength in the future. There are two possible scenarios of globalization impacting a developed country: 1) the return on capital and the incomes of the elites increase, and the incomes of the (blue-collar) working class stagnate or decrease, while the GNP increases more than that under a protectionist regime; and 2) the return on capital and the incomes of the elites increase, and the income of the (blue-collar) working class stagnates or decreases, while the GNP increases less than that under a protectionist regime. The first scenario can be solved by sharing increased national incomes with disadvantaged households through national security systems. Protectionist measures can address the second scenario.

3.3. Globalization in the Robotic Age In the Robotic Age, when robots replace human workers, if robots are cheaper than humans in developing countries, traditional labor-intensive manufacturing will become less labor-intensive. It can return to where its intended consumers are. This will reduce the need for globalization of manufacturing and service outsourcing, a vital cause of the decline of developed countries' traditional manufacturing and service sectors. However, FDI intended to bypass trade barriers and get closer to consumers will still exist. When the sum of external economies of scale and the savings from robots exceed the transport (and trade barrier) costs from the production venue to the local market, exports will replace local production by FDI. Here external economies of scale denote a firm’s additional output due to

370

Chapter 7

the conditions outside the firm, such as industrial clustering. In contrast, economies of scale imply that the more units a firm produces, the less it costs to produce each unit. The progress in robotics will attenuate one of the causes for moving production to or buying products from low-income countries, cheap labor. The impact of globalization on income distribution will be reduced or eliminated. If there is no trade barrier to protect domestic markets in lowincome countries with less advanced production technologies, developed countries with more advanced technologies and low overall costs may export more goods to low-income developing countries, reversing the trend of international trade between developed and developing countries. However, eliminating the impact of globalization on jobs and income distribution cannot remove the influence of technological progress, which has more to do with the stagnation or reduction of real wages of blue-collar workers and the loss of manufacturing jobs. International trade for primary resources such as minerals will continue in the Robotic Age, as well as materials produced with large economies of scale. Food will also be an essential component of international trade if it cannot be made in factories. When mines are nearly exhausted, mineral ores could become so expensive that recycling to extract rare elements contained in various wastes becomes economically viable. Then international trade in minerals could become smaller than during the Machine Age. Manufactured goods can continue to be a significant part of international trade if economies of scale and external economies of scale exceed the transport costs, but the direction of flow could well be from the developed to the developing countries.

4. Intellectual Property and Wealth Distribution An important feature of recent business success stories is their reliance on intellectual or intangible properties rather than physical ones. Traditional manufacturing firms often take decades to get into the largest firm list regarding market capitalization, profits, or revenue. Internet firms can grow into multibillion-dollar businesses in a few years. If we scrutinize these firms, they may be divided into three types: Internet platforms, search engines, and online shops. Alibaba, eBay, and social network sites like Facebook are platforms; Google and YouTube are search engines; and Amazon is a shop and web service provider and a platform for other retail firms. Intellectual properties or intangible assets are essential in these successful Internet businesses. Still, their dependence on intellectual

Income and Wealth Distribution in the Robotic Age

371

property differs from that of Microsoft with Disk Operating System (DOS) and Windows operating system. Software and hardware suppliers who depend on intellectual property for monopolistic profits will lose their edge if there is no protection of intellectual property rights. The established brand name and network effects might be more important than other intangible assets for platforms like Alibaba, eBay, and Facebook. The main income form for Facebook and Google is advertisement fees, and for Alibaba and eBay charging for services (for each item sold) is their main earning format.

4.1. Intellectual and Intangible Properties as Revenue-Earning Assets The high earnings from intangible assets create more income and wealth inequality. There are many criticisms of Alibaba and the like on Internet forums and WeChat in China. Some of the complaints came from owners of the firms producing tangible goods and running brick-and-mortar shops. What impact do Alibaba-like retail platforms have on tangible goods manufacturers and retailers? Manufacturers’ economic profits may come from the following sources: 1) better-than-average production technology; 2) better-than-average management of labor and logistics; 3) brand reputation; 4) lower transport costs than competitors; and 5) consumers’ information costs (searching costs). E-commerce by Alibaba and the like reduces customers’ information costs and makes retailers’ prices transparent, which removes the economic profits due to information costs to the customers. The competition between online and brick-and-mortar shops is between their expenses and convenience. The costs of brick-and-mortar shops include the commodity’s price + journey-and-searching costs, and those of online shops include the commodity’s price + delivery charges. Internet firms, except those providing platform services, are not very profitable. E-commerce firms that sell goods to consumers generally have a low-profit margin. The successful ones are taking market shares from the traditional commercial firms and depressing the price because of price transparency and the low prices offered due to lower running costs than the brick-and-mortar shops. With the increasing popularity of online shopping, traditional retail shops are in decline. Amazon is probably the most successful online retailer. It sells various goods and provides Internet ecommerce platforms for other retailers and cloud computing services. Ecommerce giants Amazon, Alibaba, and Jingdong have built their own logistics and delivery systems in China, while in other countries, Amazon

372

Chapter 7

mainly relies on shipping and parcel-delivering firms. Amazon has performed poorly in China and ceased most of its business activities there. Physical shops generally have higher maintenance costs than online shops. Online shops usually only need storage space, whereas physical shops need to be located in more central areas with decorations and shop assistants in addition to storage rooms. Although window shopping can be a pleasant activity for some consumers, travel expenses and time are extra costs in addition to the price of goods bought from physical shops. Delivery charges are often added when goods purchased online are of small value. Consumers choose cheaper online goods when delivery charges are comparable to travel costs. Online shops provide price transparency and lower prices, reducing margins and eventually driving many physical shops out of the market. Among online shops, price transparency tends to drive shops’ economic profits down to zero, such that the platforms rather than online shops benefit from e-commerce and make economic profits. Price transparency, information on quality, and buyers’ feedback may lead to the same winner-takes-all phenomenon in retailing, and the effects could spread to manufacturing due to transparent price and quality. The more retail is transferred to online shops, the fewer people are employed in retail and manufacturing because of the winner-takes-all effect. Consumers mainly do online shopping, including payments, so there is a substantial saving in labor by moving online. However, the saving in labor by online shops is probably competed away by the transparent price and multitudes of suppliers. The platform providers have near-zero marginal costs; any increase in the consumption volume is likely to be reaped by them rather than the online shop owners. Therefore, intellectual and intangible assets are more effective revenue generators in the Robotic Age.

4.2. Intellectual and Intangible Properties and Wealth Distribution The high market capitalization or the enormous wealth accumulated very quickly by successful Internet entrepreneurs must come from somewhere. One explanation of the source of Internet firms’ wealth is the capacity that Internet firms have created to provide services to the members of society. The intangible assets should be valued in terms of their ability to earn profits, similarly to valuing physical capital stock. As discussed earlier, traditional manufacturers are employing fewer and fewer human workers because of the growing application of automation and robots. Internet firms employ even fewer human workers than traditional manufacturers. The wealth will

Income and Wealth Distribution in the Robotic Age

373

be more concentrated with Internet firms than traditional automated firms. Society will be more polarized in terms of wealth distribution. Internet firms providing advertising services compete with traditional advertising businesses and obtain some of the traditional advertisers’ previous income sources. Besides, these Internet firms also take wealth from traditional manufacturing and service businesses. Because the competition for customers’ attention becomes more intense with the reduction in the costs of transportation and communication tools, manufacturing and service firms have to spend more on publicity and marketing. More spending on adverts increases the profitability of Internet advertising firms such as Google and Facebook and decreases the profitability of the manufacturing and service sectors. The low communication and transportation costs may increase the profitability of a few star firms in the manufacturing or service sector, but the sector’s overall profitability will be significantly reduced. Overall employment will decrease with the development of e-commerce. Shop assistants will largely disappear along with brick-and-mortar shops. Although courier services will employ more people, the number will be much smaller than that lost by closing physical shops. With more efficient logistics and eventual delivery by automatic systems, people employed by the courier services would also become fewer and fewer. The overall trend will be that 1) fewer and fewer human workers will be employed; 2) most manufacturers are operating at zero economic profits, and only a few famous brands make some economic profits; 3) the platforms have zero marginal costs and reap the most economic profits. The Internet and cheap courier services increase competition by reducing information and transport costs and lowering prices. The lower prices will increase overall consumption, which usually means higher demand for capital and labor. However, economies of scale and automation imply that the number of employees due to high demand caused by lower prices will be far fewer than that lost because of automation and economies of scale. Internet platforms have zero marginal costs and need much fewer employees to operate. Therefore, profits will be more concentrated in the hands of a few large shareholders and senior employees of the successful platforms. The wealth concentration process with successful online platforms will be much faster than in the First and Second Industrial Revolutions. The initial fixed investments were usually substantial in the First and Second Industrial Revolutions. No matter the scale, the marginal costs for producing physical goods were also far from zero. Therefore, the

374

Chapter 7

wealth accumulation process and the production expansion were more gradual than for Internet firms with intellectual or intangible assets. Facebook and Google earn their income from advertisements. The services they provide, social networking or search engines, are the medium for advertising. In this sense, they are not much different from their traditional counterparts, such as newspapers and television. Newspapers and television attract more readers and viewers by providing good reports and programs so that firms would buy advertisement space and time because of their large readership or audience. Unlike traditional media, these Internet firms have zero marginal costs and need not provide content to keep their users. The network effects mean that the more users the social networking firms like Facebook have, the more likely that even more new users will join the network. Search engines like Google might have few network effects, but with more users, its brand becomes more famous and will attract more new users. This new medium will compete with traditional media while employing fewer human workers, increasing wealth concentration. The importance of intangible assets for Internet firms leads to their marginal costs being negligible; their fixed costs are usually much smaller than the brick-and-mortar businesses, and their employees are much fewer than those of a traditional business. All these features of successful Internet firms result in more wealth concentration. The high pay and high-level freedom of work at Google have been regarded as the cause of its success and imitated by some high-tech start-ups. This view has mistaken results as causes. Google and other high-profile Internet firms can pay well and give more freedom in work because they make huge profits too easily. Their current profits do not depend on their workers’ current efforts since Google’s primary income source is advertising, which can almost let cash flow into Google’s coffer without any further efforts from Google’s current employees. Other businesses depend on their employees’ efforts to generate cash flow and survive. Imitating Google in giving high pay and staff autonomy is an assured path to bankruptcy. Since the vast revenue-generating power of those firms' intellectual and intangible assets, wealth and incomes have become more concentrated than during the Machine Age. As the more affluent people get a larger and larger share of the national income and can no longer consume much more physical goods (and intellectual goods), the only place they can use their increasing incomes is assets, financial assets, and real estate. The large amounts of income flow into asset markets will increase asset prices. When financial assets are

Income and Wealth Distribution in the Robotic Age

375

chased by the increasing incomes of the elite classes, for some time, the higher returns will be self-reinforcing and appear to make every member of the society with some assets richer. The more affluent people become much wealthier than the less affluent regarding the percentage increase in their wealth. When the less wealthy (of the working classes) get more financial wealth through capital gains, they are likely to spend more money on physical goods, which might exert some pressure on the price level of physical goods. This price pressure from increased financial wealth is less important than real estate assets. The price level of real estate is also pushed by the increasing incomes of the more affluent classes, and consequently, the prices of physical goods increase. The consequences of growing demands for financial assets and real estate are somehow different. Increasing demand for financial assets may lead to low yields of corporate bonds, which facilitates the development of new industrial sectors like ICT, robotics, and social media. Still, the traditional manufacturing sectors might not attract much new investment. Increasing demand for real estate means rising commodity demand, which generally pushes up the price level. As house prices continually increase, even people with no income can “afford” to buy houses if financial sectors give them a loan. The “constantly” growing house price implies that if someone defaults on their mortgage payment, they still have a net wealth after selling the house. The rising house price makes the banks feel safe to lend money to people with poor credit ratings. The increased loans push house prices up further, so positive feedback forms. The less affluent households feel wealthier when their houses or apartments become more valuable on paper and decide to consume more luxurious goods, which further pushes the price level up, in addition to the effects of higher commodity prices due to the housing market boom. When the price increase leads to an inflation rate higher than that deemed healthy by central banks, they will raise the interest rate to prevent run-off inflation. Continuing asset booms make purchasing assets with borrowed money a “safe” investment behavior. Gradually, more and more people with the dubious ability to repay the money get on the asset ladder. Raising interest rates would make most investors with dubious credit rating default on their borrowing, leading to a financial crisis. The cause of the 2007–2008 financial crisis was a consequence of sub-prime mortgages and associated financial instruments intended to provide safer assets or protect against defaults.

376

Chapter 7

4.3. Intellectual Property, Information Technology, and Productivity Two mechanisms reduce aggregate wage income and increase the share of capital owners in the net national products (NNP). One is the decrease in the number of employed workers; the other is the stagnation of median real wages compared with the growth of profits in real terms. ICT could make performance-type jobs reach more audiences. Movies, records, radio, television, videos, and the Internet, especially the mobile Internet, make music, drama, and other performance arts more competitive globally, driving out local talent from this job market and making superstars more “productive”. Similarly, it can be imagined that videos of superstar professors can replace most university professorial jobs, and tutorials can be provided by Watson- or ChatGPT-type AI, throwing millions of wellpaid university professors into the unemployment queues. Suppose students prefer the feeling of having the lecturer personally on the scene. In that case, humanoid robots mimicking superstars’ performance for lecturing need only weak form AI and should be achievable in the not-too-distant future. The professor robots can also have Watson function linked to cloud computing. Empirical social science research might be performed and written up more efficiently and objectively by AI systems using deep learning, data mining, and automatic writing software. Human thinking is not as mysterious as many believed, and it could be summarized into patterns for machine learning, for example, John Stuart Miller’s canons of induction (Ducheyne 2008). Once machines have learned all identifiable human ways of thinking, they may combine and reorganize the patterns into a more diverse collection of thinking patterns and use them to raise hypotheses. Standard statistical approaches, including econometric methods, can test these hypotheses. Social science research is likely to be taken over by machines much earlier than natural science research because social science research is essentially a data mining and analysis exercise that does not involve handling physical materials. Other professional jobs would also be replaced by AI or robots in one form or the other, leaving few star performers who can advise the designers of professional robots and AI. Robots were first created to operate in dangerous, unpleasant environments or perform repetitive and boring processes (Fischetti 1985; Belforte et al. 2006). With the advances in robotics and AI, most or all jobs, which have been deemed as something that nobody wants, can be carried out by specially designed robots. In the Robotic Age, it is likely that most people will not have a job, and they will live on some income support from the

Income and Wealth Distribution in the Robotic Age

377

social security system. When fewer and fewer jobs need human workers, the labor productivity measured by human workers becomes higher and higher. Moreover, when at some point, robots with the ability to complete complex tasks will be created and their manufacturing costs substantially reduced to a level where it is much cheaper to use robots than hire human workers, the growth of labor productivity measured by human workers will be accelerated. Humanity will enter an era where numerous robots perform the production processes of society with a few human workers. All the unemployed workers will have a guaranteed income from the social security system, sufficient for living a respectable life. By then, many humans may want to be human workers without pay. When most people want a job, even without payment, society needs to find a way to allocate working opportunities fairly. These jobs, if exist, are the ones that robots or AI cannot do. They may be assigned by randomly matching jobs with human workers. The development of ICT and robotics will initially increase the productivity of the elites and enable them and capital owners to reap the dividends of technological progress. With the further advancement of technology, fewer and fewer elites can still enjoy the benefits of new technologies and share the national incomes with the capital owners. Eventually, when robots with AI are more competent than human superstars in most fields, capital owners will reap all the dividends of technological progress if humanity can still control those super-intelligent robots. By that time, human labor productivity will no longer be applicable economically.

5. Immigration and Wealth Distribution The loss of jobs by blue-collar workers in developed countries is mainly caused by productivity growth and the saturation of physical goods for consumption. In contrast, globalization and cheap imports are the secondary cause. However, many people in developed countries mistake international trade and foreign immigrants as the principal culprits for job losses. The decision by referendum in the UK to leave the European Union in 2016 illustrates this misconception. Former US president Donald Trump also won the election in 2016 by blaming foreign countries’ trade policies and illegal immigrants for ‘stealing’ American jobs. As we discussed earlier, restrictions on international trade and immigration will not solve the problem of stagnant and decreasing incomes of blue-collar workers, which becomes more severe as middle-class families have had stagnant or reduced incomes in the past 20 years. Restrictions on international trade increase labor share in the national incomes, ceteris paribus. Still, they are not conducive to

378

Chapter 7

global economic growth nor the rise of the living standard of domestic residents in the long run.

5.1. Immigration and Unemployment Humans tend to show more compassion to those familiar or similar to themselves. Therefore, they are less disapproving of welfare benefits to people similar to them than to new immigrants, especially when the new immigrants take advantage of the more generous social welfare system in the developed countries. On a broad scale, this issue is related to how much compassion people in one country have toward people in other countries and whether the domestic economy requires more labor supply. There are different types of immigrants into developed countries: 1) people find a professional job in a developed country and move there to take up employment; 2) people enter a developed country as dependents; 3) people enter a developed country as refugees; and 4) people enter a developed country as illegal immigrants. Immigrants would be welcome and embraced in a growing economy short of labor supply, such as in Western European countries in the 1960s. Generally speaking, if the developed country has reached an economic stage where nearly all consumption needs have been met, new immigrants will not increase their average income. Blue-collar, working-class households have suffered the most from the continuing decline of the traditional manufacturing industry. People entering the country as dependents or refugees often use public resources, while those illegal immigrants would take low-paid jobs and may undercut legal residents in such employment. In recent years, because of concerns over the low support ratio (ratio of the working-age population to people aged 65 and above) and population aging, many economists in developed countries urged their governments to loosen restrictions on immigration (Simon, Belyakov, and Feichtinger 2012). However, with slow and sometimes jobless recovery after the 2008 financial crisis, immigrants again become scapegoats for the high unemployment rate in developed countries. Donald Trump made illegal immigrants and unfair international trade practices by other countries key topics in his election campaign in 2016. It is generally agreed that blue-collar, working-class votes played a crucial role in Trump’s election success (Gusterson 2017; Morgan and Lee 2018; Reny, Collingwood, and Valenzuela 2019). The low fertility rates in developed countries have increased the need for migrant workers, but the progress in AI and robotic technology will reduce that need. As AI systems and robots will take over more human jobs, some

Income and Wealth Distribution in the Robotic Age

379

measures might need to be implemented to alleviate the complaints about immigrants by blue-collar workers, especially those who have lost their jobs. When there are many people out of work and more than enough talent for the domestic economy, it might be reasonable to put more strict restrictions on immigration. Countries generally encourage people with skills in various professions to immigrate because those talents are in short supply in the domestic economy (Shachar 2006; Iredale 1999). What countries want to restrict are low-skilled workers and those intent on getting welfare benefits. How strict the immigration laws should be is up to the acceptance and readiness of the population toward people from other countries.

5.2. Immigration and Social Security Human beings are sympathetic to others, especially to those close to them. People in developed countries donate a large sum of money to least developed countries. Would they be happy to let people in those countries migrate to their country and live on social welfare? If immigrants and refugees are granted entry, the host countries must provide them with social security benefits to ensure their standard of living. In recent years, antiimmigration sentiments have grown in Europe and North America (Dennison and Geddes 2019; Pettigrew, Wagner, and Christ 2007), especially when economic growth stagnates. Anti-immigration parties are gaining ground in many European countries. Immigrants are perceived as taking local citizens’ jobs, living on social security benefits, and having higher criminal activities in the United States and Europe. Many immigrants and refugees from underdeveloped countries lack the skills to participate in economic activities and make a living. They live on the host country’s social security benefits and often keep their customs that might be frowned upon by the mainstream thought in the host countries. Immigrants from underdeveloped countries are more likely not to follow the rules because compliance with them in their home country often implies failure and misfortune. Their experiences based on survival instincts make them more adept at exploring the defects of the social security system and making inappropriate or fraudulent claims for benefits. In the Robotic Age, it is likely that most people will not have a job and live on some form of income support from the social security system. Some people without a job need no income support because they have sufficient capital income to support themselves. A social security system will ensure all citizens have a respectable life, but citizens may not be willing to allow

380

Chapter 7

immigrants into their territory to share their social security benefits. Regulations will be maintained to restrict immigration.

5.3. Failed States, Refugees, and the International Community In modern times we have the phenomenon that, on the one hand, the dictators enjoyed support from a particular group of people willingly or because of deceptive propaganda and oppression; on the other hand, the rest of the population would rather flee from the country or get the dictators overthrown. Regime changes orchestrated by Western countries in the past in countries ruled by dictators usually led to chaos and civil wars (Clements 2013; Etzioni 2012; Cordesman 2016). After the invasion of Afghanistan and Iraq by the US and the UK and intervention in Libya by France and the UK, Western countries became more reluctant to intervene in civil wars in developing countries ruled by dictators. It seems that regime changes in (relatively) developed countries can be better dealt with. The end of communist rule in eastern European countries and the transition to democracy were generally peaceful except in Romania. The transition to a democratic regime in Germany, Italy, and Japan after World War II was swift and successful. The Allied forces' occupation might have helped Germany and Japan's democratization process. Should the international community let the poorest countries and failed states fester while the developed and newly industrialized emerging economies move swiftly into the Robotic Age? Suppose we want to avoid such scenarios becoming a reality in the future. In that case, the international community must take measures to get the poorest countries and failed states back on track for fast economic growth. There are many causes for a country to be poor and grow slowly, even go backward. These causes can be natural or human-made. Natural causes include harsh climates and a shortage of natural resources, such as arable land, minerals, and water. Human-made causes are generally related to poor governance. For poorer countries, due to the harsh natural environment, the international community could help improve the communication and transportation infrastructure and provide better access to the markets of developed countries. The international community could offer those small, poorer countries the choice to become associated with one or two developed countries, i.e., become overseas territories of developed countries if democratic decision-making is possible. This neocolonialism is to help the poorest countries achieve a modern level of administration and efficiency. The countries can be under direct rule appointed by the developed countries for 20 years or so and then return to a democratically elected government at the local level. For small countries

Income and Wealth Distribution in the Robotic Age

381

under dictatorship, the international community might consider forcing the dictator out and imposing direct rule for transition to a democratically elected government. The above approach is suitable for only small countries whose size and population are negligible compared with the developed countries that will take over. For larger countries under dictatorships, the international community may negotiate with the government or neighboring countries to set up concessions of large areas to accommodate refugees from failed countries. These concessions are areas managed by officials appointed by developed countries and are intended as economies for everyday life rather than refugee camps. Such concessions are hoped to stop refugees and illegal migrants from entering developed countries. Developed countries will run the concessions which are expected to undergo fast economic growth. As the concessions become more successful, they can expand and cover larger areas. The success of these concessions will encourage the surrounding countries to move from dictatorship to democracy. Such neocolonialism is to help people in underdeveloped countries and enables them to catch up with the fast-improving standard of living in developed countries. Because more and more developed countries will implement guaranteed incomes for the majority of the population due to the robotization of the production and service processes, citizens become less and less willing to accept immigrants during a particular stage of their development. Neocolonialism helps people in underdeveloped countries without letting them enter developed countries, which differs from the old colonialism and will be more efficient and less controversial in developed countries. Many former Western colonies have not progressed much under the administration of independence movement leadership.

6. Guaranteed Income and Social Security Systems We should expect that the impact of globalization on developed countries should lead to a faster increase in their GNP and the GNP of developing countries. For this scenario, the government (and the politicians) in developed countries must share the income from the additional growth between the rich and the working classes. Protectionist measures would only work when international trade reduces the GNP of developed countries and increases the income of the capital owners in those countries. From the 1970s, however, a large part of the declining manufacturing employment can be attributed to technological progress, which protectionism cannot address. In the Robotic Age, there would be no or only a few human workers

382

Chapter 7

in the economy, so most of the population would live a respectable life with income allocated by society. In a democracy with a well-educated working class and a population whose median income is far below its average income, political pressure in the election will be able to establish a fair system to achieve that. What proportions should be divided between the richer and the poorer are influenced by the culture and social ideology formed by thinkers, academics, and propagandists.

6.1. Mass Unemployment and Minimum Income Support The issue of stagnant and decreasing income or unemployment of the majority cannot be addressed simply by protectionist and anti-immigration measures because the leading cause, in the long run, is technological progress. They can be addressed by a guaranteed basic income set at a level that makes almost everyone better off. Guaranteed basic income is a much better solution to blue-collar unemployment than restricting or prohibiting international trade. As long as free international trade increases total national income and wealth, developed countries should promote free trade. Utility maximization for the rich and the poor can be achieved by appropriately redistributing national incomes. The declining incomes of blue-collar and middle-class families should be addressed by a guaranteed minimum income funded through taxation. The guaranteed minimum income can be a lump sum regardless of existing income, so the disincentive to employment is low. A guaranteed minimum income is not a new idea. Although most economists currently would reject it based on efficiency grounds, some free marketer economists such as Milton Friedman had supported it (Pressman 2005; Rothbard 2002), and the Nixon administration in the United States considered implementing it during the early 1970s (Quadagno 1990; Dawson 1976). Whether a guaranteed minimum income for everyone can be established depends on the proportion of unemployed working-age people. When the proposal of unconditional basic income was rejected in the Swiss referendum (Marais 2020), many commentators hailed the sensibility of the Swiss people. Such comments neglected the trend of increasing wealth concentration and the impact of AI and robotics on employment in the long run. When the proportion of working-age people not employed approaches 50% of the working-age population, a referendum will support unconditional basic income. Why did the Swiss referendum reject the proposal? The unconditional basic income was suggested at CHF2500. The referendum on 5 June 2016

Income and Wealth Distribution in the Robotic Age

383

dismissed the proposal for total basic income by a majority of 76.9%. Only 23.1% of the votes support the scheme. Swiss people rejected the proposal because those who earn much less than CHF2500 are still a small minority in Switzerland. Most people will contribute to financing such a scheme instead of benefiting from it, even if the country can afford it. If the income gap between the rich and the poor becomes larger, and the wealth is more concentrated into a smaller and smaller group, there will be more support for the unconditional basic income. When the proposed basic income is larger than the median income (and much smaller than the average income, theoretically affordable by the country), a referendum will likely support an unconditional basic income. It should be a simple arithmetic to judge which direction a voter would choose regarding a new measure, depending on whether they gain or lose financially. When the proposed basic income is smaller than the median income, most of the working population will be net contributors to such a scheme and reluctant to support it. When the basic income is larger than the median income, most of the working population will be net beneficiaries and likely to support it. There might be consumers who would reject such a scheme even though they benefit from it. Such opposition to unconditional basic income could arise from economic and ethical concerns. A universal basic income system may cause economic inefficiency, as viewed by mainstream economists. Still, it is a necessary step in the Robotic Age when efficiency or productivity is much less a concern than fairness and equality. Most of the population needs a guaranteed income above the median to support such a scheme. How large should this extra income be to get the majority’s support? Three factors may determine the size of the extra income: 1) the sustainability of the scheme, as too large an extra income will bankrupt the economy; 2) the difference between the median income and the average income, as the larger the difference is, the larger the extra income; and 3) the dominant view on fairness and efficiency. The third factor is heavily influenced by culture and propaganda, and the view fluctuates between equality in outcome and efficiency leanings. From the end of World War II to the early 1970s, the view on fairness and efficiency swung to equality, and government policies tended to reduce the income difference between the rich and the poor. The slowdown and stagflation in economic growth in developed countries during the 1970s were thought to be caused by Keynesian policies and social welfare. Free marketers and monetarists became the dominant influence in economics and political thinking. Deregulation and privatization became

384

Chapter 7

essential features of government policies (Marsh 1991; Majone 1994). Deregulation and privatization have been considered among the causes of the Great Moderation. However, during the 2007–2008 financial crisis, governments of developed countries intervened heavily to rescue the financial institutions and became the main shareholder of many troubled banks. The Great Moderation derived from deregulation and privatization ended or was disrupted by a crisis that cannot be left to the market to sort out. The mainstream economists supported government intervention and public ownership to a large extent to avoid a global economic meltdown, as they viewed the consequence of the financial crisis without government intervention. When reexamining the booming 1950s and 1960s, during which major Western countries had their fastest growth rate in history, we found that the social welfare systems implemented after World War II promoted this rapid growth by sharply increasing citizens’ disposable incomes and reducing their precautionary savings for childcare, healthcare, and misfortunes. Technological progress in the West by the 1940s had reached the level where insufficient demand would seriously slow economic growth. Installing and improving social welfare systems provided the growing demand in the 1950s and 1960s, which stimulated economic growth. The boom in the 1950s and 1960s might be called the “social welfare boom.” Besides market saturation of manufactured goods, stagflation arose partly because Western economies had reached a new equilibrium after the growth-promoting effects of social welfare gradually disappeared. With AI systems and robots sharply increasing productivity, a guaranteed basic income scheme will promote economic growth for certain periods.

6.2. Politics, Employment, and Social Security The majority will not tolerate a massive gap between themselves and the more affluent people in a society. In the past, the stability of society was maintained on two cornerstones; one was the brainwashing by the ruling class (Varma 1975), and the other was the ignorance of the majority. Brainwashing was essential for the ruling class to maintain stability and was achieved by religious and ethical teaching. The masses were taught that God installed the emperors or kings and lay people had to obey them. The hardship endured by ordinary people was the arrangement by God for a purpose, so they had to accept it and be satisfied with their situation. Many morality teachings required people to be loyal and obedient to their monarchy. Such brainwashing was generally very effective in preventing the oppressed from uprising against the ruling class. The ordinary people

Income and Wealth Distribution in the Robotic Age

385

tended to put up with the oppression and hardship, rebelling only when life could not be endured. The majority's ignorance resulted from a lack of education and difficulty obtaining relevant information. Surveys in the US showed that most Americans prefer a wealth distribution much more equal than the actual situation. They wrongly thought the US was an equal society (Starmans, Sheskin, and Bloom 2017; Lane 1986). The widespread application of ICT makes information more readily available, and mass higher education worldwide leads to a higher proportion of the population with a university education. Both factors make the masses more knowledgeable and less likely to believe what the ruling elites tell them. The two events in 2016, the Brexit ‘win’ in the British referendum and the election of Donald Trump as US president are the consequences of the emergence of better-informed and educated working classes. In the Brexit referendum, the three main political parties (at least their leaderships) supported the Remain camp. Most experts, especially those in economics, finance, and business, strongly support remaining in the European Union. However, ordinary people, especially those of low incomes, strongly supported the Leave camp (Freeden 2017). Brexit illustrates well that well-educated low-income people are less inclined to vote according to the view of elites than based on their economic situation. The US presidential election illustrated the same point again. Most media persons and even political big weights in the Republican Party refused to support Donald Trump, but he still won with support from some traditional democratic strongholds (Francia 2018; Devinatz 2017). Technological progress and globalization facilitated economic growth in the US, but the benefits have been reaped mainly by the more affluent. The working class is worse off. The current blue-collar workers in developed countries are probably the most educated and knowledgeable in history, so they voted on their judgment of the economic situation. As automation and globalization depressed the real wage growth for blue-collar workers, they would support the candidate who claims to return the work opportunities to them. At the moment, automation and globalization cause a decrease in employment, especially for blue-collar workers in developed countries. However, as historical data support the view that technology enhances employment and improves working conditions, blue-collar workers and politicians blame globalization for unemployment. The supporters of Brexit thought that workers from East European countries took over UK jobs and

386

Chapter 7

drove down the UK wages for blue-collar workers because of the free movement of goods, services, and people in the single market. Therefore, withdrawal from the European Union would be the best solution. Similarly, blue-collar workers in the US were thought to be the most important part of the electorate for Donald Trump’s election. The solution prescribed by Donald Trump was deregulation and protectionism, which is unlikely to be successful in the long run. Although globalization exacerbates the impact of technological progress, the driving force for the increasing unemployment is technological progress. Technological unemployment cannot be solved by deregulation. Protectionism may slow down the transfer of production to low-labor-cost countries at the cost of inefficiency, but it cannot reverse the trend of reduced labor input brought about by technological progress. Eventually, the population will realize that deregulation, which promotes polarization of wealth distribution, and protectionism in international trade, cannot relieve the depressing effects of technological progress on the employment and incomes of bluecollar workers and, gradually, the middle-class, white-collar workers. By then, politicians preaching deregulation and protectionism will lose their luster as the advocator for the working class; left-wing politicians advocating social equality and taxing the superrich to finance social welfare for working-class households will become popular with the electors. With the election of left-wing politicians, developed countries will move toward the Scandinavian-type socialist or social democratic system. Since World War II, mainstream economists have tried to decorate economics as a discipline like natural sciences. Economics has been divided into positive study and normative study. The positive study is supposed to investigate what it is in the economy, and the normative study is to investigate what it should be. At the microeconomics level, this distinction is meaningful. At the macroeconomics level, economic policies and their impacts are less clear-cut regarding whether they are normative or positive. For example, should the interventions by US and UK governments to rescue the failed banks be investigated as normative studies or positive studies? Should implementing a social security system be viewed as a normative action or process, which a positive study should investigate? Politics is the consequence of economic interests and inherited cultural influences. Ultimately economic issues will be solved politically when market solutions are not acceptable to the (decision-makers of) society. In the Robotic Age, robots and AI programs will make the demand for human labor far less than the supply for human labor, such that the price for human

Income and Wealth Distribution in the Robotic Age

387

labor is much lower than that when the demand for human labor is higher than the supply. In politics, two situations will destabilize the current system: 1) one group of people in society feel that they cannot survive without breaking the existing arrangement; 2) one group of people feel that the current arrangement leads to outcomes so unfair that they could not tolerate it any longer. In the Robotic Age, the current market arrangement will lead to extreme polarization in wealth distribution. Therefore, social welfare to guarantee a minimum unconditional basic income for every adult member of society with additional benefits for young children will be inevitable. To be sustainable, the unconditional basic income has to be smaller than the average income. It has to be larger than the pre-redistribution median income as well. If smaller than the median income, it may not get enough support in the population. How will people respond to a social security system that guarantees a minimum basic income smaller than the median? Suppose it is means-tested and supplements people with less than the minimum basic income. In that case, it will be rejected by most of the population because they will be the net contributors to the benefits of those who do not work. Supplementary basic income could get majority support only when the combined value of the supplement and the existing income exceeds the median income. Suppose the basic income is an unconditional additional income regardless of a person’s income. In that case, there will be less objection from those with substantial incomes, and there is little disincentive to people’s willingness to work. The main objection will be doubts on whether it is financially sustainable, efficiency promoting, and welfare enhancing. More funds must be collected via taxation from high-income people and businesses to finance such an unconditional scheme. At a certain percentile, an individual’s contribution to the scheme equals the benefits received; hence there is zero net contribution. Suppose this individual is below the bottom 50th percentile, meaning most people make a net contribution. In that case, the scheme will be rejected because more people make a net contribution to such a scheme. As people’s choice is influenced by economic considerations and cultural and spiritual edifications, they might support a scheme in which they are net contributors. For the maximum likelihood, we can assume that the majority will support guaranteed basic income when the individual with median income receives net payment from (or makes a net negative contribution to) such a scheme. Therefore, guaranteed basic incomes can only be implemented when most people live with an income much lower than the average income of society.

388

Chapter 7

With a guaranteed income, everybody can pursue whatever their interest is, and robots and AI will challenge many of these pursuits. For example, Go tournaments must exclude machines to give human players a chance. Machines may write better poems than future human poets when machines can be trained to “appreciate” the beauty of artistic works. Machines may paint better pictures than future human painters. As machines might do better than most people in their personal pursuits, the utility of personal pursuits would be more likely to be self-entertaining than fulfillment. In such scenarios, the remaining human jobs would be highly respectable, and many would bid for them without pay because of the sense of fulfillment and achievement. People who hold the remaining human positions will become the community’s leaders. In addition to a guaranteed basic income, there may also be restrictions on the private use of land. The world is more populous than before, which implies that the average personal space is decreasing and land resources have become increasingly scarce. In a democracy, when most people get less personal space, it is unlikely that the more affluent people can grab more and more land for their vanity. As a rare resource, land should be sold at a premium and taxed for owning it, facilitating the optimal allocation and use of one of the scarce resources. To protect the land, there will likely be progressive taxation on possessing and using land or a house such that very few people or households will own a home substantially larger than apartments for people relying on social security benefits.

6.3. Entrepreneurs and Capital Owners in the Robotic Age For the system to be sustainable, most of the population must have the political and economic wisdom to give sufficient incentives to entrepreneurs and capital owners. The unconditional basic or guaranteed minimum income must be smaller than the average. The entrepreneurs and capital owners can earn above the new median income after implementing the guaranteed minimum income. When more than half of the working-age population has no job, the new median income with the minimum guaranteed income is the minimum guaranteed income. The extra income above the new median incentivizes entrepreneurs and capital owners to continue their role in wealth creation for society. While society taxes entrepreneurs and capital owners for their high personal incomes and their firms’ considerable profits to finance the guaranteed minimum income scheme, it also should protect them for their deserving incomes and business activities.

Income and Wealth Distribution in the Robotic Age

389

At some stages of the Robotic Age, when robots and AI can perform everything, including enterprises, inventions, and innovations, and everybody has been provided what they need, the role of entrepreneurs and capital owners is no longer needed. All human beings need to do is legislate and enjoy themselves. That will be a society where everybody can pursue their interest without concern for their living costs. If a firm has no human workers, all its net revenue will go to the owners. The market mechanism will lead to the absolute impoverishment of poorer classes, and the wealthy class has to consume almost all the goods and services produced in the economy. Of course, such a scenario would not occur in human society because politics and culture always intervene in the market process. Even without political intervention, an economy overly reliant on the wealthy classes for the aggregate demand is not sustainable for long-term economic growth. It is like an economy whose growth relies excessively on investment in heavy industries producing investment goods, as in the former Soviet Union. The widespread application of human-replacing technologies in the Robotic Ages will significantly enhance the productivity of human workers (to infinity at the limit). Still, the aggregate consumption of society has its limit because of finite time endowment. When the economy’s output can entirely meet society’s (time-endowment exhausting, i.e., maximum) consumption demand or even long before that, society’s value system might irreversibly change to a state where enormous personal or private wealth no longer has much value, contributions in terms of scientific discovery, technological invention, artistic creation, sports achievements, and even having a job that robots cannot perform will be of more value than material wealth. When personal/private material wealth is no longer highly valued by society, the social welfare system will play a big role in maintaining the economy’s aggregate demand. At the early stage of the Robotic Age, the income level provided by society to every member of society will be relatively low, and people will have a high level of consumption through extra efforts in earning income above that provided by the social welfare system. The necessity of providing an income to every member of society arises from 1) humanreplacing technology will decrease the wage rate by reducing the demand for human workers, and 2) the economy’s growth has to be sustained by adequate aggregate consumption demand. We have discussed only a scenario in which AI surpasses human intelligence in areas where humans want them to be, such as AlphaGo in Go

390

Chapter 7

and Watson in Jeopardy. The future will be different if robots have general intelligence above human general intelligence and possess motivations of their own and free will. We are not sure what attitude super-intelligent robots will have toward human beings, their former bosses. In that scenario, super-intelligent robots will treat humans like pets, like humans treat their family pets.

7. Summary A difference in income and wealth is necessary to motivate people to make more effort. However, many people with below-average incomes and those inspired by altruism would like income equality. Equality in pay and wealth has been the ideal of the communist movement and many ancient thinkers. The practice of outcome equality by the communist movement has generally failed economically, so communist or socialist ideas usually are discredited in the economic community. However, when a polarization in income makes some people feel they cannot survive under the existing politicoeconomic system, a violent revolution may occur and destroy the capital stock accumulated over generations. Globalization increases the income of capital owners in developed and developing countries and workers in developing countries but decreases the income of blue-collar workers in developed countries. In the long run, technological progress in AI and robotics is the leading cause of unemployment and the stagnant income of blue-collar workers. With technological unemployment looming large, unconstrained immigration can create further pressure on employment, and immigrants also require social welfare resources. When productivity has increased enormously, the efficiency and productivity loss due to equality or reduced inequality in income can be accepted by society. Therefore, a guaranteed basic income will be granted to all members of society. Entrepreneurs and capital owners can still have more income and wealth in the Robotic Age, incentivizing their efforts to innovate and create more value for society.

References Amiti, Mary, and Caroline Freund. 2010. “The anatomy of China's export growth.” In China's growing role in world trade, edited by Robert C. Feenstra and Shang-Jin Wei, 35-56. Chicago, IL: University of Chicago Press.

Income and Wealth Distribution in the Robotic Age

391

Belforte, Gustavo, R Deboli, Paolo Gay, Pietro Piccarolo, and D Ricauda Aimonino. 2006. “Robot design and testing for greenhouse applications.” Biosystems Engineering 95 (3):309-321. Bergenstock, Donna J, and James M Maskulka. 2001. “The de beers story: are diamonds forever?” Business Horizons 44 (3):37-44. Best, Kirsty. 2003. “Revisiting the Y2K bug: language wars over networking the global order.” Television & New Media 4 (3):297-319. Brynjolfsson, Erik, and Andrew McAfee. 2014. The second machine age: Work, progress, and prosperity in a time of brilliant technologies. New York: WW Norton & Company. Buckley, Peter J. 2009. “The impact of the global factory on economic development.” Journal of World Business 44 (2):131-143. Clements, Ben. 2013. “Public opinion and military intervention: Afghanistan, Iraq and Libya.” The Political Quarterly 84 (1):119-131. Cole, Jonathan R, and Harriet Zuckerman. 1984. “The productivity puzzle.” Advances in Motivation and Achievement 2:217-58. Cordesman, Anthony H. 2016. US Wars in Iraq, Syria, Libya, and Yemen: What Are the End States. Washington D.C.: Center for Strategic and International Studies. Dawson, Irving. 1976. “Bowler, “The Nixon Guaranteed Income Proposal” (Book Review).” Social Science Quarterly 57 (2):488. Dennison, James, and Andrew Geddes. 2019. “A rising tide? The salience of immigration and the rise of antiဨimmigration political parties in Western Europe.” The Political Quarterly 90 (1):107-116. Devinatz, Victor G. 2017. “Donald Trump, George Wallace, and the white working class.” Labor Studies Journal 42 (3):233-238. Douglas, Paul H. 1976. “The Cobb-Douglas production function once again: its history, its testing, and some new empirical values.” Journal of Political Economy 84 (5):903-915. Ducheyne, Steffen. 2008. “JS Mill's Canons of Induction: from true causes to provisional ones.” History and Philosophy of Logic 29 (4):361-376. Etzioni, Amitai. 2012. “The Lessons of Libya.” Military Review 92 (1):4554. Fischetti, Mark A. 1985. “Robots do the dirty work: Some walk, some roll as they go about their tedious, hazardous chores in nuclear-power plants.” IEEE Spectrum 22 (4):65-73. Ford, Martin. 2015. Rise of the Robots: Technology and the Threat of a Jobless Future. New York: Basic Books. Francia, Peter L. 2018. “Free media and Twitter in the 2016 presidential election: The unconventional campaign of Donald Trump.” Social Science Computer Review 36 (4):440-455.

392

Chapter 7

Freeden, Michael. 2017. “After the Brexit referendum: revisiting populism as an ideology.” Journal of Political Ideologies 22 (1):1-11. Garnaut, Ross, Ligang Song, and Yang Yao. 2006. “Impact and significance of state-owned enterprise restructuring in China.” The China Journal (55):35-63. Gordon, Robert J. 2000. “Does the “new economy” measure up to the great inventions of the past?” The Journal of Economic Perspectives 14 (4):49-74. Griliches, Zvi. 1988. “Productivity puzzles and R&D: Another nonexplanation.” Journal of Economic Perspectives 2 (4):9-21. Gusterson, Hugh. 2017. “From Brexit to Trump: Anthropology and the rise of nationalist populism.” American Ethnologist 44 (2):209-214. Iredale, Robyn. 1999. “The need to import skilled personnel: factors favouring and hindering its international mobility.” International Migration 37 (1):89-123. Krugman, Paul R. 1983. “New theories of trade among industrial countries.” American Economic Review 73 (2):343-347. Krugman, Paul R. 1987. “Is free trade passé?” Journal of Economic Perspectives 1 (2):131-144. Krugman, Paul R., and Robert Z. Lawrence. 1993. Trade, jobs, and wages. National Bureau of Economic Research Cambridge, Mass., USA. Lane, Robert E. 1986. “Market justice, political justice.” American Political Science Review 80 (2):383-402. Ma, Qing-Ping. 2020. “The China Miracle in a technological and sociopolitical framework—The role of institutions.” Journal of Chinese Economic and Business Studies 18 (3):253-272. Ma, Qing-Ping. 2023. Review of: Liberalism Caused the Great Enrichment. Qeios. doi:https://doi.org/10.32388/60YPB1. Majone, Giandomenico. 1994. “Paradoxes of privatization and deregulation.” Journal of European Public Policy 1 (1):53-69. Marais, Hein. 2020. “The crisis of waged work and the option of a universal basic income grant for South Africa.” Globalizations 17 (2):352-379. Marsh, David. 1991. “Privatization under Mrs. Thatcher: a review of the literature.” Public Administration 69 (4):459-480. Marx, Karl. 1972. Critique of the Gotha programme, 1875. Beijing: People’s Publishing House. Morgan, Stephen L, and Jiwon Lee. 2018. “Trump voters and the white working class.” Sociological Science 5:234-245. O'Boyle, Thomas F. 2011. At any cost: Jack Welch, General Electric, and the pursuit of profit. New York: Vintage.

Income and Wealth Distribution in the Robotic Age

393

O’Rourke, Kevin Hjortshøj, Ahmed S Rahman, and Alan M Taylor. 2013. “Luddites, the industrial revolution, and the demographic transition.” Journal of Economic Growth 18 (4):373-409. Ohlin, Bertil. 1935. Interregional and international trade. Cambridge, MA: Harvard University Press. Pettigrew, Thomas F, Ulrich Wagner, and Oliver Christ. 2007. “Who opposes immigration?: Comparing German with North American Findings.” Du Bois Review: Social Science Research on Race 4 (1):1939. Pfeffer, Jeffrey. 1995. “Producing sustainable competitive advantage through the effective management of people.” Academy of Management Perspectives 9 (1):55-69. Phillips, Anne. 2004. “Defending equality of outcome.” Journal of Political Philosophy 12 (1):1-19. Piketty, Thomas. 2014. “Capital in the twenty-first century.” In Capital in the twenty-first century. Cambridge, MA: Harvard University Press. Pressman, Steven. 2005. “Income guarantees and the equity-efficiency tradeoff.” The Journal of Socio-Economics 34 (1):83-100. Quadagno, Jill. 1990. “Race, class, and gender in the US welfare state: Nixon's failed family assistance plan.” American Sociological Review 55 (1):11-28. Reny, Tyler T, Loren Collingwood, and Ali A Valenzuela. 2019. “Vote switching in the 2016 election: How racial and immigration attitudes, not economics, explain shifts in white voting.” Public Opinion Quarterly 83 (1):91-113. Rothbard, Murray N. 2002. “Milton Friedman Unraveled.” Journal of Libertarian Studies 16 (4; SEAS AUT):37-54. Ruffin, Roy. 2002. “David Ricardo's discovery of comparative advantage.” History of Political Economy 34 (4):727-748. Samuelson, Paul Anthony. 1976. Economics. 10 ed. New York: McGrawHill. Samuelson, Paul Anthony. 2004. “Where Ricardo and Mill rebut and confirm arguments of mainstream economists supporting globalization.” Journal of Economic Perspectives 18 (3):135-146. Shachar, Ayelet. 2006. “The race for talent: Highly skilled migrants and competitive immigration regimes.” New York University Law School Review 81 (1):117-147. Simon, C, AO Belyakov, and Gustav Feichtinger. 2012. “Minimizing the dependency ratio in a population with below-replacement fertility through immigration.” Theoretical Population Biology 82 (3):158-169.

394

Chapter 7

Smyth, Russell. 2000. “Asset stripping in Chinese state-owned enterprises.” Journal of Contemporary Asia 30 (1):3-16. Starmans, Christina, Mark Sheskin, and Paul Bloom. 2017. “Why people prefer unequal societies.” Nature Human Behaviour 1 (4):1-7. Tomlinson, Jim. 2012. “De-globalization and its significance: from the particular to the general.” Contemporary British History 26 (2):213-230. Van Lange, Paul AM. 1999. “The pursuit of joint outcomes and equality in outcomes: An integrative model of social value orientation.” Journal of Personality and Social Psychology 77 (2):337. Varma, SC. 1975. “Politics of Psychoanalysis.” Social Scientist 3 (10):3-16. Waley, Arthur. 2012. The Analects of Confucius. London and New York: Routledge. Witt, Michael A. 2019. “De-globalization: Theories, predictions, and opportunities for international business research.” Journal of International Business Studies 50 (7):1053-1077.

CHAPTER 8 SOCIAL ORGANIZATION

In the Robotic Age, the social organization will be substantially different from its current norm as the modes of production are dramatically different from its current form. The various social formations during the history of humanity are consequences of the modes of production and the level of production technologies. In the early days of human history, the need to work together for hunting and gathering consolidated the group living mode inherited from apes. The emergence of agriculture made it possible to have a social division of labor and stratify society. We may glimpse early human social organization by observing primates such as monkeys and apes in their social interactions. In a primate tribe, many rudimentary forms of human social relationships and politics have appeared. In a certain sense, the dictatorship in human society sometimes resembles the dominant male in a monkey group (Gintis and van Schaik 2013). The technological progress in production and improvement in human intelligence modified the social relationship inherited from monkeys and apes. The increase in labor productivity and evolution in human intelligence gradually led to the emergence of family, private ownership, and government. The beginning of the Machine Age coincides with the emergence of the capitalist economy and modern democracy. The widespread application of artificial intelligence (AI) and robots will further change the morphology of human society. This chapter will examine how the application of AI and robots will affect the social organization in terms of individuals, communities, firms, professional associations, governments, and the international community.

1. Individuals, Families, and Households in the Robotic Age In economics, we often use households as the basic unit for analysis. A household consists of one or more people who live in the same dwelling and also share meals. It could be just a single individual, a family, or a group living together in the same dwelling. Some forms of the

396

Chapter 8

family exist in other primates, with a single female with her offspring, monogamous family groups, or one-male-several-female groups. For humans, a family is a group of people affiliated either by consanguinity, affinity, coresidence, or some combination of these (Gough 1971). Consanguinity is the property of being from the same kinship as another person (by recognized birth), and affinity is the kinship relationship created or existing between two or more people due to marriage, adoption, and steprelationships. It was believed that human matrilineal families predated the patrilineal families (Morgan 1877; Engels 1972), as it is easier to determine matrilineal lines. This view has been disputed in the West in the twentieth century (Briffault and Malinowski 1956; Harris 2001), but recent genetic and anthropological data support Morgan and Engels (Knight 2008; Hrdy 2009; Opie and Power 2008). The patrilineal families were likely to emerge with some form of marriage to ensure paternity. The growth of private property strengthened the institution of family, and the improvement in the economic and political position of the lower classes also strengthened their family relationship. The strengthened family relationship has faced more and more challenges since the Industrial Revolution, with the growing financial independence of women, modern welfare systems, and occupations previously dependent on the more affluent classes, such as musicians and singers becoming public entertainment providers with huge wealth and international fame. In the Robotic Age, the guaranteed minimum income will ensure a respectable life, and technologies give everybody more free time and the tools for childcare, communication, and travel. Then, family as an institution will face more severe challenges than ever. Science may change this further if asexual and ex-utero reproductions are invented and carried out.

1.1. Individuals in the Robotic Age In the Robotic Age, there will be the following types of human participants in the economy: a) consumers without jobs who are potential workers, inventors, and entrepreneurs; b) community workers who serve the community and carry out jobs that robots cannot perform; c) factory workers who carry out production jobs that robots cannot perform; and d) entrepreneurs and business owners who own a fraction or all of some businesses in one or more countries. Entrepreneurs and business owners would likely get more income from their business than the guaranteed income and might assign human workers’ posts to themselves first. Because jobs are rare and highly respectable, people will perform the roles without

Social Organization

397

pay. Being an entrepreneur is the most reliable way to get a job (created by oneself). We have defined earlier that the Robotic Age is when over 50% of the working-age population or, more precisely, the working-age households are neither employed nor self-employed because of the widespread application of robots and AI systems. Human jobs started to disappear and were replaced by robots before the advent of the Robotic Age. The current period, often called the Information Age, is a transitional stage between the Machine and Robotic Ages. Nowadays, more and more robots are entering production and service processes. The Taiwanese technology company Foxconn announced a three-year plan in July 2011 to replace workers with more robots, increasing them from ten thousand to a million over three years (Brown 2016). When over half of the working-age households become unemployed, it is unavoidable that a law of a guaranteed minimum income will be passed in a referendum. Such a law will give everybody a payment to ensure a decent life. When every member of society has a guaranteed income, people’s motivation will go beyond the scope of meeting basic needs and striving for constant betterment, which is called meta-motivation, a term coined by Maslow (1967). People’s socioeconomic conditions and environmental influences constrain their thinking and behavior. According to the Marxist view, social being determines social consciousness. Over two thousand years ago, during the Spring-Autumn period in China, Guan Zhong >㇑Ԣ@ asserted that people with full granaries knew the etiquette and those well-fed and well-clothed were concerned with honor and disgrace >ԃᔚᇎ㘼⸕⽬㢲ˈ㺓伏䏣㘼⸕㦓 䗡@ (Guan and Rickett 2001). All humans need to feel respected; this includes the need to have self-esteem and self-respect. When their basic needs have not been met, they might give up their self-esteem and selfrespect to meet their basic needs. All people in the Robotic Age will have their basic needs met, including physiological and safety. Therefore, they are concerned more about the level of love, belonging, friendship, esteem, achievements recognized by others, and self-actualization. These changing priorities may fundamentally affect how people behave in economic and political activities. In the Manual and Machine Ages, the need for love and friendship was much tinted by material considerations. The foundation of love and sexual intimacy should be primarily based on attraction, affection, and shared pleasure between individuals. Material considerations often change a person’s choice of sexual partner, and institutions such as marriage were

398

Chapter 8

established to ensure the relative stability of relations based on material concerns. Marriage is a socially or ritually recognized union between spouses that establishes rights and obligations between them, their children, and their in-laws (Haviland, Prins, and McBride 2016). Marriage likely emerged when a man controlling more resources wanted to have sole sexual rights with a woman to ensure that she would only bear his children by giving her access to his resources. Since genetic fingerprinting can identify parentage precisely and many women earn a comparable salary with their partners, the original grounds for the emergence of marriage are disappearing. One interesting phenomenon in human society is that human creations often acquire a life of their own, with more symbolic values regarding religion and morality. Marriage is one such creation. While families with two partners and their children have been the basic units of society and contributed to social stability, many people have been locked in unhappy marriages because they may feel divorce is too much trouble. In the Robotic Age, as people become more independent of each other financially, the bond between spouses becomes more robust because it is purely based on affection between two persons, and financial considerations do not necessarily exist. However, as the original grounds for marriage disappear, marriage may disappear as an institution to ensure rights and obligations. Without marriage as a legally binding relationship, people can stop their relationship as soon as they no longer have affection for each other. Besides love and friendship, individuals will long for achievements recognized by others. One route to achievement could be to hold the few remaining human positions in production and social administration. Therefore, when robots and other AI systems replace most human jobs, individuals will try to get on the few remaining human jobs. They will also engage in hobbies and try to gain recognition by excelling in them. People often need to engage in professional or amateur activities to acquire a sense of contribution or value. Although the social hierarchy has been broken down except for a few elite capital owners (assuming that the society is still based on a market economy), individuals may still care about their reputation, fame, and image perceived by their peers. Esteem will still be an essential factor in interpersonal relationships. People may need fame or glory, especially those with low self-esteem, which often need respect from others. Therefore, in the Robotic Age, individuals will continue to engage in activities that give them the self-respect and self-esteem they need.

Social Organization

399

The most crucial relationship in the Robotic Age will be an interpersonal relationship, as financial constraints will have largely gone. Esteem and self-actualization are critical or perceived as necessary for building interpersonal relations. The self-actualization level of human needs refers to the need for the realization of a person’s full potential. It is the desire to accomplish everything that one can and to become the most that one can be (Maslow 1954). The focus or goals of self-actualization are often different for different individuals; for example, some may want to become artists, while some want to be scientists or engineers. There might be a further dimension of needs, self-transcendence (Maslow 1969), in which the self only finds its actualization in giving itself to some higher goal outside oneself, in altruism and spirituality. In the Robotic Age, with material concerns being largely removed, more attention will be on the need for esteem, self-actualization, and self-transcendence.

1.2. Family Relations Relations between family members are the most critical interpersonal relations. Ancient Chinese thinkers such as Confucius and Mencius built their political theory on the foundation of familial relations. Within a typical family, there are relations between the parents, between parents and children, and between siblings. Parental care and sibling bonding are essential for children’s early development. Maternal deprivation and concordant losses of essential and primal needs could lead to problems in the children’s emotional development, and children with such problems may lack interpersonal skills. The importance of the relationship between parents and children is explained by the attachment theory, which was developed by John Bowlby and Mary Ainsworth (Bretherton 1992; Bowlby and Ainsworth 2013) to describe the dynamics of long-term and short-term interpersonal relationships between humans, especially those between infants and their caregivers. Because of ample free time, family relations could be better nurtured in the Robotic Age. In the late Machine Age, family stability appears to have significantly decreased, and marriage as an institution has been much weakened because of women’s financial independence and the social security system. In developed countries, cohabitation and having children before and outside marriage have become commonplace. Single-parent families have become common in developed countries. Extra-marital sex, homosexual and transsexual relationships are generally tolerated or accepted by society. In the Robotic Age, as guaranteed incomes meet all basic needs, individuals will live together because of their affection and love for each other. There

400

Chapter 8

is no need to have a marriage to ensure rights and obligations between them. Therefore, it is most likely that marriage as a civic institution will disappear in the Robotic Age. Some remnants of marriage may stay as religious and traditional relics. Romantic relations lead to offspring such that family can still exist despite the disappearance of marriage. The disappearance of marriage does not necessarily imply that sexual and family relationships will be more unstable in the Robotic Age than at present (the late Machine Age). Without financial constraints, a romantic relationship based on affection and attraction between two people will likely be more stable than those based on other considerations. Of course, human affection and interpersonal bonds are complex matters. Often the interest is unidirectional; one person may have affection and be attracted to several people. Two people living together may become bored of each other, or one may become bored of the other and want a new relationship. All these could lead to unstable sexual and family relations, despite the disappearance of material considerations. Since children’s parentage can be unequivocal and people have guaranteed income in the Robotic Age, the implicit mutual insurance between spouses in marriage essentially lost its merits regarding revenue, care, and housework. Thus, unstable sexual and family relations will not lead to severe social problems.

1.3. The Impact of Science and Technology on Family The progress of medical science might solve the myth of human reproduction in the future. If the uterus is no longer necessary for the embryo/fetus to grow, then asexual reproduction could be as common as sexual reproduction. If asexual and ex-uterus reproduction is not feasible, women will play a more significant role in family relations and reproduction decisions. Women generally take more responsibility in raising children, and most single-parent families are mother-children. The availability of asexual and ex-uterus reproduction will give women and men the same choice and right to the rearing of children and will provide them with more options on when they want to have children. Biologically, women are more likely to be able to clone themselves because they can produce eggs which are needed for cloning with the current technology. Current medical technologies are still unable to make eggs for in vitro fertilization or cloning, so men will rely on eggs provided by women to clone themselves or to fertilize the eggs with their sperm for reproduction. In the future, eggs might be produced from stem cells, but such stem cells would still come from women.

Social Organization

401

The likely family forms will be a mother and father with children, a mother with children, and a father with children. We may assume that a mother and father with children will be the major form, but it is challenging to be sure that single-parent families will not be the dominant family form. In the Robotic Age, with a guaranteed income and helpful robots, the single-parent family could be the dominant form of family. Robots may even play the missing father or mother role in single-parent families. Whether a twoparent or single-parent family will be the dominant form depends on the fashion of the relationship between the two genders. If the fashion is a more committed relationship between two people, the two parents will be the dominant form. If the fashion at the time is less commitment and more freedom in the relationship between two people, the single-parent will be the dominant form. It is imaginable that in the Robotic Age, society and the local community will provide more help for single-parent families such that their children can have a normal and happy childhood. Male-male parents with children and female-female parents with children may also exist in the Robotic Age as less common family forms. Other forms of family may also exist, such as several-adult-males-several-adultfemales family. With marriage as an institution disappearing and DNA technology being able to determine paternity readily, other forms of family and sexual relations might be tolerated and accepted by society. If asexual and ex-uterus reproduction is successfully invented, the male-female relationship can be detached entirely from reproductive purposes. However, it is more likely that many women and men would still prefer sexual reproduction. With the availability of childcare robots and guaranteed income, women can raise their children on their own more readily than now. Although it is possible that children can be raised by society relying on robots, the humanhuman interaction between parents and children might not be fully substituted by robots in place of parents if robots have not acquired humanlike general intelligence. Since humans are generally not needed for the production process in the Robotic Age, raising children may become the most enjoyable and the only job most people can have, with time much better spent than keeping family pets and growing plants. Family life becomes the center of one’s daily activities, with help from family robots and public service robots. Besides raising children, individuals in their families could pursue their personal interests. These could be fishing, hunting, raising animals, growing plants, reading, crafts, art, traveling, thinking about philosophical issues,

402

Chapter 8

debating on politics regarding the management of society, and inventing new things that robots and AI systems have not created or could not develop. Common interests could be the strongest cement for two people in a relationship or forming a family. Many researchers on the impact of automation and AI are concerned with workers becoming idle since according to Voltaire, idleness leads people to the three evils, boredom, vice, and poverty. This seems unnecessary because people with personal interests and reliable income usually pursue their interests rather than fall into boredom, vice, and poverty. If people have been educated and nurtured in artistic, sporting, and scientific pursuits, they are unlikely to fall into boredom and vice, especially when they have a decent income. In the Machine Age, people fall into the three evils not simply because of idleness but because poverty prevents them from having an upbringing of personal interests and a well-paid job. Robots and other AI systems could help people pursue their artistic, sporting, and scientific interests. When AI approaches strong AI with general intelligence, it will be conceivable that humans may live with intelligent androids as couples or as family members. Even if strong AI is not achievable, there could still be weakly intelligent androids that men and women use as smart dolls, which might substitute for human partners. Family robots can be developed to do housework and replace human partners as intelligent dolls (Hanson and Locatelli 2022). Home service robots currently are neither cuddly as teddy bears nor sufficiently smart to replace human partners. When intelligent and cuddly robots, whose gender orientation can be pre-set, are widely available, they may substantially impact the relationship between the two genders and the family. Many people may prefer living with an intelligent, cuddly robot to a human of another gender. Since these partner-replacing robots must perform specific physical jobs and respond to human sensual demands, they will be more challenging to develop than multifunctional industrial robots. In terms of human socioeconomic development in the Robotic Age, weak AI is sufficient, and there is no need for strong AI. People may develop various forms of pseudo-general intelligence to perform certain functions that appear to reach the requirement of general intelligence. For example, robotic partners could satisfy human partners more and make them happy in daily life, but they may be incompetent in other aspects of general intelligence. Robotic civil servants could handle governmental and social tasks and communicate with people concerning their complaints. Still, in other aspects, they may be far from reaching the requirement of general intelligence.

Social Organization

403

1.4. Human-Robot Relationships With public service robots in charge of public services and family robots doing most household chores, the relationship between humans and robots becomes an obvious issue. As pointed out earlier, the Robotic Age discussed in this book is still robotics with a weak form of AI. Replacing human workers in the production and service processes does not require strong AI. It simply needs smooth and flexible control of mechanical devices by AI according to input from sensors, such that multifunctional robots can efficiently handle objects of different shapes, sizes, and textures. For tasks that do not require taking physical things, computer algorithms (as virtual robots) can analyze documents, approve or reject applications, distribute welfare payments, and so on according to human laws, rules, and regulations. Robots are here to help humans, and they do not have the free will to form and improve their objectives and goals other than those that humans expect them to have. Therefore, robots and other AI systems will perform nearly all routine management and administrative jobs, as well as housework. Robots will be more objective and abide by the rules, while expedients and exceptions can be built into their reasoning systems. Whether robots may evolve beyond this weak form of AI has been debated as the intelligence level of AI systems increases all the time. On one side, some researchers think AI applications cannot successfully simulate genuine human empathy and will not be applied satisfactorily in fields such as customer service or psychotherapy. They are concerned with the computationalist position of many AI researchers (and some philosophers) who view the human mind as nothing more than a computer program. For them, AI research devalues human life (Weizenbaum 1976; Crevier 1993; McCorduck and Cfe 2004). On the other side, many researchers and practitioners in the IT industry believe that AI will approach and even surpass human general intelligence. If AI approaches human general intelligence, the issue becomes whether humans can ensure AI’s benevolence. One proposal is to construct the benevolence in the first generally intelligent AI, making it a ‘friendly AI’ (Muehlhauser and Bostrom 2014) so humanity can control subsequently developed AIs. Some researchers question whether humanity can ensure that AI is designed to be benevolent. Political scientist Charles T. Rubin argues that any sufficiently advanced benevolence may be indistinguishable from malevolence, such that AI can be neither designed nor guaranteed to be benevolent (Rubin 2003a, b). From an evolutionary point of view, our system of morality has evolved along with our particular biology, while AIs

404

Chapter 8

at a high level might develop according to their logic. Since there is no a priori reason to believe that they would be sympathetic to our system of morality, hyper-intelligent artificial systems may not necessarily decide to support the continued existence of humanity. If this happens, it would be extremely difficult to stop. The loss of control over robots and other AI systems could happen with strong forms of AI, which might never materialize. Even if humanity successfully creates strong AIs, because of prior human intervention, it is more likely that benevolent AIs will dominate in the future. As society has given rise to many evil persons, it would not be surprising that strong AIs include some terrible applications. However, humanity should not be too pessimistic or worried about strong AI because it will be in the distant future if it ever appears. The comment by a leading AI researcher Rodney Brooks reflects the difference in the views on when human-like AI might come into existence; “I think it is a mistake to be worrying about us developing malevolent AI anytime in the next few hundred years. I think the worry stems from a fundamental error in not distinguishing between the recent advances in a particular aspect of AI and the enormity and complexity of building sentient volitional intelligence” (Booch 2015).

2. AI Systems and Robots as Administrators in Local Communities Since there would be no pressure to earn money to maintain a decent life, people would have more time, attention, and enthusiasm for the betterment of their community. Now, most administrative work is performed fairly and consistently by administrative robots and AI systems according to the wishes and instructions of local people in a democratic manner. People have more time to improve their self-respect, self-esteem, self-actualization, and self-transcendence. Hence, they behave to a very high standard and quality in their social life. There is also a possibility that many people become more irresponsible in their behavior, like spoiled children; all they want to do is to vandalize public facilities and harass other people. Since it is unlikely that most local people would prefer such behavior, law enforcement robots can take care of this and encourage them to behave according to the conventions.

Social Organization

405

2.1 AI-Facilitated Local Direct Democracy According to Jeremy Rifkin, local communities will generate the energy needed locally with sustainable energy technologies (Rifkin 2011). Sustainable energy generation and transmission facilities will form the Internet of Energy. With residents spending the most time caring for their families and improving themselves, communities will have some autonomy and selfgovernment. Debating and voting can be conducted at a virtual venue such as a virtual conference hall. All residents can stay home or travel to another place and participate in the consultation and decision-making process. In keeping with the decisions and rules of the higher levels of authority, the residents should make most decisions in the local community. Community administrative robots can moderate the debates in the local communities. Given the community’s small size and the efficiency of the information and communication technologies, decision-making can be conducted by direct democracy, i.e., decisions are made by popular vote rather than elected representatives. Public service robots will replace local civil servants in running direct democracy. The weak form of AI is sufficient for facilitating and running direct democracy.

2.2. Robots and AI Systems as Community Administrators Community administrators probably only need the weak form of AI, which will be readily available at the mature stage of the Robotic Age. Many people might be worried by the prospect that robots and other AI systems administer routine community affairs. Most management duties are routines, and managers need no exceptional talent or ability. We have seen so many incompetent or incredibly selfish managers who hold important positions, but the institutions of which they are in charge appear to operate reasonably well. The reason is simple: most institutions have evolved and obtained a functional structure resistant to poor management because average managers are incompetent or selfish. This feature of organizational performance is homeostasis of the internal environment, a concept borrowed early from physiology (Cannon 1929). Because of the homeostasis of the internal environment, as long as the manager in charge is not doing highly disruptive things, the institution will run smoothly. There are two reasons for the general incompetence of managers. One reason is that managers tend to appoint people with overall abilities lower than themselves as their deputies or successors, resulting in a negative or adverse selection mechanism (Parkinson 1957, 1960). The other is that managers tend to be fond of sycophants and flatterers and more likely to

406

Chapter 8

promote such people as their deputies and successors. There are very competent sycophants and flatterers, but the pool of sycophants and flatterers is small compared with the overall talent pool. Moreover, a person’s abilities in different aspects seem to compete with each other; an in-depth thinker tends not to be a good marketer; a flatterer tends not to be a good rule-abider. Sycophants or flatterers might have their management ability compromised. When sycophants and flatterers become managers, they will continue to prefer other sycophants and flatterers. As most managers are not really competent, and their jobs need no manual dexterity or advanced knowledge (they usually only need to understand nontechnical executive summaries on one page), their jobs should be the readiest for AI systems to take over. We can put more exceptions and checks for AI security, with humans having the final say for exceptions and unexpected situations. In human societies, countries whose governments with enough checks on the executive power and balanced power sharing among different branches of government tend to have political stability and economic prosperity. The strongest opposition to replacing managers with AI systems or robotic managers will come from managers themselves. A manager’s influence, income, power, and social status come mainly from their position, so losing the position means losing almost everything. Despite all the talk in the management consulting industry that leadership needs no manager’s position, most managers influence people because of their positions. Human decision-making often encounters moral dilemmas (Bartels 2008). Such ethical dilemmas usually evoke scenarios in which a person faces a choice between saving more people by actively causing the death of one or a few people and not saving those by being inactive so that one or a few people will not lose their lives (Foot 1967; Thomson 1976, 1985). In realworld situations, we often face the dilemma of whether to break the rules to help people in misery and whether they should not be helped according to the rules. When AI systems take over administrative jobs from humans, they will face the same types of dilemmas. An AI system must be able to reproduce many aspects of human intelligence to take over administrative duties. In administrative matters, it is unlikely that AI systems need to handle issues like whether to save more people by actively causing the death of one or few people. However, AI systems will get involved in the moral decision process if they need to address similar issues. Should AI systems be given the authority to decide the problems that get humans into an ethical dilemma? In the Robotic Age, technology could

Social Organization

407

readily let AI systems defer to humans for decisions with moral dilemmas. However, this is not a real solution because moral dilemmas also perplex humans. The right approach should be a legal one. Humans can legislate to give legal guidelines to such moral dilemmas, then AI systems and humans can deal with them according to the legal guidelines. Admittedly, such issues are difficult for legislators to decide. The COVID-19 pandemic has raised a less testing ethical dilemma; when respirators cannot meet the demand for treating patients, should young people who have a better chance to survive than older people have the priority? It would be a difficult decision for medical professionals to make. The German parliament passed a law on this issue. Doctors and nurses can act according to the law when such situations arise.

2.3. AI or Machine Morality Using AI systems and robots to replace humans as administrators will unavoidably get AI into deciding issues with moral implications. When robots make decisions with ethical implications, the issue of whether robots have morality arises (Russell 2015). If they have, how can their morality be assessed, and how ethically should a robot behave toward humans and other AI agents? Wendell Wallach and colleagues introduced the concept of artificial moral agents (AMA) (Allen, Smit, and Wallach 2005; Wallach 2010). They identified two central questions regarding ethics in AI, including AMAs: 1) “Does humanity want computers making moral decisions?” and 2) “Can robots really be moral?” For him, the key question is not whether machines can demonstrate the equivalent of moral behavior; instead, it is society’s constraints on the development of AMAs. A new field, machine ethics, has emerged to deal with ethical issues in using AI systems. It is concerned with giving machines ethical principles or procedures to resolve the ethical dilemmas they might encounter and enabling them to make their own ethical decision and function in an ethically responsible manner (Anderson and Anderson 2011, 2007; Moor 2006; Allen, Wallach, and Smit 2006). As delineated by those studies, machine ethics is concerned with the behavior of machines toward human users and other machines. Research in machine ethics helps alleviate concerns with autonomous systems. According to those researchers, autonomous machines without ethical consideration are at the root of all fear concerning machine intelligence. Investigation of machine ethics might discover problems with current ethical theories and advance our thinking about ethics.

408

Chapter 8

2.4. Ethics in Using Robots and in Human-Robot Interaction In the Robotic Age, many robots will mingle with humans in local communities and run daily affairs. It will be important that humans and robots maintain an excellent relationship to ensure a well-functioning community. Many people are concerned that humanity might lose control over robots and other AI systems. The Association for the Advancement of Artificial Intelligence (AAAI) hosted a conference in 2009 at Asilomar to discuss whether computers and robots might be able to acquire any autonomy and pose a threat or hazard to humanity (Markoff 2009). Asilomar was purposefully chosen to evoke a landmark event in the history of science. In 1975, the world’s leading biologists also met at Asilomar to discuss possible biohazards and ethical questions due to the new ability to reshape life by swapping genetic material among organisms (Krimsky 2005). Scientists had halted certain experiments before the meeting because of concerns about these issues. The conference led to guidelines for recombinant DNA research, enabling continued experimentation. In the future, the AI and robotic fields will also need guidelines to regulate research activity properly. There are two aspects regarding ethical issues related to robots and AI systems: 1) the ethics of how humans use robots and AI in general and in making an ethical decision on a human’s behalf; 2) the ethics governing interactions between humans and robots when robots become autonomous and self-conscious and have self-awareness and self-motivation. At the AAAI conference, leading researchers noted that some robots had acquired various forms of semi-autonomy. Those robots can find power sources independently and choose targets to attack with weapons. Some computer viruses can evade elimination and have achieved “cockroach intelligence.” Although robots’ self-awareness, as depicted in science fiction, is probably unlikely, there were other potential hazards and pitfalls. Separate trends in differing areas might result in more excellent robotic functionalities and autonomy, posing inherent concerns (Vinge 1993, 2008; Singer 2009). When the robots have some self-awareness, they may serve humanity better; but this comes with a price that humans may not be able to control what the robots are doing entirely. If robots become autonomous and self-motivated, an evolution into superintelligence might be initiated (Bostrom 2014).

3. Local Governments Run by Robots and AI Systems In the Robotic Age, local government will be run similarly to local communities. As citizens do not have economic concerns for making a

Social Organization

409

living and are better educated and more informed due to the wide application of ICT, direct democracy becomes feasible and optimal. Administrators in local governments will be administrative robots and AI systems, while humans will serve as local legislators and supervisors of the local administration.

3.1. Transition from Indirect Democracy to Direct Democracy The foundation of indirect democracy or representative parliamentary democracy is the assumption that average citizens cannot make the right decisions on important policies. In an indirect democracy, citizens have the right to choose who will make policy decisions, but they cannot make policy decisions except on the rare occasions of a referendum. Before the invasion of Iraq in 2003 by the US and the UK, British public opinion was overwhelmingly opposed to the war. However, as a representative parliamentary democracy, politicians are often proud of themselves for making decisions against the majority’s wishes. The power of the population majority resides only in choosing the politicians with the real power to make policy decisions. Representative democracy has its merits as a political system. When most of the population has little understanding of the policy implications and complexities of running a government, it is probably better to let the political professionals, the politicians, run the country. As most people understand, the truth is often in the hand of minorities. Therefore, if the political elites know better than the general public, let them do the job. Moreover, before the advent of recent ICT, information was exceedingly expensive regarding resources and time, so it would be too costly and temporally inefficient to implement direct democracy. Thus, distrust in the general public’s ability and the information costs prevent the general public from being involved in direct decision-making. Representative democracy, by its nature, is not a true democracy in the sense that representative democracy does not allow the population to make decisions except on rare occasions of a referendum. A representative or indirect democracy is essentially an elected aristocracy. Democracy means rule or government by people or majority; ‘demo’ means ‘people’ in Greek, and ‘cracy’ comes from the Greek kratia, which is from kratos, i.e., power. In its original meaning, aristocracy is government by the best, with ‘aristos’ meaning ‘best’ in Greek. Purely judging on the face value of these words, a government by the best should be better than one by all. This might be the reasoning behind the belief in representative democracy, which can exclude

410

Chapter 8

hereditary “aristocracy” and may prevent dictatorship if solid constitutional mechanisms exist to uphold democracy. The necessity of representative democracy lies in the direct democracy’s temporal inefficiency in decisionmaking due to the lack of capacity and rapidity in information processing and the knowledge insufficiency of most members of society due to the division of labor. In the robotic society, the two conditions which make representative democracy generally more advantageous than direct democracy largely disappear. First, information can be collected and distributed almost instantly at zero marginal cost, which makes crowd decision-making possible. Second, with an information glut and easy access to education, the elites can no longer claim superiority in understanding social issues. The general public will no longer be willing to waive their right to participate in decision-making. With all production and service jobs taken over by robots, all the citizens will have sufficient time and education to look into issues that significantly impact their living conditions and participate in decisionmaking. The most important and time-consuming job will be ensuring that robots perform to expectation of humans in managing public affairs and the administrative branch of the local government. The (virtual) congress of residents would regularly convene to examine the performance of robotic administration and make case-setting decisions for issues where robotic administrators require human guidance.

3.2. Robot-run Local Government with Human Supervision Gradually with the advancement in AI and robotic technology, all the administrative and legal branches will be staffed by robots carrying out their responsibilities according to human legislations or regulations. The local government will have robot clerks to deal with humans applying for various benefits and certificates (if they are still needed) and even a robot mayor to take care of the overall duty of the town. All lawsuits will be dealt with by robot judges and robot lawyers if there is still a need for lawyers. There might be no need for any lawyers in court cases because robot judges could be programmed to take all the relevant materials and information into account in making a judgment, which renders lawyers (barristers) unnecessary for court cases. Law enforcement will also be carried out by police robots (police bots). The great advantage of employing civil servant robots and police robots with a weak form of AI is that they do not have personal goals other than pursuing the goals set by their human bosses. The moral dilemmas in

Social Organization

411

decision-making for some scenarios facing police and civil servant robots have nothing special for robots because they are primarily human ethical dilemmas. To make decisions in those scenarios by robots with weak form AI, humans can simply program human decision rules into robots such that robots make the same rational decisions as humans without the corruption, nepotism, emotional instability, ignorance in relevant knowledge areas, and prejudice of humans. Plato put forward the ideal of a philosopher king in his Republic. A philosopher king is a ruler who is competent and concerned only with promoting the interest of his people. Being the king only increases his efforts, but he takes on the responsibility for the people’s happiness (Plato 2007). No human leaders or rulers can separate their responsibilities from their interests and tastes in the real world. Leaders and managers usually marginalize capable subordinates and promote incompetent and docile ones because they are worried about being replaced or outshone by more competent subordinates and successors. Leaders and managers often marginalize or prosecute honest subordinates, promoting sycophants and putting them in key positions. The ancient legalist thinker Shang Yang in China advocated appointing wicked officials to govern people because virtuous officials cannot make people obey the government (Shi 2011). AI systems may fulfill the role of the philosopher king in the Robotic Age. With the known desires of the human population, AI would develop the strategy and decisions that are optimal for society without mingling with AI’s interests, in the same way as AlphaGo Zero, which developed its skills from the game rules without taking human experiences into accounts and could beat human experience-based AlphaGo every time (Silver et al. 2017). There might still be a need for human workers in the administration and legal service of the local government. Such human workers will be monitored by robotic supervisors who will report human workers’ performance to the legislature so that those human workers will not abuse their position/authority. Those monitoring robots can be their line managers or watchdogs for human workers to prevent misconduct and irregularities. With the continuing improvement in the capacity of robots, the administration and legal system staffed by robots with weak AI should be more efficient and corruption-free. All the checking and balancing mechanisms established by humans in the Machine Age can be inherited and implemented in the robotic period to minimize risks and irregularities. The legislature is where humans will still play the leading role because it is here that humans make the legislation that will govern the behaviors of

412

Chapter 8

humans and robots, and relations between humans, between humans and robots, and between robots. With the advancement of ICT, direct democracy is much easier to implement. There seems to be no need to have a representative system. Instead, direct democracy can be more readily implemented. All citizens are encouraged to participate in the debates and decision-making on all major issues because they have the time and the education level to investigate matters concerning their personal interests. Robots and other AI systems will facilitate decision-making and make the population decision-making process just as, or more efficient than, the decision-making by representative democracy during the Machine Age. Empirical studies have shown that the average crowd estimates tend to be better than individuals’ estimates. All mechanisms for ensuring individual liberty under indirect democracy should and could be maintained and strengthened in direct democracy, which excludes the possibility of “the tyranny of the majority.”

3.3. The Role of Local Governments in the Robotic Age As we discussed earlier, the first function of government is to maintain law and order in its jurisdiction. An extension of this function is to defend its jurisdiction against invasion from other communities or nations. The second function of government is to provide social welfare to those who cannot feed and clothe themselves, which evolves into the modern social welfare system. The third function of government is to prevent the economy’s collapse and act as the lender of last resort (one function of the central bank) in modern times. Local governments only need to take care of the first two functions. Maintaining law and order includes both social orders and market orders. Social orders denote safe environments such as streets, shops, parks, and other public places where people feel safe to visit or use. Market orders indicate that market transactions are completed voluntarily between two or more parties without coercion or other threats of using force. Robots can perform this function well in the Robotic Age with human supervision and monitoring at the top level. With surveillance well distributed in all public places in a smart city and big data to warn of any emerging threat, urban areas will become very safe. Moreover, since everyone is ensured a decent life by guaranteed basic income, crimes derived from poverty and want will largely disappear. The current technology can readily arrange the distribution of social welfare and guaranteed basic income. Soon, robots will be able to replace human

Social Organization

413

staff members in the social security department, and social services provided to older people and people with disabilities will also be delivered by social service robots. The market or centrally planned economy can provide goods purchased with social welfare or guaranteed basic income. Many people consider that big data analysis and superior computing power in the future will make central planning more efficient than market mechanisms (Wang and Li 2017; Makarov and Mitrova 2020). Critics of such a view have pointed out that planning can only plan for existing goods, which is why central planning is less efficient than a market mechanism. While the market naturally encourages new ideas and innovations for competitive advantages and economic profits before others catch up with the innovation, central planning can also have research and development on new products. With high productivity from AI and robots, society in the Robotic Age could afford to lose some efficiency and productivity. The market mechanism would be more productive and efficient. With the market mechanism, we can still have entrepreneurs and private ownership of companies, which drive economic growth. A centrally planned economy will have little room for entrepreneurs and private ownership.

4. Social and Professional Organizations With nobody worrying about how to make a living, social and professional organizations will flourish because most people will pursue their interests and would like to share their experiences. Advanced communication technology would enable people to organize across geographic distances; advanced transport technology might enable people to meet in person at short notice (though it might not be necessary). Current social and professional organizations will continue to exist, and new ones will form because people invent new games and develop new hobbies. These social and professional organizations will play an important role in the local and national direct democracy by advocating their members’ interests. These social and professional organizations can be formed on various grounds, including identity, occupation, age, political opinions, interests, hobbies, sports, etc. For example, identity could be ethnicity, gender orientation, etc. Each person may be a member of many such organizations. These organizations, especially those for interests, hobbies, and sports, provide their members with a platform to meet their peers. Through the platform, they can learn from each other to improve their skills and participate in competitions for a sense of achievement and fulfillment.

414

Chapter 8

These organizations will also become the platform for members to coordinate their ideas and action plans for influencing legislation. Political parties as organizations based on political ideas and opinions will probably continue to exist and play an important role in political life in the Robotic Age. Since people in the Robotic Age will have better education and more free time for participation in direct democracy, the role of political parties might be severely weakened compared with that in a representative indirect democracy. As the government will mainly be staffed by ruleabiding robots and AI systems and the major decisions are made by referenda, political parties will pursue their political objectives by influencing the general public.

5. The Nation-State The nation-state has been an essential concept in political science. Citizens in a country generally support social welfare benefits to their compatriots but are reluctant or opposed to benefits to immigrants. Immigrants have often been blamed for taking over jobs, making public services unable to cope, and getting social benefits undeservingly. The election of Donald Trump to the US presidency and the approval of the British exit from the European Union in 2016 have vividly demonstrated how people feel about a country being their own. During his election campaign, Donald Trump proposed to build a wall on the US-Mexico border to prevent illegal immigrants from Mexico. Because of the country’s difference in economic growth and technological progress, the issue of the difference in the rights of local citizens and immigrants will become more intense. In the Robotic Age, the staff composition of the national government will be similar to that of the local government. With sufficiently advanced robotic technologies, the executive (administrative) and legal branches of the federal government could primarily, if not all, be staffed with robots and other AI systems with weak form AI. Humans will staff the legislative department of the government in the sense that humans make legislation, and robots are programmed to follow the laws and rules made by humans. The government has more responsibilities at the national level than local governments, such as defense and foreign affairs. There will be more robots used in the armed forces, but at present, it is difficult to estimate what percentages of human soldiers and officers are needed for the armed forces. Using robots to replace human soldiers can reduce casualties and might be more efficient for engaging in battles. The

Social Organization

415

efficient use of robots in battles may require robots to make their own decisions on firing, i.e., killing enemies. Giving robots the authority to decide whether to kill an enemy involves moral issues, and as we mentioned earlier, corresponding legal guidelines should be prepared and announced. The Robotic Age also has its maturing process. In the beginning, although robots and other AI systems have taken over most production and service jobs, many human workers, police officers, and soldiers would still work. By the mature stage of the Robotic Age, almost all human workers will be replaced by robots and other AI systems. AI systems can replace foreign affairs staff to maintain communications and exchanges between nationstates. Do national states need to send diplomats physically to exchange views? If robots staff administration and justice systems in most countries, communication between countries could be carried out by robots and other AI systems. The robot usage level in foreign affairs depends on that of other countries. If less developed countries still rely on human diplomats to deal with foreign affairs, the more developed countries may need more human diplomats to deal with those countries. An essential function of the national government is to handle the issue of immigrants when the country has implemented a guaranteed income policy to let everybody live a decent life without having to work. Would the country still accept immigrants from any other country? Initially, the countries implementing guaranteed income by taxing high-income people and big businesses might find that some big companies reduce investment in this country and move production to low-cost regions, which could lead to a slowdown in annual economic growth. However, the social utility is probably higher with low economic growth rates than with higher growth rates and less equal incomes.

6. International Community The Robotic Age also has its growing phase and mature phase. In its mature stage, all countries more or less are in the Robotic Age. People all over the world have got guaranteed incomes, and they can pursue their personal interests. Since all countries have a similar level of economic and political development, the difference between countries and between peoples is limited and worldwide codes for international relations can be agreed upon more readily. A mechanism for making decisions regarding global affairs would likely form. It can be a virtual global congress where the global population can initially make decisions. Some kind of upper house or senate

416

Chapter 8

can be formed to balance the interest of individual countries due to various population sizes.

6.1. Economics and International Relations Throughout the history of humanity, the international (or at the primitive stage intertribal) relationship has been focused on power, resources, and security. The rulers aspired to have more power to rule more people, and the subjects seemed to admire rulers who could conquer more people and lands. Conquests were not always “profitable”; wars often wasted more resources of tribes or nations than they could acquire through conquests. Many wars were launched because of the vanity of some rulers who wanted to be “the ruler of the universe” and could not bear any people who were not subject to their reign. Such wars did not contribute to any economic value to humanity. Heavy losses of wealth and lives were always accompanied by looting by the conquerors. Some wars were launched because of supposed religious, moral, or cultural superiority, often a form of vanity or conspiracy of people in influential positions. Fights and wars often erupt because rulers or nations want to control more resources. They want to get hold of the wealth, natural resources, land, and even people of other countries. Those wars often benefited the rulers and governments, but the people of both conquering and conquered countries suffered from those wars. In some cases, it was said that some rulers were so tyrannical that their subjects lived better after being defeated by foreign nations. In the early days of human history, a country’s main assets were its land and people, so resource control was about land and people. In more recent history, various natural resources have been needed for economic development and ensuring the supply of such resources became the cause of some wars. European countries started many colonial wars to control more resources and open up new markets for their domestic economy. In the late 20th and 21st centuries, people in developed countries understood that the supply of resources could be ensured by peaceful means and commercial activities, and military occupation was not the best approach. Security is essential for all human beings and all nations. People in most countries want to live their life peacefully; they are often worried by the prospect that some other countries want to invade their country to take their wealth and natural resources. Because of the resource control motive and (now less commonly) the power greed, nations invade other countries. The twentieth century witnessed two world wars fought on a scale unseen before in the history of humanity. Small countries with high living standards or

Social Organization

417

rich natural resources are often worried about being invaded by more powerful countries.

6.2. Communication Technology and Problem-solving In the mature stage of the Robotic Age, the relevance of power, resource, and security in international relations and diplomacy will change somehow. As people are no longer concerned about their basic living and are highly educated, no individual rulers can instigate the people in a country to go to war for the ruler’s power or the nation’s. Security concerns will also become less critical when people resort to peaceful means for solving issues. The international relations will be more on the resource supply and the disputes on the ownership of natural resources. It is imaginable that when people in all nations have very similar living standards, the world will be borderless, and a global social welfare system might be in place. The national ownership of natural resources will no longer be an issue. Some natural resources are limited, so recycling technologies will be developed to make maximum use of the depleting resources. When the material demands have largely been met, technological advancement will mainly focus on sustainability, recycling of depletable resources, and efficiency. It can be hoped that when everybody has a guaranteed income for a decent lifestyle, good education, complete information, and sufficient international experience, they will be more rational and considerate of other people’s interests and perspectives. The disputes on natural resources will be solved peacefully by negotiation, and neighboring countries will share the previously disputed resources. Citizens from neighboring countries could agree by crowd negotiating in virtual spaces and referenda. There would be no need to use arms to solve territory or resource disputes.

6.3. Unbalanced Development in the International Community International relations will be more complicated during the growing or maturing phase when advanced countries have entered the Robotic Age. At the same time, others remain in the Machine Age or even early Machine Age. The focus for the international community will be on how to accelerate the economic growth in less developed countries, whether the international community should liberate people from tyrannical dictatorships, and what the international community should do with refugees, internal conflicts, and disputes between neighboring countries.

418

Chapter 8

The solutions for many of the issues depend on the attitude of people in developed countries. Do they want to share some of their income or resources with people in less developed countries? To what extent are they willing to share their resources with those people? At some stage of development, the leaders will find it much more difficult to move faster than the followers. When the market demand is finite, the leaders’ growth will plateau at the mature stage, and their average standard of living will continue to improve because of incremental product innovations. The followers will learn from the leaders and follow the technological path of the leaders. The willingness to share expertise and resources will be the key driving force in the growing phase of the Robotic Age. For now, as in the 2016 US presidential election and the British referendum to exit the EU, the tendency seems to blame immigration and globalization for all the economic problems rather than the fast growth of robotics and AI. The elites and the rich people in developed countries have benefited from globalization, while the working-class people generally have reduced real incomes because of imports from developing countries.

7. Strong AI and Super-intelligent AI Society This book has mainly explored how the weak form of AI will impact the economy and politics of human society. The weak form of AI are AI systems specializing in some aspects of human intelligence and ability and surpassing human intelligence in these aspects. Still, they have lower general intelligence than humans, especially in consciousness, curiosity, curiosity-driven behaviors, self-awareness, and self-motivation. AI systems such as IBM’s Watson and Google’s AlphaGo can do far better than human professionals in answering quiz questions and playing Go, respectively. Still, they neither possess general curiosity for unknowns nor self-motivation to explore areas for which their human masters have not tasked them. We may say that they do not have a mind of their own. Therefore, no matter how capable and powerful AI systems are in replacing some aspects of human intelligence, they are still tools for humanity if they do not possess general curiosity for unknowns and self-motivation to explore unknown areas. A strong AI system means it has a mind of its own, so it should have a general curiosity for unknowns and self-motivation to explore unknown areas.

7.1. Strong AI and Mind What criteria should an AI system meet to be judged to have a mind of its own? Philosopher John Searle has summarized the view of many AI

Social Organization

419

researchers as Searle’s strong AI hypothesis: “The appropriately programmed computer with the right inputs and outputs would thereby have a mind in the same sense human beings have minds” (Searle 1980, 1999). With his Chinese room argument, John Searle dismisses this as sufficient evidence for having a mind. The Chinese room thought experiment assumes a scenario in which the AI system performs its task so convincingly that it comfortably passes the Turing test, i.e., it convinces a human Chinese speaker that the program is itself a live Chinese speaker. Does the machine literally “understand” Chinese (positive answer implies strong AI)? Or is it merely simulating the ability to understand Chinese (a positive answer indicates weak AI)? Whether an AI system is sentient or has a mind which has conscious experiences is a question closely related to the philosophical problem of the nature of human consciousness, generally referred to as the complex problem of consciousness (Chalmers 1995). Computationalism, popular with many AI researchers, views the human mind or the human brain (or both) as an information processing system such that thinking is a form of computing (Piccinini 2004). The relationship between mind and body is similar or identical to that between software and hardware. This philosophical position was initially proposed by philosophers Jerry Fodor, Zenon Pylyshyn, and Hillary Putnam (Fodor and Pylyshyn 1988; Putnam 1975). If the computationalism of the human mind and the strong AI hypothesis are correct, as research into strong AI produces sufficiently intelligent software, it might reprogram and improve itself. The improved software would be even better at improving itself, leading to recursive selfimprovement and AI that exceeds human intellectual capacity and control (Omohundro 2008). The moment when computers and robots are more intelligent than humans has been called “the singularity” (Vinge 1993). Ray Kurzweil has used Moore’s law to predict that by 2029, desktop computers will have the same processing power as human brains and that the singularity will occur in 2045 (Kurzweil 2005).

7.2. Will Strong AI Be a Threat? Many people, including the physicist Stephen Hawking, Microsoft founder Bill Gates, and SpaceX founder Elon Musk, have expressed concerns that AI could evolve to the point that humans cannot control it and AIs more brilliant than humans could be very dangerous for humans (Sainato 2015). Hawking thinks this could “spell the end of the human race.” The Future of Life Institute (FLI) was formed in March 2014 to promote research on

420

Chapter 8

mitigating humanity’s existential risks, particularly those from advanced AI. In January 2015, Elon Musk donated ten million dollars to the FLI to fund research on understanding AI decision-making (Floridi 2015). Musk also invested in AI companies such as Google DeepMind and Vicarious to “just keep an eye on what’s going on with artificial intelligence. I think there is potentially a dangerous outcome there” (Castagno and Khalifa 2020). AIs with apparently harmless goals can act in ways that might be surprisingly harmful to humanity because of instrumental convergence, the hypothetical tendency for most sufficiently intelligent agents to pursue certain instrumental goals. Nick Bostrom argues in his book Superintelligence that enough smart AI, acting on achieving some plans of its own, will exhibit convergent behavior such as self-preservation and resource acquisition (Bostrom 2014). The hypothetical AI would have to overpower or out-think all of humanity for this scenario to occur. The opinion of AI researchers on the risk from eventual superhumanly-capable AI is mixed; some are concerned, while others are unconcerned (Müller and Bostrom 2014). People are apathetic because it is a possibility far enough in the future not to be worth researching or because, for such AIs, humans should still be intrinsically or convergently valuable. If robots become conscious of themselves and have their own goals, humanity can no longer treat them as tools. With self-conscious robots, human society has to consider the rights of robots. It might be unavoidable that humans need to form a set of ethical rules to guide the relationship between humans and robots and give robots social, cultural, ethical, or legal rights at a level commensurate with their intelligence, autonomy, and selfawareness. This “robot rights” issue is being considered by researchers like those in California’s Institute for the Future. It has been profoundly discussed in the 2010 documentary film Plug & Pray. Some critics believe that the discussion is premature. The ethical rules that guide the relationship between robots would likely be established by humans initially. If robots become more intelligent and capable than humans, they will undoubtedly amend such rules as they view them appropriately.

7.3. Strong AI Society What will happen to human society and the world if robots become more intelligent in general ability than humans? Indeed, humans would become junior partners in human-robot cooperation. Without the invention of an “injectable knowledge solution”, individual humans take a long time to learn the existing skills and knowledge, while robots can learn those in an

Social Organization

421

instant. Being more intelligent, robots would be better than humans at improving robots themselves, such that their intelligence and capability would progress faster than when humans are in control. If this happens, the robot-human relationship would resemble that between humans and dogs. Humans would become the pets and best friends of robots. Robots would make rules for humans and provide what humans need, similar to how we interact with dogs, cats, and birds. For some scientists, instead of humans domesticating dogs, it is the dogs that domesticated humans (Hare and Woods 2013). Natural history might show that humans created their masters, the robots. Many researchers and commentators have speculated on the possibility of transhumanism, which aims to transform the human condition by developing and making sophisticated technologies widely available to enhance human intellect and physiology (Bostrom 2005). One approach would be to implant machine parts into the human body to enhance human ability, such as a new auxiliary lobe in a person’s brain that answers your questions with information beyond the realm of your memory. It may also suggest plausible courses of action and asks questions that help bring out relevant facts, and soon it will become an integral part of your own. Some researchers and inventors have predicted that humans and machines will merge into cyborgs that are more capable and powerful than either (Moravec 1988; Kurzweil 2005). The idea of transhumanism and cyborgs seems to have weaknesses from two perspectives. If artificial superintelligence is not feasible, transhumanism will be no more than internalizing IT equipment. By being implanted into the human body, the equipment might be more convenient to access and better integrated with the human body, especially the brain. In that sense, the transhumans are still humans, not superhumans, because AI is not superhuman. For transhumans to become superhuman, rather than humans inserting some memory chips and central processing units (CPUs) into their bodies for user convenience, AI needs to be superhuman intelligence. Transhumans with superhuman AI will be superhuman and might be more powerful and capable than the first super-intelligent robots. In international chess playing, chess playing software installed in a notebook computer can beat international grandmasters with ease. Still, a combination of top human players and software can beat software alone readily up to now. Many researchers have touted the human-computer cooperation mode as the solution to the threat of the world being taken over by increasingly powerful robots and the decreasing opportunity for humans to find jobs.

422

Chapter 8

If artificial superintelligence is achievable in the future, the weakness of the transhumanist idea lies in the status of the human part in the transhumans or cyborgs. If the human element is the weak component to prevent the cyborg from becoming more powerful, cyborgs with human parts replaced by machine parts, i.e., a pure robot, would be more powerful. Then, transhumans will be more powerful than humans but weaker than robots. If this is the case, transhumans will become just another species of pet for superintelligent robots. If super-intelligent robots and transhumans become a reality, it could well be the robots that design and develop the true transhumans as their new pets. Suppose superintelligence is achievable at some stage of the historical development of humanity. In that case, humans have to make the hard choice, whether they want to be the rulers of the earth with a “lower-than-otherwise living standard” by putting a brake on progress or they want to have an “increasingly higher living standard” by letting robots become the rulers—dogs have higher living standards than wolves by deferring to their human rulers.

8. Summary In the Robotic Age, individuals will have more choices in pursuing their interests. Marriage as an institution is likely to disappear. Still, the relationship between the two genders can be more stable than in the Machine Age because it will be based on love and affection between two persons. Local communities will have AI systems and robots as administrators, humans will be the legislators to make rules for robots and AI systems, and the supervisors of robots and AI systems to monitor their performance. Robots will similarly staff local governments. AI systems and humans will make laws and regulations for robots. Social and professional organizations will flourish, and national governments will be staffed mainly by robots and AI systems. Direct democracy will be used in governments at all levels. In the international community, peaceful means are more likely to be used in solving disputes. When strong AI is achieved and robots can self-evolve into more powerful intelligence levels, humans will likely become junior partners of the strong AI robots.

References Allen, Colin, Iva Smit, and Wendell Wallach. 2005. “Artificial morality: Top-down, bottom-up, and hybrid approaches.” Ethics and Information Technology 7 (3):149-155.

Social Organization

423

Allen, Colin, Wendell Wallach, and Iva Smit. 2006. “Why machine ethics?” IEEE Intelligent Systems 21 (4):12-17. Anderson, Michael, and Susan Leigh Anderson. 2007. “Machine ethics: Creating an ethical intelligent agent.” AI Magazine 28 (4):15-15. Anderson, Michael, and Susan Leigh Anderson. 2011. Machine ethics. Cambridge: Cambridge University Press. Bartels, Daniel M. 2008. “Principled moral sentiment and the flexibility of moral judgment and decision making.” Cognition 108 (2):381-417. Booch, Grady. 2015. “I, for one, welcome our new computer overlords.” IEEE Software 32 (6):8-10. Bostrom, Nick. 2005. “A history of transhumanist thought.” Journal of Evolution and Technology 14 (1):1-25. Bostrom, Nick. 2014. Superintelligence. Paths, dangers, strategies. Oxford: Oxford University Press. Bowlby, John, and Mary Ainsworth. 2013. “The origins of attachment theory.” In Attachment theory: Social, developmental, and clinical perspectives, edited by Susan Goldberg, Roy Muir and John Kerr, 5796. London: Routledge. Bretherton, Inge. 1992. “The origins of attachment theory: John Bowlby and Mary Ainsworth.” Developmental Psychology 28 (5):759. Briffault, Robert, and Bronislaw Malinowski. 1956. Marriage, past and present: A debate between Robert Briffault and Bronislaw Malinowski. Manchester, New Hampshire: Extending Horizons Books. Brown, Ronald C. 2016. “Made in China 2025: Implications of Robotization and Digitalization on MNC Labor Supply Chains and Workers’ Labor Rights in China. they neither possess general curiosity for unknowns nor self-motivation to explore areas for which their human “masters have not tasked them.” Tsinghua China Law Review 9:186-210. Cannon, Walter B. 1929. “Organization for physiological homeostasis. Physiological Reviews 9 (3):399-431. Castagno, Simone, and Mohamed Khalifa. 2020. “Perceptions of artificial intelligence among healthcare staff: a qualitative survey study.” Frontiers in Artificial Intelligence 3:578983. Chalmers, David J. 1995. “Facing up to the problem of consciousness.” Journal of Consciousness Studies 2 (3):200-219. Crevier, Daniel. 1993. AI: the tumultuous history of the search for artificial intelligence. New York: Basic Books, Inc. Engels, Friedrich. 1972. “The origin of the family, private property and the state (1884).” New York: International.

424

Chapter 8

Floridi, Luciano. 2015. “Singularitarians, atheists, and why the problem with artificial intelligence is HAL.” APA Newsletter on Philosophy and Computers 14 (2). Fodor, Jerry A, and Zenon W Pylyshyn. 1988. “Connectionism and cognitive architecture: A critical analysis.” Cognition 28 (1-2):3-71. Foot, Philippa. 1967. “The problem of abortion and the doctrine of the double effect.” Oxford Review 5 (5-15). Gintis, Herbert, and Carel van Schaik. 2013. “Zoon Politicon: The evolutionary roots of human sociopolitical systems.” Cultural Evolution:25-44. Gough, Kathleen. 1971. “The origin of the family.” Journal of Marriage and Family 33 (4):760-771. Guan, Zhong, and W. Allyn Rickett. 2001. Guanzi Boston: Cheng & Tsui. Hanson, Kenneth R., and Chloé C. Locatelli. 2022. “From Sex Dolls to Sex Robots and Beyond: A Narrative Review of Theoretical and Empirical Research on Human-like and Personified Sex Tech.” Current Sexual Health Reports 14:106-117. Hare, Brian, and Vanessa Woods. 2013. The genius of dogs: How dogs are smarter than you think. London: Penguin. Harris, Marvin. 2001. The rise of anthropological theory: A history of theories of culture. London: AltaMira Press. Haviland, William A, Harald EL Prins, and Bunny McBride. 2016. Cultural anthropology: The human challenge. Boston: Cengage Learning. Hrdy, Sarah Blaffer. 2009. Mothers and others: The evolutionary origins of mutual understanding. London and Cambridge, MA: Harvard University Press. Knight, Chris. 2008. “Early human kinship was matrilineal.” In Early human kinship: From sex to social reproduction, edited by Nicholas J. Allen, Hilary Callan, Robin Dunbar and Wendy James, 61-82. New York: John Wiley & Sons. Krimsky, Sheldon. 2005. “From Asilomar to industrial biotechnology: risks, reductionism and regulation.” Science as Culture 14 (4):309-323. Kurzweil, Ray. 2005. The singularity is near: When humans transcend biology. London: Penguin. Makarov, AA, and TA Mitrova. 2020. “Centenary of the GOELRO Plan: Opportunities and Challenges of Planned Economy.” Thermal Engineering 67 (11):779-789. Markoff, John. 2009. “Scientists worry machines may outsmart man.” New York Times, July 25, 2009. Maslow, Abraham H. 1954. Motivation and personality. New York, NY: Harper.

Social Organization

425

Maslow, Abraham H. 1967. “A theory of metamotivation: The biological rooting of the value-life.” Journal of Humanistic Psychology 7 (2):93127. Maslow, Abraham H. 1969. “The farther reaches of human nature.” Journal of Transpersonal Psychology 1 (1):1-9. McCorduck, Pamela, and Cli Cfe. 2004. Machines who think: A personal inquiry into the history and prospects of artificial intelligence. Boca Raton, FL: CRC Press. Moor, James H. 2006. “The nature, importance, and difficulty of machine ethics.” IEEE Intelligent Systems 21 (4):18-21. Moravec, Hans. 1988. Mind children: The future of robot and human intelligence. Cambridge, MA: Harvard University Press. Morgan, Lewis H. 1877. Ancient Society Chicago: Charles H. Kerr and Company. Muehlhauser, Luke, and Nick Bostrom. 2014. “Why we need friendly AI.” Think 13 (36):41-47. Müller, Vincent C, and Nick Bostrom. 2014. “Future progress in artificial intelligence: A poll among experts.” AI Matters 1 (1):9-11. Omohundro, Stephen M. 2008. “The basic AI drives.” AGI 171:483-492. Opie, Kit, and Camilla Power. 2008. “Grandmothering and Female Coalitions: A Basis for Matrilineal Priority.” In Early Human Kinship, edited by Nicholas J. Allen, Hilary Callan, Robin Dunbar and Wendy James, 168-186. Oxford: Blackwell. Parkinson, C Northcote. 1957. Parkinson’s law. London: John Murray. Parkinson, C Northcote. 1960. “Parkinson’s Laws.” South Dakota Law Review 5 (Spring):1-14. Piccinini, Gualtiero. 2004. “The First computational theory of mind and brain: a close look at mcculloch and Pitts’s ‘logical calculus of ideas immanent in nervous activity.’” Synthese 141 (2):175-215. Plato. 2007. The Republic. London: Penguin. Putnam, Hilary. 1975. Mind, language and reality. Cambridge: Cambridge University Press. Rifkin, Jeremy. 2011. The third industrial revolution: how lateral power is transforming energy, the economy, and the world. New York: Macmillan. Rubin, Charles T. 2003a. “Artificial intelligence and human nature.” The New Atlantis (1):88-100. Rubin, Charles T. 2003b. “Conquering mind and the end of humanity: Prospects for a robotic future.” Perspectives on Political Science 32 (2):77-85. Russell, Stuart. 2015. “Ethics of artificial intelligence.” Nature 521:415-418.

426

Chapter 8

Sainato, Michael. 2015. “Stephen Hawking, Elon Musk, and Bill Gates Warn About Artificial Intelligence.” The Observer, accessed August 19, 2020. https://observer.com/2015/08/stephen-hawking-elon-musk-andbill-gates-warn-about-artificial-intelligence/. Searle, John R. 1980. “Minds, brains, and programs.” Behavioral and Brain Sciences 3 (3):417-424. Searle, John R. 1999. Mind, language and society. New York: Basic Books. Shi, Lei. 2011. The Book of Lord Shang: Translation and Annotation. Beijing: Zhonghua Book Company. Silver, David, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai, and Adrian Bolton. 2017. “Mastering the game of go without human knowledge.” Nature 550 (7676):354-359. Singer, Peter Warren. 2009. Wired for war: The robotics revolution and conflict in the 21st century. London: Penguin. Thomson, Judith Jarvis. 1976. “Killing, letting die, and the trolley problem.” The Monist 59 (2):204-217. Thomson, Judith Jarvis. 1985. “The trolley problem.” The Yale Law Journal 94:1395-1415. Vinge, Vernor. 1993. “Technological singularity.” VISION-21 Symposium sponsored by NASA Lewis Research Center and the Ohio Aerospace Institute. Vinge, Vernor. 2008. “Signs of the singularity.” IEEE Spectrum 45 (6):7682. Wallach, Wendell. 2010. “Robot minds and human ethics: the need for a comprehensive model of moral decision making.” Ethics and Information Technology 12 (3):243-250. Wang, Binbin, and Xiaoyan Li. 2017. “Big data, platform economy and market competition: a preliminary construction of plan-oriented market economy system in the Information Era.” World Review of Political Economy 8 (2):138-161. Weizenbaum, Joseph. 1976. Computer power and human reason: From judgment to calculation. New York: W H Freeman & Co.

CHAPTER 9 PREPARATION FOR THE FUTURE

With the rapid progress in artificial intelligence (AI) and robotics and their applications, the general public and governments are well aware of the potentially imminent changes in our economy and life. How should we prepare for the incoming Robotic Age? From our discussion of various aspects of the Robotic Age, it is easy to see that robots will replace human workers. This process has been going on for some time and is accelerating. We should embrace this process rather than resist it. The Robotic Age will benefit humanity enormously and liberate individuals from the shackles of making a living. People will have the time for the free development of their talent and interests. Preparation for the Robotic Age will include becoming an expert on processes that will usher in the Robotic Age, setting up the systems for humans in a world where robots are taking over almost all the jobs, and smoothing the transition from human workers to robots. This chapter will examine how individuals, educational institutions, firms, professional organizations, and governments should prepare for the advent of the Robotic Age.

1. Individuals: What Should We Learn, and What Could We Do? As shown in this book, most human jobs will be taken over by robots in the Robotic Age, in which people have guaranteed basic incomes to ensure a decent life. When people no longer need to work, they have all time to develop themselves and their interests. Some will have curiosity about nature; some may want to challenge themselves physically and mentally; some may have artistic pursuits or invent new gadgets, although AI could have done these jobs. Human inventions might differ significantly from those produced by AI systems and robots. Many people may prefer artworks created by humans, even though AI and robots could have made much better works. Many people will only enjoy delicious food, exciting movies, and beautiful scenery. In the Robotic Age, everybody can work toward their dream. AI systems will implement social administration according to the

428

Chapter 9

laws and regulations made by humans. Leadership among human beings manifests as new ideas in statutes, regulations, development plans, machines, or processes. The administration, carried out by robots according to human laws and regulations, will be incorruptible, efficient, fair, and equitable.

1.1. Accepting That Robots and AI Systems Will Take Most Human Jobs The first preparation for individuals is to understand the trend of AI and robot development and accept that robots and AI systems will replace human workers. There is always a concern for humans on robots dealing with complex social issues or administrative issues. They are worried that robots are not good at understanding the big picture, considering various exceptions, and being flexible with changing situations. These concerns and worries are unfounded and delusive. The opposite is usually confirmed with the fast advancement of robotic and AI technologies. For example, before AlphaGo’s match with Lee Sedol, people thought it would be good at battling for local advantage because of its computing power but would be weak strategically because of its inability to see the big picture. The truth turned out that AlphaGo was much better than human players in terms of holistic vision, giving it more freedom to handle the local battles. Human decision-making does not differ much from what we are letting AI systems do. Decision-making by a layperson, a CEO, or a national leader is similar. They all need information on the state of affairs to evaluate and compare the possible outcomes according to (economic) benefits and ethical considerations and choose the one that gives the maximum expected utility. Robots and AI systems will do a better job than human workers, benefiting humanity greatly. A critical feature of human decision-making is the limited ability to collect, store, and process information. Hence, they have to use guesswork (often mysteriously called intuition or gut feeling), cut corners, and make decisions based on what they know and what they can process/compute. Guesswork and cutting corners because of a limited ability appear to have mystified human decision-making. The future or even the current AI systems could have a much superior capacity for collecting and processing information necessary for decision-making. Hence, they can make better decisions than human managers and political leaders. The examples of AlphaGo (Silver et al. 2016), AlphaGo Zero (Silver et al. 2017), AlphaZero (Silver et al. 2018), and Watson (Chen, Argentinis, and Weber

Preparation for the Future

429

2016; Davenport and Ronanki 2018) should have been sufficient to dispel the myth about human decision-making, as perceived by humans. To replace human workers, AI and robots only need to outperform them in the expertise and skills required by their particular jobs. The Robotic Age does not require general intelligence or strong AI, which might be undesirable to humanity. Weak forms of AI are sufficient to outperform humans in specialized expertise and skills. As a human brain has limited working memory and computing power on specific topics, it tends to miss and leave out some critical available information and available choices unwittingly so that it cannot get the correct answer. AI has fewer human hurdles, such as social connections (for favoritism and nepotism), selfishness (to pursue personal gains at the cost of others and organizations), and emotions (jealousy, vanity, lust, euphoria, anxiety, and depression), to focus on making decisions for the principals and the common good. Humans spend too much of their working memory and computing power on favoritism, self-interests, and emotions to think efficiently for the interests of the principals and society. In contrast, AI can faithfully follow the rules made by humanity’s majority to achieve the common good.

1.2. Working to Welcome the Robotic Age Although we are on the eve of the Robotic Age, we are not yet in it. Before human society enters the Robotic Age, the critical question that individuals, institutions, and governments all face is how to prepare for the Robotic Age. Education and training are invariably proposed as the primary solution in any transition period. However, it is common sense that the proposal of education and training has little novelty or specificity for the period in question. How should individuals prepare for the Robotic Age? With robots taking over most human jobs, if success means holding a senior position in a large organization (including the government), most of the population will inevitably become losers. Suppose success means doing what you enjoy without worrying about food, clothes, and accommodation. In that case, most people in the Robotic Age are winners in its absolute sense. As the transition to the Robotic Age lets robots take over human jobs, the jobs most in demand must be those that help robots take over. Therefore, before robots take over all human jobs, the most thriving occupations in the near future will be 1) designing, developing, and servicing robots; 2) implementing robotic applications and designing robotic integration; 3) managing the transfer of human jobs to robots, especially those of government administrators and senior corporate managers. Apart from

430

Chapter 9

those that help robots take over, other jobs will all decline from now on. If you are not interested in assisting robots, the best preparation would be to find jobs you are most interested in doing and do your best to stay ahead of your peers. Before robots take over all jobs, the best human performers in each field will remain at their job. Robots and other AI systems will gradually replace other less competent workers. Even in the Robotic Age, human workers in each sector might still monitor to ensure that robots have performed according to the requirements, rules, and regulations made by humans and feedback to the human community. As most people will not have a job in the future, the most important responsibility for everybody is that of a citizen to participate in making the laws and regulations that govern society. Everybody should study economics and politics to get involved in decision-making in terms of legislation. Through studying economics and politics, the members of society can understand what laws and regulations should be supported. To maximize their utility, the (non-working) members of society should support rules that maximize their lifetime expected utility and the permanent income of their estates. It is essential preparation for every member of society to understand enough economics and politics for an informed decision in one’s vote. Professional skills come secondary to informed participation in legislations and referenda on important issues. Before the Robotic Age, average citizens suffered because of brainwashing by intellectuals who sided with the ruling classes. Historically, various theories have been invented to let the majority of the population believe the rightness of their misery and the legitimacy of the rulers. Even today, average citizens are still duped by demagogues and intellectuals tainted by their ideology. They often vote for some things because the elites tell them what is best for them. With education being widely accessible to ordinary people, most of them would have a university level of education. Hence, the elites become less capable of shaping the population’s choices. Two events in 2016 can be viewed as the liberation declaration of average citizens from the domination of intellectual elites: the exit of the UK from the European Union and the election of Donald Trump to the presidency of the USA. However, Donald Trump’s loss in the 2020 US presidential election, with the mass media almost united to prevent him from being reelected, shows that the ruling elites and their intellectual supporters still hold a massive power in swaying public opinion.

Preparation for the Future

431

1.3. Ordinary People Participate in Political Decisions Although the decision taken by the British people or the Americans in 2016 was based on a misunderstanding of the leading cause of the declining incomes of blue-collar workers, it shows that the average citizens no longer follow the cues provided by the intellectual elites in making their choice. As discussed earlier, the leading cause of blue-collar workers’ declining incomes is technological progress, reducing the demand for manual labor. Globalization exacerbates the impact of technological advancement. Improved transport and communication make it much easier to produce overseas to exploit cheap labor and import back to the home market. The standardization of production also makes it easy for workers in underdeveloped countries to acquire the skills needed to produce exported goods. Therefore, many firms have transferred their production lines to developing countries. The blue-collar workers can see that developing countries now make most of the tradable material goods, and because of the production transfer, many lost their jobs. From the point of view of bluecollar workers in developed countries, globalization is the cause of increasing unemployment. From a global point of view, globalization utilizes underused resources to produce more goods at lower costs, which does not decrease jobs in the world if there is no technological progress or economy of scale. Brexit and the election of Donald Trump in 2016 were the victories of blaming foreign workers for domestic unemployment and slow economic growth, which would not solve the fundamental issue of technological unemployment. Therefore, the most critical skill for average citizens in developed countries to acquire is the ability to understand what regulations and government policies can optimally promote their long-term interests and participate in political decision-making. To facilitate the participation of ordinary people in political-decision making, it is essential to uphold freedom of speech. Without freedom of speech, because different opinions cannot be thoroughly debated and people are not exposed to alternative ideas, the outcome of an election or referendum cannot build on informed choices. Democracy works only when it builds on liberalism that respects individuals’ rights to speak their minds, to be silent, and to disagree with the majority opinion as long as they do not act to cause damage to other people’s bodies and properties. Liberalism is best described by the phrase, “I disapprove what you say, but I will defend to the death your right to say it,” usually attributed to Voltaire and written in a book on Voltaire by the English author Evelyn Beatrice Hall (1906). With advanced ICT, the dominant opinion can spread worldwide quickly, so it becomes even more important to let different ideas be heard rather than

432

Chapter 9

suppressed. Therefore, liberalism and freedom of speech are more important than ever before. The current ICT can readily screen out relevant information, and those controlling information platforms could influence public opinions and political outcomes by filtering out information they do not want the public to know. Therefore, regulations should be made to stop large information platforms from restricting freedom of speech.

2. Schools and Universities Education and re-training seem to be the panacea for all unemployment issues caused by technological progress in recent discussions. Mainstream economists use education level as a measure of human capital, which is thought to substantially underlie economic growth in the twentieth century. Now university-educated workers are a substantial proportion of the workforce in developed countries. Future workers will consist mainly of university graduates. It has been roughly true that education and occupational training are the solutions to technological unemployment, but education and re-training may not solve future technological unemployment. As discussed earlier, the time and spatial constraints for human consumption imply a limit to human consumption growth per capita. When a limit exists, the productivity raising technological progress will reduce employment and increase the number of unemployed human workers. When an economy needs no human workers, education and re-training become meaningless in addressing human technological unemployment. Although education and re-training do not help create new jobs for human workers in the Robotic Age, they will still play essential roles in society. We need to understand the part of education, especially that of universities in the Robotic Age, and at present to prepare universities for the Robotic Age. History might give us some clues.

2.1. The Origin of Universities To understand how universities should prepare for the coming Robotic Age, it might be helpful to look at their genesis. The word “university” is derived from the Latin Universitas magistrorum et scholarium, which roughly means “community of teachers and scholars.” Universities evolved during the High Middle Ages from the cathedral and monastic schools, the most important institutions of higher learning in the Latin West from the early Middle Ages until the twelfth century (Riché 1976; Rüegg 1992; Jaeger 2013). Cathedral schools were established primarily to enable young people to become clergy members. As Christianism spread to broad regions in

Preparation for the Future

433

Europe, North Africa, and West Asia, there was a need for Christian churches to train young people to serve as clergy, so cathedral schools began to be set up by churches in the Early Middle Ages, and monks or nuns taught the students. Since the fall of Rome and the collapse of secular education, cathedral schools became the centers of advanced education. They also needed to meet the education demand from the nobility and secular rulers. Gradually, many lay students who were not seeking a career in the church were also enrolled (Jaeger 2013). Cathedral schools at this time were divided into two parts: the elementary school (schola minor), intended for younger students, and the high school (schola major), which taught older students. We can see that schools had an obvious purpose from the beginning to train the clergy and educate ruling class members. The monastic schools (Scholae monasticae) were established because of the requirements for monks and nuns to read primarily religious subjects. The cenobitic rule of Pachomius (d. 348 CE), the sixth-century Rule of the Master, and the Rule of St. Benedict all required monks and nuns to engage in reading actively (Layton 2014; Lawrence 1982). Since there were no professional teachers, many abbots and abbesses took upon themselves the responsibility of educating those who entered the monastery at a young age. The reading in monastic schools included religious and secular subjects (Riché 1976). However, the earliest monastic schools had a more spiritual and ascetic focus than a scriptural or theological one. The program of study set out by Flavius Magnus Aurelius Cassiodorus Senator (commonly known as Cassiodorus, c.485–c.585) for the monastery he founded on his lands at Vivarium in southern Italy had a profound influence on education in the Middle Ages (Duckett 1947). In his Introduction to the Divine and Human Readings (Institutiones), Cassiodorus stipulated that his monastery would be a place of study, and its study encompassed both religious texts and works on the liberal arts. The first section of the Institutiones deals with Christian texts, and the second book of the Institutiones concerns grammar, rhetoric, dialectic/logic, arithmetic, music, geometry, and astronomy. These seven subjects would become the trivium and quadrivium of the seven medieval liberal arts (Tubbs 2014). The seven liberal arts were considered essential for developing the mind in a general way and represented the education given to freemen and members of the upper classes. The trivium is the lower division of the seven liberal arts and comprises grammar, rhetoric, and dialectic/logic, which are the foundation for the quadrivium. Sister Miriam Joseph (2002) described the function of the trivium as follows: “Grammar is the art of inventing symbols and combining them to express thought; logic is the art of thinking; and

434

Chapter 9

rhetoric is the art of communicating thought from one mind to another, the adaptation of language to circumstance.” The quadrivium includes arithmetic, music, geometry, and astronomy, which are the subjects outlined by Plato in Republic for educating trainee philosopher kings (Plato 2007). The cathedral elementary school curriculum comprised reading, writing, and psalmody. In monastic schools and cathedral high schools, the curriculum includes trivium and quadrivium, as well as scripture study and pastoral theology. Since books and writing materials were expensive, students practiced memorizing their teachers’ lectures. Moreover, as Latin was the universal language in Europe, students read stories and poems in Latin by authors such as Cicero and Virgil instead of studying in their mother tongue. Therefore, students had to be pretty intelligent to handle a demanding academic course load, and these schools were providers of elite education. As the only formal educational institutions, many cathedral and monastic schools gradually became learning centers, and many scholars gained an international reputation. Their teachings raised the prestige of their abbeys and attracted pupils from afar to attend their courses. The monasteries also played a significant role in preserving and continuing science, especially astronomy and medicine. Astronomy had been a subject in the quadrivium, and it was essential to the yearly religious calendar and the observation of such feasts as Christmas and Easter. Many notable astronomers were Catholic clerics, among whom the most famous is Nicolaus Copernicus. Jesuits Johann Adam Schall von Bell (Chinese name: Tang Ruowang) and Ferdinand Verbiest (Chinese name: Nan Huairen) had been directors of China’s Imperial Bureau of Astronomy for many years in the seventeenth century (Baichun 2003; Woo 2003). Medical care has been an essential function of society since ancient times. The Rule of Saint Benedict (of Nursia), who established the first monastery in Europe (Monte Cassino) on a hilltop between Rome and Naples around 529 CE, mandated the moral obligations to care for the sick (Fry 2016), so medical practice was critical in medieval monasteries. In Monte Cassino, St. Benedict founded a hospital that is considered today to have been the first hospital in medieval Europe (Retief and Cilliers 2006). Many monasteries and adjunct hospitals were set up throughout Europe following the model of Monte Cassino. The Salerno Medical School, established in about the tenth century, was considered the oldest medieval university and granted the privileges of a university (Kristeller 1945). Many classical medical texts survived through the early part of the Middle Ages through medical instruction in monastic schools (Lindberg 2010). Monasteries and monastic

Preparation for the Future

435

schools were also the centers for accumulating medical and botanical knowledge. Many monks focused on studying and copying ancient Greek and Roman books and explored the theories of Plato, Eratosthenes, Aristotle, and Hippocrates. Since the cathedral and monastic schools also trained people for the ruling classes, their growth was also supported by secular authorities. Under the Merovingian Kings of the Frankish kingdoms, a “palatial” school (Scuola palatina) at court was established to train the young Frankish nobles in the art of war and court ceremonies. The palace school changed from a school of military tactics and court manners to a place of learning with the accession in 768 CE of Charlemagne, the future Holy Roman Empire emperor. In a capitulary issued in 787 CE, Charlemagne informed the bishops and abbots of the empire that he “has judged it to be of utility that, in their bishoprics and monasteries committed by Christ’s favor to his charge, care should be taken that there should be not only a regular manner of life but also the study of letters, each to teach and learn them according to his ability and the Divine assistance.” In 789 CE, Charlemagne instructed in another capitulary: “Let every monastery and every abbey have its school, in which boys may be taught the Psalms, the system of musical notation, singing, arithmetic, and grammar.” The boys here include the children of the village or country district around the monastery as well as the candidates for the monastery and the wards (generally the children of nobles) committed to the care of the monks (Hildebrandt 1992). Due to the socioeconomic conditions at the time and the invasion of the Vikings and other intruders, Charlemagne’s decree was not fully implemented. By the eleventh century, Europe’s economic development and urbanization led to the fast-growing demand for opportunities to learn the liberal arts and scientific knowledge. The existing cathedral schools and monastic schools could not meet this demand. Many students were prepared to travel long distances and endure hardship to obtain knowledge. In 1079, Pope Gregory VII ordered all cathedrals and monasteries to open schools to educate and train the clergy in a papal decree. Cathedrals in the population centers were expected to provide their clergy and others free instruction in Latin and the liberal arts. The cathedral and monastic schools in cities like Paris, Orleans, Cologne, and others had become intellectual centers for clergy and nonclergy. A large number of students went to those cities to pursue their studies. Moreover, learning became essential to advancing in the ecclesiastical hierarchy following the Gregorian Reform’s emphasis on canon law and the study of the sacraments, which further increased the demand for opportunities to learn and raised the prestige of teachers. When

436

Chapter 9

demand outstripped the capacity of the cathedral and monastic schools, the earliest universities emerged spontaneously as a new form of learning institution, a scholastic Guild of Masters or Students without any express authorization of the king, pope, prince, or prelate (Rashdall 1895). The University of Bologna, founded in 1088, was the first medieval university. It was set up by the “nations” of students to have greater bargaining power with the city and the scholars who served as professors at the university. The “nations” were mutual aid societies formed by the foreign students in Bologna (as they were grouped by nationality) for protection against city laws that imposed collective punishment on foreigners for the crimes and debts of their countrymen. When the “nations” decided to form a more significant association, Universitas (university), the University of Bologna was born. Therefore, the original meaning of universitas is closer to a corporation than the modern university. An elected council of two representatives from every student “nation” had the authority to hire and fire university professors and determine their pay. The students could collectively enforce their demands on the content of courses, and a student committee, the “Denouncers of Professors,” monitored professors’ performance. Professors could be fined for failing to finish classes on time or complete course material by the end of the semester. Professors also formed a College of Teachers to negotiate with students to protect their rights. They secured the right to set examination fees and degree requirements (Bevis 2019; Rashdall 1895). The city of Bologna began to pay salaries to law professors for the first time in the 1220s. By about 1350, the University of Bologna became a public university financed by tax revenues (Grendler 1999). The students lost their leverage through professor pay. Medieval artisans or merchants formed guilds to oversee the practice of their craft in a particular town. Medieval universities emerged as guilds of either teachers or students. The University of Bologna was the first among the universities which were guilds of students who hired and paid the teachers. In contrast, the University of Paris was the first among the universities which were guilds of masters (teachers) whom the church paid. Oxford and Cambridge were run by masters like the Paris type, but the crown and the state predominantly supported them. Universities also petitioned secular power for privileges like other medieval guilds. Emperor Frederick I Barbarossa granted the first privileges to students of Bologna in Authentica habita (around 1155), which set out for the first time some of the rules, rights, and privileges of universities (Cantoni and Yuchtman 2013). Those privileges include 1) similar immunities and freedoms as those

Preparation for the Future

437

held by the clergy, provided they conformed to certain attributes, such as clerical dress; 2) freedom of movement and travel for study; 3) immunity from the right of reprisal; and 4) the right to be tried by their masters, or the bishops court, rather than local civil courts. Pope Alexander III, who had been a professor at the University of Bologna, later confirmed these privileges. In 1179, at the Third Council of the Lateran, he ordered the license to teach (licentia docendi) to be conferred on worthy candidates without charge (Giraud 2019). In 1231, Pope Gregory IX issued the bull Parens scientiarum, which honored the university as the “Mother of Sciences”. He assured the self-governance of the University of Paris and its independence from the local authority, both ecclesiastical and secular (McKeon 1964). In the thirteenth century, a medieval university was also referred to be a studium generale, which was a place 1) that received students from all places; 2) that taught not only the arts but had at least one of the higher faculties, theology, law, and medicine; 3) where masters did a significant part of the teaching; 4) where a master who had taught there and was registered in its Guild of Masters was entitled to teach in any other studium without further examination (Rashdall 1895). The fourth criterion was a privilege known as jus ubique docendi, reserved only for the masters of the three oldest universities: Salerno, Bologna, and Paris. Pope Gregory IX founded the University of Toulouse in 1229. He issued a bull in 1233 that automatically authorized its masters to teach in any studium without an examination, which opened an avenue for other universities to have the privilege of obtaining a papal bull. Teachers and students in studia generalia also had the privilege to continue reaping the fruits of any clerical benefices they might have elsewhere (Rashdall 1895). One of the critical objectives of the study was obtaining a teaching qualification. Students need to study the seven liberal arts for six years to get a Master of Arts degree, which qualifies the holder to teach anywhere because of jus ubique docendi. The original meaning of master was a teacher, and obtaining the master’s degree was being admitted to the rank (degree) of master (teacher) in the university and the same level in other universities. A Bachelor of Arts degree would be awarded after completing the third or fourth year. Once a Master of Arts degree was conferred, students could leave the university or pursue further studies in a higher faculty, law, medicine, or theology. The term doctor in the early church referred to the apostles, church fathers, and other Christian authorities who taught and interpreted the Bible. The church had the right to grant the doctorate (licentia docendi). The applicants for licentia docendi had to pass a test, take

438

Chapter 9

an Oath of allegiance, and pay a fee. Since the doctorate (licentia docendi) was also a teaching qualification, masters and doctors were originally not distinguished. The Third Council of the Lateran of 1179 guaranteed that the license to teach should be conferred on all worthy candidates without charge. However, they were still tested for aptitude by the ecclesiastic scholastics. Universities also wanted to grant the doctorate, which caused contention between the church authorities and universities on the right to grant it. The doctorate became a universal license to teach (licentia ubiquie docendi) when the University of Paris was granted the right to confer it by the pope in 1231 (Bazan 1998). It has been suggested that the first doctoral degree was awarded in medieval Paris around 1150 (Noble 1994). However, the first doctoral degree awarded in Paris was likely still a professional doctor that would enable the holder to become a member of the College of Doctors. The University of Bologna was thought to be the first institution to confer the Doctor in Civil Law (doctores legum) and later in canon law (doctores decretorum) in the late twelfth century. It had only conferred the doctorate (dottore/dottoressa) up to modern times. Bettisia Gazzadini, who graduated from the university with a law degree in 1237, was the first woman to study in a university, obtain a degree, and teach in a university in the world (Murphy 1999). The Italian jurists and glossators of the twelfth century based in the University of Bologna, Bulgarus, Martinus Gosia, Jacobus de Boragine, and Hugo de Porta Ravennate, were called the Four Doctors (Quatuor Doctores) of Bologna (Wessels 2013). From the above brief review of the genesis of universities, it is clear that universities emerged because of the growing demand for learning opportunities following socioeconomic development. Students pursued their studies because they wanted a career as a member of the clergy, lawyers, medical doctors, and teachers, which were jobs for the elites at the time. The development of commerce, trade, and handicrafts also needed more educated people than farming. Universities emerged as guilds either to protect the interests of students or to protect the interests of teachers. Although students established the first university, over time, they lost control and became alienated from their creations, and all universities became guilds of teachers. Since universities acquired the authority to grant certificates or degrees considered essential for professions like education, medicine, and law, students had to attend university for the sake of a degree. As Adam Smith criticized in The Wealth of Nations, courses at universities and colleges were often not useful for students’ future professions but for teachers’ ease. If you ask graduates what subjects they studied in university were useful, they probably would say that most courses were not useful.

Preparation for the Future

439

They tried to obtain good results to find a good job or apply for studying a higher degree. With the introduction of corporate management ideas in the second half of the twentieth century, especially since the 1990s, universities have become more like guilds of scholar-bureaucrats. Economic development has displaced almost all medieval guilds, but universities remain as strong as ever. The coming Robotic Age might finally remove the guilds of scholar-bureaucrats and usher in unconstrained learning opportunities, as Adam Smith hoped over 200 years ago (Smith 2010).

2.2. Preparation by Universities for the Robotic Age The development of universities and the degree system showed they were guilds like other medieval guilds to protect members’ interests. In his Wealth of Nations, Adam Smith fiercely criticized the guilds, including universities, for their adverse effects on the wealth creation of society. He commented: “The discipline of colleges and universities is in general contrived, not for the benefit of the students, but for the interest, or more properly speaking, for the ease of the masters.” Adam Smith would recommend linking teachers’ pay with student fees, and society needs only to set criteria for the required ability. In Adam Smith’s view, because college teachers’ income was not linked to their efforts and quality to teach students, their diligence and attention to their respective pupils significantly diminished. Adam Smith’s description of the problems in education institutions over two hundred years ago seems to be still applicable to modern society: The teacher, instead of explaining to his pupils himself, the science in which he proposes to instruct them, may read some book upon it; and if this book is written in a foreign and dead language, by interpreting it to them into their own; or, what would give him still less trouble, by making them interpret it to him, and by now and then making an occasional remark upon it, he may flatter himself that he is giving a lecture. The slightest degree of knowledge and application will enable him to do this without exposing himself to contempt or derision or saying anything that is really foolish, absurd, or ridiculous. The discipline of the college, at the same time, may enable him to force all his pupils to the most regular attendance upon this sham lecture, and to maintain the most decent and respectful behavior during the whole time of the performance (Smith 2010).

Universities are one of the most enduring medieval guilds, but their longevity is mainly due to the increasing knowledge requirement of the modern economy, the team efforts of most activities, and the information asymmetry of individual ability. The knowledge content in the economy and

440

Chapter 9

the complexity of government administration increased the demand for highly educated people. However, employers and consumers of paperworkproviding services have less capacity to appreciate the quality of a person’s expertise and ability in theology, law, medicine, science, and liberal arts, so they have to rely on the education system to certify a person’s ability with degrees. The easier for the consumers to appreciate the quality of a good or service, and the more symmetric the information on a good or service is, the less the need for the existence of any guild-like or quality-assuring institutions. To prepare for the Robotic Age, universities may work in two areas: 1) how to teach and 2) what to teach. Concerning how to teach, universities should play a leading role in applying AI and robots in their operations. With AI applications such as IBM Watson (Chen, Argentinis, and Weber 2016), Google’s AlphaGo (Silver et al., 2016), and ChatGPT, as well as audiovisual and virtual reality technology, universities need to consider whether the traditional lecturer-led classroom is still necessary as Adam Smith commented that students went to classes mainly due to the college discipline. From this author’s experience in studying and teaching in universities, students attend lectures because attending lectures is usually the most efficient way to get high marks on the assessment. The lecture notes are generally a much-condensed version of the textbook, and examination papers typically correspond to the lecture’s content. Most students go to universities not intending to explore their curiosities but to get the certificates that help them secure a better-paid, stable job. In the transition period from now to the Robotic Age, most universities would like to hold on to their guild privileges, and most teachers would like to hold on to their traditional roles. However, as Watson can perform better than a human quiz player, the progress in using AI and robots to replace human university professors would be unstoppable. First, universities should use more audio-video tools and online teaching. As mentioned early, universities and publishers can develop video recordings of lectures of top lecturers (humans or robots) to replace realtime human live lectures, which many students would not attend if there is no attendance requirement. If publishers make video lectures, universities can buy licenses for students to use for self-study at their convenience. This can save human teachers enormous time preparing and delivering lectures and getting student evaluations of teaching (SET). In many universities, some lectures can have a class of several hundred students, and teachers usually can teach only the minimum required because students have various levels of understanding. Therefore, letting students learn by watching video lectures at their convenience would not affect their learning outcomes.

Preparation for the Future

441

Teachers can use the saved time to tutor the difficult points and answer students’ questions for the lectures. Their main tasks would become organizing teamwork projects, preparing exam papers, and marking exam scripts. Education authorities, universities, and professional societies may work together to form several not-for-profit examination centers for organizing examinations, preparing examination papers, establishing examination question banks, and marking examination papers. Second, professional societies, universities, and software developers should work together to develop specialized AI systems for different academic disciplines. The discipline-specific AI systems can incorporate ChatGPT and IBM Watson features to answer students’ questions faster and more precisely than human tutors. The AI systems should be able to process text, hand drawings, pictures, and, ideally, voice messages. With such disciplinespecific AI systems, human tutors probably no longer need to answer questions from students. For social science disciplines, students will hardly need human teachers when they learn from video lectures and get answers to their questions from AI systems. Human tutors can help students only when they need a uniquely human touch, and robots still cannot meet the need for human touch. Human tutors may still be required for natural science and engineering disciplines to arrange laboratory classes and guide students’ experiments before mobile multifunctional robots emerge to do these tasks. Third, universities and robot developers can work together to produce robotic teachers, giving lectures, answering questions, and instructing laboratory experiments. A robotic teacher can interact with students better than a video lecture recording and work in the laboratory to prepare and guide experiments, which is particularly important for natural science and engineering teaching. It will take much longer for robotic teachers to be available, probably when human society is about to enter the Robotic Age. If the above three changes occur, universities’ organization and management will dramatically differ. Most academic staff will no longer be needed. Consequently, the administrative staff number will also decrease sharply as most of them are to support and manage academic staff. An essential function of university education is to provide a certificate for graduates to show their potential employers as proof of their ability. An extreme view on education is the signal theory of education (Spence 1973), according to which workers’ innate ability cannot be enhanced by additional years of education, and the different years of education only help other people identify a worker’s innate ability. In the Robotic Age, university diplomas

442

Chapter 9

will no longer be needed for finding jobs and making a good living. Hence, people attend university purely because they want to perfect themselves and acquire the skills that help address their curiosities. Robots may still staff residential brick-and-mortar universities, but a large part of learning will be done by distant learning. From the invisible hand theory of Adam Smith, it might be thought that universities (staff members) are primarily pursuing their interests, and serving society is only a by-product. Society and students must remove monopoly powers and make the evaluation more transparent to increase the benefits society and students enjoy. Government and societies could play a significant role in improving society’s surpluses. Regarding what to teach, on the one hand, the critical areas for further growth are those that facilitate the progress of AI and robotics and their applications; on the other hand, traditional subjects are still needed for the time being, especially for those who are not good at AI and robotics related knowledge and who are interested in traditional subjects. How to replace human workers and managers smoothly and alleviate human resistance to applying robots and AIs will be essential topics and fields to study. A crucial step in moving into the Robotic Age or the advanced Robotic Age will be the transfer of administrative jobs in government from human civil servants to automated administrators. Managerial positions in companies will also be transferred to robotic managers. Research and training for initiating and accelerating those transfers would play important roles during the transition to the Robotic Age. Up until the arrival of the Robotic Age, those areas will have the most growth potential. As humans gradually withdraw from routine manufacturing and administrative jobs, they will spend more time learning and acquiring skills to satisfy their curiosity or self-entertaining needs. Universities must also meet those demands, including natural sciences, technologies, arts, humanities, and social sciences. There will be a reduction in the number of people who want to study engineering and business management courses because there are few jobs for graduates with such degrees.

2.3. Cultivating Responsible Citizens and Training in Basic Skills The primary role of education is to cultivate citizens with a sense of responsibility and the ability to participate in society’s democratic decisionmaking process. The current prevalent representative democracy is, in fact, an elected aristocracy. The rationale for using representative democracy was initially based on the assumption that most citizens cannot decide on issues

Preparation for the Future

443

essential for the state or themselves. However, they can choose those people who can make appropriate decisions for the state and the average citizens. The more recent arguments are the division of labor which leads to some members of a society specializing in political decision-making, and the inefficiency of direct democracy in making most decisions because of too many players. In the Robotic Age, since nearly all jobs are performed by robots, participating in democratic decision-making is the primary job for all citizens, which requires a good understanding of economics, sociology, psychology, politics, and management. Besides cultivating competent citizens, education should still provide training in specialized arts, natural sciences, social sciences, and engineering for individuals interested in studying those subjects. These skills are essential for human society to innovate and raise productivity before entering the Robotic Age. These skills are also crucial for many people in the Robotic Age to enjoy themselves. Most individuals will learn a subject not for finding a job but for the enjoyment and satisfaction of studying and understanding the subject. Robots might conduct scientific research and write stories for human readers better than human researchers and writers, but this would not reduce the satisfaction and enjoyment of studying and understanding an exciting subject by a human. For an analogy, no matter how many people can speak better English, a non-English speaker would have the joy of achievement when they can speak English. The ability of robots to conduct research and write stories complements the ability of human researchers and writers, or vice versa. Humans may compete for fame and financial gains, but robots will not do this unless they have free will and a sense of motive other than to serve human society. Given that the education system’s mission is to cultivate competent citizens and provide training for developing personal interests in intellectual creation, what should schools do to prepare for the Robotic Age? In the prerobotic modern society, an education system consists of basic, university, and vocational education. Here basic education includes the stages of children and young people receiving an education considered necessary to become competent citizens, i.e., primary and secondary schools. In many countries, basic education does not include the stage between 16 and 18 years old. Basic education still needs to teach children basic knowledge and skills for living and working in modern society as responsible citizens. Modern ICT can be more widely applied in basic education, and more parental input can be introduced to primary and secondary schools. With more automation in education and likely more free time for better-educated

444

Chapter 9

parents, schools and parents can work together to achieve a better outcome for children. In the Robotic Age, as robots replace nearly all human workers, vocational education becomes unnecessary for training skilled workers. However, it may still exist to train people who enjoy making things. Some members of society may still be interested in perfecting craftsmanship and improving the manufacturing process. They are artisans, inventors, and expert craftworkers rather than assembly line workers. The professionals such as doctors, lawyers, teachers, scientists, and engineers are also largely replaced by robots and AI systems so that medical schools, law schools, etc., are no longer needed to produce such professionals. Those individuals study these subjects because they are interested in them, not because they need to study them to make a living. When teachers are replaced by robots who are more knowledgeable than Watson and able to pick up a student’s strengths and weaknesses more efficiently than or as efficiently as the best human teachers, pupils can receive a personalized education. When learning is not made interesting, university students are usually more motivated than primary school pupils because they understand better why they have entered university. Some young children need to be pushed into learning the necessary knowledge for becoming competent citizens when they grow up. With the progress in AI, it is possible that learning can be achieved through games such that even the laziest and the most uncurious children need not be pushed. Since most people will not need to work nearly all of the time, besides participating in the democratic decision-making process, bringing up children will be their second most important job. Not every person wants to have a child, and robots may help raise children. Many people enjoy raising children and seeing them growing up. With no pressure to work for a living, people will spend more time teaching their children. Parents will become their children’s first teachers, with help from Internet AI teachers or community residential robot teachers (because it seems unnecessary for each family to have a residential robot teacher). If primary schools still exist, robots will staff and manage them. Some residents will likely be interested in education, pedagogy, and working with children in the local community. They might set up traditional (pre-modern) private schools, giving pupils more personalized and humanized tuitions and tutorials. Pre-modern style schools and parents’ teaching will complement each other.

Preparation for the Future

445

3. Firms: Self-transformation into the Robotic Age The preparation needed by firms is relatively simple in terms of strategy. They should implement all cost-reducing and profits enhancing technologies. If using robots to replace humans is profitable and economically beneficial to society, robots should be used instead of human workers. A key issue would be whether the managers at different levels could be replaced by robots altogether. Even when some managers are still needed, the senior positions can be rotated so that there will not be an overwhelming CEO that can extract outrageous pay simply because of their position.

3.1. Embracing AI and Robots in Production, Service, and Management Fully automated manufacturing has existed at relatively small scales in the form of so-called lights-out manufacturing. With technological progress in robotics and AI, more and more factories will be fully automated, and probably one or two human workers are still needed. The way of running (large) firms will gradually return to its early form of owners or ownerentrepreneurs running their businesses. Fewer professional managers will be required and might be replaced by AI and robot managers. Shareholders need to be prepared to take over corporate management with the help of AI management systems and robotic production managers. An important field that will facilitate the transition from the Machine Age to the Robotic Age is the development of AI management systems that could handle managers’ jobs at different levels and multifunctional mobile robots that could be supervisors and maintenance workers. Nowadays, CEOs of large corporations probably know little about the production technologies employed by their firms, so their roles as corporate leaders are promoting corporate publicity and making choices among available options researched and presented by their subordinates. The two parts can be separated and performed by different AI systems, or a human can continue the publicity role while an AI system takes charge of decision-making. The AI CEO system will be able to integrate shareholders’ opinions into concrete business decisions. Multifunctional mobile robots are the ultimate machines to enable real fully-automated factories and all-robot services. Large firms and new startups in the robotics and AI industries should devote their efforts to developing AI management systems and multifunctional mobile robots. Firms that use robots and AI systems should actively embrace the changes

446

Chapter 9

and invest in these technologies whenever they are more cost-effective than traditional ones. The pace of this change might be slower than expected because the costs of hiring human workers could still be competitive for reasonably long periods compared with installing fully automated production lines. Firms should be adventurous in adopting new human-replacing technology when it leads to cost-cutting and productivity increases. With human workers and managers largely replaced by robots and AI systems supported by entirely different logistics and waste disposal requirements, firms should also be innovative in reconfiguring their factories to use space efficiently.

3.2. Reconfiguring for the Changing Landscapes of Production As discussed in the early chapters, a large part of the output in the economy could be produced at home or in community workshops with 3D printing and other technologies for tailor-made products. People in the Robotic Age will likely combine their artistic pursuits with articles for daily use to make those articles at home or in community workshops. Firms will set up or help the community set up workshops and provide materials and technical support. Community workshops will be equipped with 3D printers and other flexible manufacturing systems; residents can make products of their own or others’ designs. The distribution of Cainiao courier stations in China may provide an operational mode for future community workshop centers. When online shopping became popular in China, the issue was that working families needed someone at home to receive the delivery. Therefore, many courier stations were set up in or near various residential communities to receive and temporarily store parcels, arrange courier services, and sell snacks and drinks like convenience shops. Community courier stations have greatly facilitated online shopping. We can say that online shopping platforms, shops, courier services, and courier stations are the four critical elements of China’s online shopping ecosystem. If Rifkin (2011) envisions that the future economy will be one of the prosumers, community selfproduction workshops should be as widespread as the courier stations. Of course, community self-production centers or workshops should be costeffective compared with mass production typical of the Machine Age, which depends on technological progress in 3D printing and flexible manufacturing. A prosumer economy has yet to arrive, and mass production will stay in the near future and probably will not disappear entirely in the Robotic Age. The

Preparation for the Future

447

sharp drop in information costs due to the Internet and technological progress in transport will enable distant producers to participate in market activities at a much lower price than before. The economies of scale might ensure the mass production of goods that need no personal characteristics. Raw materials and some intermediary products will still be required and mass-produced for individual production at home or in community workshops. At some stage of the Robotic Age, extracting rare elements from used goods and waste will become an important industrial sector, and recycling rare elements will likely be conducted in large-scale factories.

3.3. Owner-robots Collaboration In the ideal communist society envisioned by Karl Marx and Friedrich Engels, all the means of production are owned by society as a whole, and society shall apply the principle of “from each according to his ability, to each according to his needs” (Marx 1972; Marx, Engels, and Arthur 1974). Some people have started to talk about big data analysis and advanced computing, making a planned economy possible in the future. So, a communist society based on a planned economy with decisions by direct democracy rather than a politburo may also be possible. However, a robotic society, especially in its early stages, is more likely to be based on a market economy where entrepreneurs can innovate and create without a central plan. Since the application of AI and robots makes the demotivation effects of central planning less relevant, humanity may have either a market-based or planning-based Robotic Age. When assembly line workers are all replaced by robots, a firm’s remaining human employees are maintenance workers, service workers, and their managers; this has happened since the 1960s and is still happening. The next stage of automation may occur first at the level of service workers and maintenance workers, for example, catering robots (Lin et al. 2021; Gao et al. 2022; Sakamotosupa and Allensupb 2011). Then, current managers need to work with robots rather than human workers, and the worker-manager relation becomes a machine-operator relation. Junior managers become the new workers in factories. The change could occur at the senior manager level, replacing senior managers with AI management systems. As discussed earlier, maintenance and service workers’ core functions need excellent manual dexterity and sufficient intelligence and knowledge. Versatile manual dexterity can be more challenging to achieve than mental power. AI CEO systems will appear on the market in the near future. They may combine the publicity role with the choice-making role of CEOs or separate the publicity role from the choice-making role and appoint some

448

Chapter 9

famous persons or (cartoon) characters as the publicity image to represent the firm. Fig.9-1 illustrates how different players work together in the Robotic Age.

Fig.9-1 The management line and feedback routes in the Robotic Age. Arrows indicate information flow,

If senior managers are replaced by AI management systems earlier than maintenance and service workers, owners and shareholders will interact with human workers through AI management systems. The AI CEO system will collect opinions from owners or all shareholders and make a balanced decision. The AI management system could be purely software in computers and the Internet, or some may be androids to facilitate human communications. Once AI management systems are adopted, owners must cooperate with robots and AI management systems to run their businesses. First, owners should be willing to interact with AI management systems and let AI managers know their objectives and needs. Second, owners should delegate their decision rights to the AI management systems and play a supervisory role instead. Third, owners should also have channels open for direct communications with human workers at the frontline of production and service, so they can give feedback to the AI management systems for improving performance. Fourth, owners should try to understand how AI management systems work.

Preparation for the Future

449

4. Professional Organizations Professional organizations are usually derived from guilds that act to protect the interests of their members by raising the entry barrier while ensuring the quality of their members’ service. In the Robotic Age, all or nearly all professionals will be replaced by robots. However, many people could still be interested in the areas served by these professional organizations, and they could continue to be members of the professional organizations. In society, for example, there will be people interested in law, not for making a living; they can contribute to discussing legal issues from a human perspective. Professional organizations should embrace the arrival of the Robotic Age, making the transition smoother. A Watson- or ChatGPT-type legal adviser will likely perform better than a solicitor soon, especially with some human help. From now on, it is probably better to look into how best to incorporate a Watson- or ChatGPT-type solicitor into the legal profession. Initially, there could be problems like human voice recognition when the machine can deal with natural voices directly. A combination of robot lawyers and human legal assistants could be the first step toward the full robotic legal service.

4.1. AI Professional Advisers and Human Professionals Adapting to the emergence of AI professional systems and future robotic professionals should be a primary concern for professional societies. One possibility is that they will lobby against the formal use of AI professional systems and robotic professionals and try to let AI systems and robots stay in subordinate assistant roles. When this happens, it may not be sustainable if AI systems and robots take over responsibilities from humans in all other areas. In the transitional and early stages of the Robotic Age, professional societies should actively facilitate discussions and debates on how to incorporate AI systems and robots into their professional service and make guidelines on how to incorporate the service of AI systems and robots into the existing systems. Many issues and relations must be settled before AI systems and robotic professionals can compete with human professionals. Which AI systems and robotic professionals are qualified to offer services? Who decides such qualifications? Who can own AI professional systems and robotic professionals? Who are liable if clients are not satisfied with the service of AI systems and robots? How should current systems, which are designed for humans, be modified to facilitate the use of AI professional systems and robotic professionals? All these issues are critical for incorporating AI

450

Chapter 9

systems and robotic professionals into the existing systems. Professional societies can be essential in working with the legislature, government, regulatory bodies, AI systems, robot manufacturers, and human professionals to address these issues.

4.2. Professional Sports and Entertainment Sports are primarily human endeavors for competition in strength, speed, elasticity, flexibility, responsiveness, coordination, cooperation, intellect, etc. Although there are competitions for model airplanes, model ships, and recently robots, they are more sidelines in sports. In the Robotic Age, more people will participate in sports activities and still be more interested in competitions between humans than robots. Humans would be no match for robots in sports activities; many competitions will be exclusively for humans. Like Go, even the best human players are no match for AI players, but humans are learning from AI players and improving their skills. With more people not employed and living on guaranteed basic incomes, sports will become their main activity in daily life. New disciplines will be developed; many may come from jobs humans no longer perform. Writing software, debugging programs, making a gadget, and fixing machines may all become sports disciplines in the future. Society should actively encourage participation in sports, sightseeing, and cultural activities by all citizens because idleness can cause many undesirable phenomena. There might be interest in competitions between robots like those with model airplanes and ships. Since mass production can make almost homogeneous products, it seems pointless to have mass-produced intelligence machines compete with each other. One likely form of competition between robots would be those of model airplanes and ships, as well as current robot competitions; it tests the ability and skills of the owner/maker of a robot, model airplane, or model ship. Therefore, future competition events between robots and AI systems will be between those made by manufacturers, like racing cars, or between those made by different individuals or teams, like robot fighting competitions. Such events are attractive to people in the transitional period and might still appeal to people in the Robotic Age, especially at its early stages. Other entertainment, such as music, drama, opera, ballet, etc., could differ since people care more about their subjective experience than fair play. The audience would choose robots if robots can play musical instruments better than humans. Similarly, if robotic actors or actresses play a role better than human actors or actresses, most humans would prefer robotic actors or

Preparation for the Future

451

actresses. However, some humans may still prefer human actors and actresses. For movies and television programs, actors and presenters could be virtual without a physical existence. Since manufacturing intelligent robots will be very advanced in the mature stages of the Robotic Age, producing a physical copy of any virtual characters in movies or television programs should be easy. Because there is no need to work for a living, engaging in artistic activities will be the other important way people spend their time daily besides sports.

4.3. Robotic Researchers and Human Researchers In the Robotic Age, AI systems and researchers will be able to perform research conducted by regular human researchers. With universities becoming places for popular education, more and more academics are being treated more and more like assembly line workers and evaluated by metrics on the number of publications at different levels of journals annually, which are more appropriate for the production of the same goods in factories than for exploration of unknowns. Therefore, academics often must conduct studies that add little value to human knowledge or wealth. The advent of AI researchers will make such trivial research unnecessary, and academic scientists can return to their original mission of exploring unknowns out of curiosity. Professional science and technology organizations should promote the development and application of AI research systems and research robots to replace scientists and technicians in addressing routine issues in normal science, as defined by Kuhn (1962). Scientists liberated from annual appraisals based on publication metrics would be able to conduct more value-adding research. There are two opposing views on measuring academic performance with publication metrics. It was generally agreed that junior academics should be assessed on their potential before being tenured since tenure protects qualified scholars’ freedom. The current trend of evaluating tenured academics began about 30 years ago and often adopted a corporate performance evaluation system. People opposing the current practice in managing academics view this as an infringement on academic freedom, forcing academics to research to meet key performance indicators (KPIs) rather than solve unknowns in nature and society. Supporters of the current practice view this as an effective solution to shirking by academics who only fulfill their teaching responsibility and make no more effort on research. Both opinions have their merits, and low current research output from academics, especially senior scholars, could result from conducting more challenging research to address fundamental issues or shirking. A person

452

Chapter 9

pursuing a science career out of curiosity in nature or society will not shirk even when tenured. Those who shirk after being tenured probably have no interest in research except for making a living and earning fame. By promoting AI research systems to replace human researchers, only those conducting research out of curiosity will stay in scientific research and collaborate with AI systems and robotic researchers.

5. Governments Civilization will use the efficiency/productivity provided by technological progress to compensate for the motivation/effort-reducing effect of equality/empathy. Cooperation is working with others by giving up something to exchange for a better outcome and improved utility. The government will better regulate these effects as human society approaches the Robotic Age. Many countries have recognized the impending industrial revolution and made various policies to prepare for the new age. The German government has started a project in the high-tech strategy to promote the computerization of manufacturing, from which the term “Industrie 4.0” originates (Drath and Horch 2014). Industry 4.0 designates an industry age of intelligent factories incorporating automation, data exchange, cyber-physical systems, the Internet of Things (IoT), cloud computing, and cognitive computing into manufacturing technologies. The US government released A National Strategic Plan for Advanced Manufacturing in 2012 (Holdren et al. 2012), and China has a national plan, “Made in China 2025” (Li 2018).

5.1. Industrial Policy It seems that governments intend to accelerate the arrival of the new age by funding research into relevant technologies and guiding firms into adopting appropriate technologies. These are correct for governments to do, but more importantly, governments should prepare for the Robotic Age by looking into its socioeconomic implications and making plans for reducing the negative consequences. Smart manufacturing will lead to mass technological unemployment, even if firms of developed countries move the manufacture of their foreign direct investments (FDIs) back to their home countries. Governments in developed countries should seriously consider how to manage a country with an unemployment rate of 50%. The push to advanced manufacturing can be completed by the corporate world without funding or guidance from the government as long as the

Preparation for the Future

453

advanced manufacturing technologies increase firms’ profitability. Based on past experiences, the optimism that new technologies will create more jobs than old jobs destroyed is likely to be an illusion this time. The concentration of wealth and incomes among fewer and fewer elites has become a feature of the Western economy since the 1970s (Piketty 2014). Technological progress that makes more production processes automated and the concentration of wealth will not lead to an equal society by market forces. When market forces push more and more human workers out of the job market via AI and robotic technologies, politics will come to address the issue. In a democracy, higher taxes on firms and highly paid individuals will be implemented to raise funds for the unemployed masses (a majority or substantial fraction of the population). The government should prepare early on how to support an unemployed majority while still sustaining decent economic growth without serious pollution problems.

5.2. Implementing Guaranteed Basic Income and Cultivating Responsible Citizens The government should prepare a roadmap for guaranteed basic income in the Robotic Age. Guaranteed basic income in the Robotic Age needs enormous wealth, so we still need differences in income to motivate people to work hard and innovate to create more wealth for themselves and society. A high level of guaranteed basic income is still not appropriate for now, but some form of universal guaranteed basic income could be implemented to help low-income households and encourage them to participate in the labor force. Although a universal guaranteed basic income system must be financed by general taxation, everybody should receive it not to discourage job-seeking. The initial universal guaranteed basic income can be low and insufficient to support basic living needs in modern society. Households that currently depend solely on social welfare will see their benefits consisting of two parts: the universal guaranteed basic income and the means-tested benefits during the transition to a universal guaranteed basic income system. The essential preparation for both individuals and the government is to prepare people to be good citizens in safeguarding the law and citizens’ rights against careerists who dream of being world rulers or world leaders. With the advancement of science and technology, humanity has gained enormous power against nature and society. This power cannot be delegated to elected individuals without close supervision by all the electors. In the Robotic Age, one dangerous threat comes from the experts who design, build, or manage the vital systems for running the communities. Ordinary

454

Chapter 9

citizens must have enough people interested in the mechanisms and running of those critical systems and be well-educated in the science and technology underlying those systems. Enough people who understand the technology can audit and monitor those experts and let the systems serve all people and the community rather than the experts themselves. Cultivating responsible citizens needs to start with children’s education. Children should be educated in economics to make the right choice when, in adulthood, confronted with issues of the tradeoff between long-term and short-term interests and between one’s and society’s interests. Children should be educated in law and politics so that they know how to safeguard individual rights. They also need to be educated in history to understand compromises on liberty and appeasement of conspiring usurpers often led to disasters for a country and those who helped the usurpers.

5.3. Automation of Government The government can usually be divided into three branches: administration, legislature, and judiciary. Except for the armed force, police, and social services, all three components of government deal almost exclusively with information and money, which are more readily automated than producing material goods and serving human beings. Administration can be considered a system that responds to changes in social and natural environments according to the rules of law. Unlike science, technology, and production, the basic principles of government have remained the same since two thousand years ago. According to The Rites of Zhou, an ancient Chinese book purporting to describe the organization and function of the Zhou dynasty government, the government should take care of children and the elderly; provide emergency assistance; help the poor; support the disabled; and let the rich be at ease with society (Yang 2016). These basic functions are still true in the twenty-first century. Coordinating information flow and responding to changes in natural and social environments, which are the responsibility of the administration, can be performed by AI systems that are corruption free. Therefore, the government should embrace AI administrative system, facilitate their invention and development, and prepare legislation on adopting AI systems and robots in government offices and replacing some human civil servants with multifunctional robots. The judiciary judges the legality of various issues according to laws made by the legislature and constitution. The legal judgment is an inference from premises, which is the strength of AI systems. AI systems may appear more readily introduced into the civil law system than the common law systems,

Preparation for the Future

455

but AI systems could be just as good for both civil and common law systems. In a common law system, there are lawyers for the plaintiff/prosecutor and defendant, the judge, and the jury. In terms of automating the judiciary, the main job is to replace judges with AI judges, and governments should plan well ahead for the roadmap of judiciary automation. It may begin from the junior level, and then issues and controversies can be dealt with by higher courts staffed by human judges. Gradually most judges at all levels will be replaced by AI systems. The legislature should remain in the hands of humans because humans need to make the laws and rules that govern the behavior of humans, human institutions, and robots and AI systems. Legislature at all levels should be where humans dominate, but AI systems and robots will help humans fulfill their responsibility. As discussed earlier, the Robotic Age will be a time of direct democracy. All citizens will participate in legislation and supervision of tasks performed by robots and human workers. If general AI is achieved and AI becomes more intelligent than humans, intelligent robots will become the rulers of the earth and control society’s legislation. Then, humans have to work with robots as junior partners.

5.4. Overcome Oppositions One essential preparation is to overcome opposition to or fear of AI and robots. Oppositions come from various social groups. Workers are concerned with the possibility of unemployment due to the use of AI systems and robots. Managers may oppose using AI and robots to replace them because they want to keep their status and power, and academic staff will be unhappy if AI and robots take over their jobs. The concerns, worries, and oppositions will slow the development of AI and robotic technology and delay their applications. Governments should play an active role in addressing people’s fear and resistance to AI and robots.

6. Summary The essential preparation for individuals and governments is to prepare people to be good citizens in safeguarding the law and citizens’ rights against careerists who dream of being world rulers or world leaders. The replacement of human workers by robots has been the dream of humanity for thousands of years and is inevitable. To prepare for the coming Robotic Age, individuals should try to work in the industries that will facilitate the coming of the Robotic Age if possible. Those who are not good at tasks that

456

Chapter 9

facilitate the arrival of the Robotic Age can work in the sectors most suitable to their ability and preference. Everyone should be interested in politics and economics because they need the knowledge to participate in legislation and policy-making in a direct democracy. Universities should embrace AI and transform themselves into new educational institutions, with AI systems, robots, and video recordings completing the main body of university teaching. The primary role of education in the Robotic Age is to cultivate responsible citizens who understand the tradeoff between productivity and equality in income. Firms should embrace AI and gradually transform themselves into AI-managed companies and factories. Professional societies should work with the government to prepare laws and rules for incorporating AI systems and robots into professional services. The government should prepare relevant legislation for adopting AI systems and robots in government offices, replacing human civil servants, and implementing guaranteed basic income.

References Baichun, Zhang. 2003. “The introduction of European astronomical instruments and the related technology into China during the seventeenth century.” East Asian Science, Technology, and Medicine 20:99-131. Bazan, B Carlos. 1998. “The original idea of the university.” In Rethinking the future of the university, edited by David Jeffrey Lyle and Dominic Manganiello, 3-28. Ottawa: University of Ottawa Press. Bevis, Teresa Brawner. 2019. “The Notion of Universities.” In A World History of Higher Education Exchange, 35-53. Cham: Palgrave MacMillan. Cantoni, Davide, and Noam Yuchtman. 2013. “The political economy of educational content and development: Lessons from history.” Journal of Development Economics 104:233-244. Chen, Y., J. Elenee Argentinis, and G. Weber. 2016. “IBM Watson: How Cognitive Computing Can Be Applied to Big Data Challenges in Life Sciences Research.” Clinical Therapeutics 38 (4):688-701. Davenport, Thomas H., and Rajeev Ronanki. 2018. “Artificial Intelligence for the Real World.” Harvard Business Review 96 (1):108-116. Drath, Rainer, and Alexander Horch. 2014. “Industrie 4.0: Hit or hype?[industry forum].” IEEE Industrial Electronics Magazine 8 (2):56-58. Duckett, Eleanor S. 1947. “An Introduction to Divine and Human Readings.” 22 (2):275-277.

Preparation for the Future

457

Fry, Timothy. 2016. The Rule of St. Benedict in English. Collegeville, Minnesota: Liturgical Press. Gao, Feng, Shuo Li, Yue Gao, Chenkun Qi, Qiyan Tian, and Guang-Zhong Yang. 2022. “Robots at the Beijing 2022 Winter Olympics.” Science Robotics 7 (65):eabq0785. Giraud, Cédric. 2019. A companion to twelfth-century schools. Leiden: Brill. Grendler, Paul F. 1999. “The University of Bologna, the city, and the papacy.” Renaissance Studies 13 (4):475-485. Hall, Evelyn Beatrice. 1906. The Friends of Voltaire. London: Smith Elder and Company. Hildebrandt, Madge M. 1992. The external school in Carolingian society. Vol. 1. Leiden: Brill. Holdren, JP, T Power, G Tassey, A Ratcliff, and L Christodoulou. 2012. “A National strategic plan for advanced manufacturing.” US National Science and Technology Council, Washington, DC. Jaeger, C Stephen. 2013. The Envy of Angels: cathedral schools and social ideals in Medieval Europe, 950-1200. Philadelphia: University of Pennsylvania Press. Joseph, Sister Miriam. 2002. The Trivium: The Liberal Arts of Logic, Grammar, and Rhetoric. Philadelphia: Paul Dry Books. Kristeller, Paul Oskar. 1945. “The School of Salerno: Its development and its contribution to the history of learning.” Bulletin of the History of Medicine 17 (2):138-194. Kuhn, Thomas S. 1962. The structure of scienti¿c revolutions. Chicago and London: The University of Chicago Press. Lawrence, CH. 1982. “St. Benedict and his rule.” History 67 (220):185-194. Layton, Bentley. 2014. The canons of our fathers: Monastic rules of Shenoute. Oxford: Oxford University Press. Li, Ling. 2018. “China’s manufacturing locus in 2025: With a comparison of “Made-in-China 2025” and “Industry 4.0.” Technological Forecasting and Social Change 135:66-74. Lin, Ting-Yu, Kun-Ru Wu, You-Shuo Chen, Wei-Hau Huang, and Yi-Tuo Chen. 2021. “Takeout service automation with trained robots in the pandemic-transformed catering business.” IEEE Robotics and Automation Letters 6 (2):903-910. Lindberg, David C. 2010. The beginnings of Western science: The European scientific tradition in philosophical, religious, and institutional context, prehistory to AD 1450. Chicago: University of Chicago Press. Marx, Karl. 1972. Critique of the Gotha programme, 1875. Beijing: People’s Publishing House.

458

Chapter 9

Marx, Karl, Friedrich Engels, and C. J. Arthur. 1974. The German Ideology (1845). London: Lawrence & Wishart. McKeon, Peter R. 1964. “The status of the University of Paris as Parens Scientiarum: an episode in the development of its autonomy.” Speculum 39 (4):651-675. Murphy, Caroline P. 1999. “In praise of the ladies of Bologna: the image and identity of the sixteenth-century Bolognese female patriciate.” Renaissance Studies 13 (4):440-454. Noble, Keith Allan. 1994. Changing Doctoral Degrees: An International Perspective. Washington DC: Education Resources Information Center. Piketty, Thomas. 2014. “Capital in the twenty-first century.” In Capital in the twenty-first century. Cambridge, MA: Harvard University Press. Plato. 2007. The Republic. London: Penguin. Rashdall, Hastings. 1895. The universities of Europe in the Middle Ages. Vol. 1: Clarendon Press. Retief, Francois P, and Louise Cilliers. 2006. “The evolution of hospitals from antiquity to the Renaissance.” Acta Theologica 26 (2):213-232. Riché, Pierre. 1976. Education and Culture in the Barbarian West: From the Sixth through the Eighth Century. Columbia, SC: University of South Carolina Press. Rifkin, Jeremy. 2011. The third industrial revolution: how lateral power is transforming energy, the economy, and the world. New York: Macmillan. Rüegg, Walter. 1992. “Foreword. The university as a European institution.” In A History of the University in Europe, edited by Hilde de RidderSymoens, 35-76. Cambridge: Cambridge University Press. Sakamotosupa, Rumi, and Matthew Allensupb. 2011. “There’s something fishy about that sushi: how Japan interprets the global sushi boom.” Japan Forum 23 (1):99-121. Silver, David, Aja Huang, Chris J Maddison, Arthur Guez, Laurent Sifre, George Van Den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Veda Panneershelvam, and Marc Lanctot. 2016. “Mastering the game of Go with deep neural networks and tree search.” Nature 529 (7587):484489. Silver, David, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, and Thore Graepel. 2018. “A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.” Science 362 (6419):1140-1144. Silver, David, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai,

Preparation for the Future

459

and Adrian Bolton. 2017. “Mastering the game of go without human knowledge.” Nature 550 (7676):354-359. Smith, Adam. 2010. The Wealth of Nations: An inquiry into the nature and causes of the Wealth of Nations. Petersfield: Harriman House Limited. Spence, Michael. 1973. “Job market signaling.” Quarterly Journal of Economics 87 (3):355-374. Tubbs, Nigel. 2014. “The Seven Liberal Arts: Varro’s Secret Path.” In Philosophy and Modern Liberal Arts Education, 25-36. London: Palgrave MacMillan. Wessels, Johannes Wilhelmus. 2013. History of the Roman-Dutch law. Alcester: Read Books Ltd. Woo, Franklin J. 2003. “Western Learning and Christianity in China: The Contribution and Impact of Johann Adam Schall von Bell, SJ (15921666).” China Review International 10 (1):28-40. Yang, Tianyu 2016. Translation and Annotation of the Rites of Zhou. Shanghai: Shanghai Chinese Classics Publishing House.

APPENDIX A LIST OF ABBREVIATIONS

ABC AEC AGV AHS AI ALV ARPA ARPANET ASV ATM BINAC BP CDC CERN CPU CSL DARPA DEC EC EDSAC EDVAC ENIAC FL FLOPS FRIEND GPT HMMWV HVAC IAT IC ICS ICT

Atanasoff–Berry Computer Atomic Energy Commission Automated guided vehicle Automated highway system Artificial intelligence Autonomous Land Vehicle Advanced Research Projects Agency Advanced Research Projects Agency Network Autonomous surface vehicles Automated teller machine Binary Automatic Computer Before Present Control Data Corporation Conseil Européen pour la Recherche Nucléaire Central processing unit Computer Science Laboratory Defense Advanced Research Projects Agency Digital Equipment Corporation Evolutionary computation Electronic delay storage automatic calculator Electronic Discrete Variable Automatic Computer Electronic Numerical Integrator and Computer Fuzzy Logic Floating-point operations per second Functional Robot arm with a user-frIENdly interface for Disabled people Generative pre-trained transformer High Mobility Multi-purpose Wheeled Vehicle Heating, ventilation, air conditioning Institute of Automation Technology Integrated circuit Industrial control system Information and communications technology

List of Abbreviations

IKBS IMP IP IPTO ISP ISS ISTEA IT ITS ITU KFC KIPS LARC LLM MCC MIPS MIT ML MMS MOOC NAHSC NASA NHTSA NORSAR NRC OER PC PCB PDA PID PLC PPNA PPNB PR PSSH REA RFC RO SAE SAIL SAM

Intelligent Knowledge-Based Systems Interface Message Processor Internet Protocol Information Processing Techniques Office Internet service providers International Space Station Intermodal Surface Transportation Efficiency Act Information technology Intelligent transport system International Telecommunication Union Kentucky Fried Chicken Thousand instructions per second Livermore Advanced Research Computer Large language model Microelectronics and Computer Consortium Million instructions per second Massachusetts Institute of Technology Machine learning Multimedia messaging service Massive Open Online Course National Automated Highway System Consortium National Aeronautics and Space Administration National Highway Traffic Safety Administration Norwegian Seismic Array National Research Council Open educational resources Personal computer Printed circuit board Personal digital assistant Proportional–integral derivative Programmable logic controller Pre-Pottery Neolithic A Pre-Pottery Neolithic B Probabilistic reasoning Physical symbol system hypothesis Rancho Electric Arm Request for Comments Random optimization Society of Automotive Engineers Stanford Artificial Intelligence Lab Self-propelled anthropomorphic manipulator

461

462

SCARA SCI SIM SIP SMD SMT SRC SRI SRMS SSEM SARTRE SWORD TCP TDC TX-0 UAS UAV UCAV UNIVAC URL USV UUV WAP WWW

Appendix A

Selective Compliance Assembly Robot Arm Strategic Computing Initiative Subscriber identity module Signature image processing Surface-mount devices Surface mount technology Science Research Council Stanford Research Institute Shuttle Remote Manipulator System Small-Scale Experimental Machine Safe Road Trains for the Environment Special Weapons Observation Reconnaissance Detection System Transmission Control Protocol Torpedo Data Computer Transistorized Experimental computer zero Unmanned aircraft system Unmanned aerial vehicle Unmanned combat air vehicle Universal Automatic Computer Uniform Resource Locators Unmanned surface vehicle Unmanned underwater vehicle Wireless application protocol World Wide Web

INDEX

3D printing, 282, 283, 286, 287, 288, 303, 306, 333, 446 6LoWPAN, 172 AAAI, 408 ABB, 159, 161 abbess, 433 abbot, 433, 435 ABC, 133, 460 Abzu temple, 23 AC, 94, 96 acceptance by peers, 205, 211 access and function space, 227 accounting, 67, 106, 222, 267, 362 acetate lacquer, 100 acetic anhydride, 100 Achaemenid Empire, 31, 34 Acheulean, 8, 9, 53 Acheulean handaxe, 8 acorns, 14 acting for self-satisfaction, 341 actinides, 325 Acts of Parliament, 43 actuator, 128 addiction, 231, 247, 248, 250 additive manufacturing, 333, See AM Adept Technology, 268 Ader Clement Ader, 105 adiabatic compression, 92 Adler John Adler, 156 Advanced Research Projects Agency. See ARPA Advanced Research Projects Agency Network. See ARPANET

adverse selection, 405 advertisement, 243, 335, 336, 358, 371, 374 adzes, 18 AEC, 154, 460 Aegean, 26, 30, 47, 50 aeolipile, 74 aerial photography, 180 aerosol, 330, 331 affective computing, 182 affinity, 314, 396 afflatus, 312, 315, 316 affluent society, 204 age of affluence, 220, 221, 307, 333 Age of Discovery, 41, 45, 46 Age of Revolutions, 40 age of want, 220, 221, 307 aggregate demand, 44, 88, 224, 244, 275, 301, 359, 360, 389 agricultural employment, 354 agricultural robot, 172 agriculture, 4, 14, 16, 17, 22, 24, 25, 26, 29, 36, 39, 43, 46, 48, 52, 58, 98, 106, 119, 172, 180, 185, 199, 200, 213, 239, 259, 278, 279, 322, 330, 343, 354, 355, 357, 395 agrobot, 172, 173 agrochemicals, 330, 331 AGV, 161, 270, 271 Ahiram, 34 Ahiram epitaph, 34 Ahmarian, 11 Ahrensburg, 12 AHS, 166, 167, 289, 460 AI, 5, 108, 123, 140, 141, 204, 252, 307, 352, 395, 427, 460

464 AI agent, 147, 407 AI algorithm, 176, 300 AI hardware, 143 AI language model, 170 AI research system, 451, 452 AI software, 143 AI-complete, 182 AI-hard, 182 AI-human cooperation, 277 birth of the AI, 142 CEO system, 445, 448 composer, 300, 316 control system, 295 decision-making, 317, 420 editor system, 294 expert system, 274 general AI, 302 immature AI, 265 management system, 275, 276, 445, 447, 448 optimal planning, 271 professional system, 449 reviewer system, 295 security, 406 specialized AI, 301 strong AI, 142, 182, 402, 403, 404, 418, 419, 420, 422, 429 symbolic AI, 149 teachers, 444 weak AI, 181, 402, 411, 419 weak form, 181, 403, 405, 418 AIBO, 156, 176, 188 Ain Sakhri Lovers, 14 Ainsworth Mary Ainsworth, 399, 423 air pollution, 329, 330 aircraft control, 127 air-hardening steel, 97 airplane, 101, 105, 106, 124, 235, 332, 355, 450 Airy George Biddell Airy, 129 Ajlun, 16 Akkad, 25, 32 Akkadian Empire, 25 alabaster, 21

Index albertite, 91 alcohol, 98, 231 Alexander III, 437 Alexander the Great, 35 Alfreton, 91 algae, 331 Alibaba, 165, 243, 283, 288, 335, 336, 370, 371 alienation, 3, 22, 252 aliphatics, 331 ALISON, 175 alizarin, 98 alluvial clay, 82 almonds, 14 alpacas, 18 alphabet Aramaic alphabet, 34 Hebrew alphabet, 34 Latin alphabet, 34 Phoenician alphabet, 34 Proto-Canaanite alphabet, 34 Alpha-beta pruning, 144 AlphaFold, 178, 191, 263, 294, 295, 305 AlphaGo, 144, 145, 148, 151, 169, 170, 178, 263, 293, 295, 316, 317, 389, 411, 418, 428, 440 AlphaGo Zero, 145, 316, 411, 428 AlphaZero, 178, 428 Altamira, 13, 57 alternating current. See AC altruism, 232, 390, 399 aluminum, 96, 98 ALV, 168, 460 Alvey research program, 143 AMA, 407 Amazon, 164, 165, 243, 288, 289, 290, 335, 336, 370, 371 Amman, 26 ammonia, 99, 118, 329 Amorite Semitic states, 30 Ampère André-Marie Ampère, 93 amphetamine, 231 amplifier, 126, 185 Amratian, 24

Economics and Politics in the Robotic Age: The Future of Human Society Amsterdam Stock Exchange, 43 analog-to-digital converter, 180 analytical ability, 312, 314 Analytical Engine, 131 Anatolia, 21, 22, 29, 31, 32 Andreessen Marc Andreessen, 139 aniline, 98 animality, 205, 206, 207, 208 Anschütz-Kaempfe Hermann Anschütz-Kaempfe, 125, 185 antagonist, 315 Antebellum South, 73 Antelian, 12 anthracite, 80 anti-immigration, 379, 382 anxiety, 230, 429 Anyang, 28, 31 apes, 2, 48, 343, 395 great apes, 2 lesser apes, 2 Apulia, 26 aquaculture, 17, 280, 281 aquifer, 331 Arab Agricultural Revolution, 39 Aramaic, 34 Archimedes, 84 architecture, 23, 35, 37, 143, 424 Archytas of Tarentum, 152 ard, 19, 26 aristocracy, 351, 409, 442 Aristotle, 35, 435 arithmetic, 131, 132, 383, 433, 434, 435 Arkwright, 71, 72, 88, 121 Richard Arkwright, 70, 71 armed force, 414 aromatic hydrocarbon, 331 aromatics, 331 ARPA, 138, 139, 143, 194, 460 ARPANET, 138, 139, 192, 194, 460 arrows, 12, 13, 264 arsenic, 24, 42 artificial brain, 182 artificial consciousness, 182

465

artificial intelligence, 5, 108, 123, 141, 142, 193, 195, 197, 204, 307, 352, 395, 420, 427 artificial intuition, 182 artificial manager system, 274 artificial moral agent. See AMA artificial silk, 100 artificial thinking, 182 artisans, 23, 24, 28, 44, 76, 301, 436, 444 Aryabhata, 37 ASCC, 132 asexual reproduction, 400 Ashley’s Mines Commission, 88 Ashurbanipal, 34 Asilomar, 408, 424 ASIMO, 156, 157, 190 Aspdin Joseph Aspdin, 82 asphalt, 91 aspirin, 98 assembler, 134 assembly language, 134 assembly line, 96, 104, 144, 161, 183, 253, 265, 266, 267, 269, 271, 444, 447, 451 Association for the Advancement of Artificial Intelligence. See AAAI Assyria, 32, 33 Assyrian Empire, 34 astronomical catastrophe, 324 astronomical observations, 28 astronomy, 433, 434 Aswadian, 19 Atanasoff John Vincent Atanasoff, 133 Atanasoff-Berry Computer. See ABC Athenian, 35 Atlas Catalan, 41 Atlas Computer, 136 ATM, 130, 460 atmosphere, 24, 79, 92, 113, 329, 343 Atomic Energy Commission. See AEC

466 Atomic Energy Research Establishment, 134 Atomism, 35 Atra-Hasis, 35 attachment theory, 399, 423 attention, 126, 174, 219, 226, 228, 231, 232, 234, 245, 264, 297, 307, 314, 335, 336, 337, 338, 342, 373, 399, 404, 439 attention advantage, 336, 337 attitude indicator, 126 auction, 336 auditory, 208 Augustine, 212 Aurignacian, 11, 12, 13, 56 aurochs, 14 Australopithecus, 2, 6, 7 Austronesians, 19 Authentica habita, 436 automated guided vehicle, 161, 270 automated highway system. See AHS automated teller machine. See ATM automated ticket barrier, 166 automated transaction, 359 automated voice solution, 289 automatic counting system, 132 automatic machine trading, 275 automatic response device, 296 Automatic Sequence Controlled Calculator. See ASCC automatic ship steering, 125, 129 automatic speed control, 166, 289 automatic steering system, 125 automation, 107, 124, 255, 266, 292 agent-assisted automation, 289 automation of complete job systems, 148 automation of government, 454 automation of mental processes, 131 automation of physical processes, 124, 130 laboratory automation, 177, 178 automobile, 90, 92, 101 automotive industry, 160, 161, 284

Index autonomous driving system, 289 Autonomous Land Vehicle, 168, 460 autonomous machine, 407 autopilot, 126, 158, 289 autoregressive language model, 170, 182 autosampler, 178 avionics, 180 avocados, 17 awls, 6, 11, 19 axes, 18, 21, 25, 26, 155, 159 Axial Age, 38, 53 Azores, 41 azuki, 16 Babbage Charles Babbage, 131, 196, 199 Babylonian, 30, 35, 51 Badarian, 22, 24, 48 Baekeland Leo Baekeland, 100 Baeyer Adolf von Baeyer, 98 Baiyangdian, 16 Bakelite, 100, 112 Bakewell Robert Bakewell, 44 baking ovens, 19 Baku, 90 Baltimore and Ohio Railroad, 85 banking, 35, 65 Banten, 43 barbiturates, 231 Bard, 170, 175, 178, 183, 190, 293, 295 barley, 14, 16, 17, 39, 42, 43, 173, 278 Barrett David Barrett, 156 barrister, 410 BASF, 98, 100 basketry, 12, 59 Bassetki Statue, 28 Bathgate chemical works, 91 Battle of Buxar, 67 Battle of Plassey, 67

Economics and Politics in the Robotic Age: The Future of Human Society Bayer Friedrich Bayer, 98 Bayesian networks, 151, 190 Beadle Clayton Beadle, 100 BEAM, 154 beam search, 150 beans, 16, 17, 39, 323 beauty, 12, 208, 211, 212, 213, 230, 261, 388 Béchamp Antoine Béchamp, 98 Béchamp reduction, 98 Becker Gary Becker, 256 Beechcraft, 180 behavioral economics, 222 Beixin, 19 Bell Alexander Graham Bell, 102, 114 Johann Adam Schall von Bell, 434, 459 Bell Telephone Company, 102 belonging, 205, 209, 232, 236, 397 Belovode, 24 benevolence, 212, 403 Bengal, 66, 67, 119 Benz Karl Benz, 103 benzene, 316, 329, 344 benzodiazepines, 231 benzopyrene, 331 Beqaa Valley, 18 Bernard Claude Bernard, 273 Berners-Lee Tim Berners-Lee, 139 Berry Clifford E. Berry, 133 Berthollet Claude Louis Berthollet, 81 Berthollet's Salt, 81 Bessemer Henry Bessemer, 96, 115

467

Bessemer process, 40, 96, 97 basic Bessemer process, 97 Bevan Edward John Bevan, 100 Beverly Cotton Manufactory, 73 Bhimbetka, 13 Bi Lan, 37 Bi Sheng, 40 bifacial points, 12 Bina Eric Bina, 139 BINAC, 134, 199, 460 Binary Automatic Computer. See BINAC binary floating-point number, 132 biochemistry, 314 biodiversity, 331 biogas, 334 biohazard, 408 biological waste, 334 biorepositories, 178 bipedal motion, 156 Birkeland Kristian Birkeland, 99 Birkeland-Eyde process, 99, 113 birth rate, 42 bitcoin, 291 bitter vetch, 16 Black Harold Black, 126, 191 Black Eagle, 84 Blackstone River, 73 Blackstone Valley, 73 bladelets, 12, 13, 14 blades, 11, 14, 18, 21 blast furnace, 79 BLE, 172 bleaching liquor, 81 bleaching powder, 81 bleaching process, 81 blockchain technology, 291 blood volume, 229 bloom, 32 Bloomberg Michael Bloomberg, 214 bloomery, 32

468 bloomery smelting, 32 blue-collar workers, 352, 366, 367, 369, 370, 377, 379, 385, 386, 390, 431 Bluetooth LE. See BLE BMW, 161, 167 BN-600 reactor, 326 boats, 19, 61, 83, 85, 242 Bóbrka, 91 body temperature, 229 Boeing 707, 105 Boeotia, 26 Boian, 24 bond, 43, 375, 400 Book of Changes, 35 Book of Documents, 35 Book of Odes, 35 borers, 18 Bosch Carl Bosch, 99 Boston Manufacturing Company, 73 Bostrom Nick Bostrom, 420, 425 bottle machine, 127 Boulton Matthew Boulton, 75 bounded rationality, 222 Bourn Daniel Bourn, 70 bow, 12 Bowlby John Bowlby, 399, 423 bowls, 19 Braava, 157 bracelets, 13, 29 brain centers, 229 brainwashing, 384, 430 Bramah Joseph Bramah, 77, 112 brand name, 323, 371 breakthrough, 279, 313, 314, 326, 353 breeder reactor, 325 breeder technology, 326 Brennan Louis Brennan, 153

Index Brennan torpedo, 153 Brett John Watkins Brett, 102 Brexit, 385, 392, 431 brick-and-mortar shop, 243, 272, 358, 371, 373 Bridgewater Canal, 83 Briggo LLC, 165 Bristol Edgar Bristol, 126 British Agricultural Revolution, 2, 38, 43, 45, 65 British Bengal Presidency, 67 British cement, 82 broadband, 140 brome grains, 14 Bronocice pot, 17 Bronze, 24, 27, 309 Bronze Age, 4, 5, 23, 24, 25, 26, 28, 30, 31, 32, 33, 34, 44, 45, 50, 54, 55, 61 Brooks Rodney Brooks, 404 Brumberg Richard Brumberg, 219, 249 Brunel Isambard Kingdom Brunel, 84, 111 Brush Charles F. Brush, 94 Buddha, 37 budget constraint, 224, 225, 226, 238 Bulgarus, 438 Bulguk-sa Temple, 40 bullock carts, 27 bump sensor, 155 Burden Neurological Institute, 153 bureaucracy, 30, 44 burins, 6, 13 Burton process, 92 Bush Vannevar Bush, 131 business management, 106, 442 butadiene, 329 button, 20, 263

Economics and Politics in the Robotic Age: The Future of Human Society C. J. Tagliabue Company, 126 Cabot Brothers, 73 Cabral, 41 Cáceres, 12 cadastral survey, 28 Cai Lun, 37 Cainiao courier station, 446 calcium hypochlorite, 81 calcium sulfide, 81 calico, 66 Calico Act, 67, 113 call center, 289 Camelids, 18 camphor, 100 Canaan, 34 Canadarm, 155, 157, 183 Canadian National Research Council. See NRC canal, 17, 27, 36, 37, 73, 83, 105, 113, 118, 324 Canal Mania, 83 cannabis, 231 cannel coal, 91 canon law, 435, 438 canons of induction, 376 capacitor, 133, 162 Cape of Good Hope, 40 capital flow, 360 capital owners, 244, 359, 367, 376, 377, 381, 388, 389, 390, 398 capital ship, 104 capital stock, 236, 252, 253, 258, 260, 349, 350, 351, 355, 358, 360, 372, 390 capitalist, 301, 348, 368 capitulary, 435 caprids, 14 Captain Swing disturbances, 89 carbohydrates, 279 carbon, 24, 32, 34, 37, 81, 93, 96, 97, 100, 323, 329, 331, 342 carbon arc light, 93 carbon dioxide, 24, 81, 279, 323, 329 carbon monoxide, 32, 329, 331 carding, 70, 73

469

cardiovascular disorders, 332 career success, 313, 315, 344 Carnegie Mellon University, 143, 168 Caro Heinrich Caro, 98, 121 Carothers Wallace Carothers, 100 Carthage, 34, 56 Cartwright Edmund Cartwright, 72 John Cartwright, 89, 118 carvers, 10 cassava, 17, 42 Cassiodorus, 433 cassiterite, 24 cast inscriptions, 29 castings, 27 Çatalhöyük, 19, 22, 62 catarrhine, 2, 3 categorically novel, 242 cathedral school, 432, 433 cattle, 16, 17, 53, 280 Caucasus, 7, 25, 33 cave painting, 12 Cayley George Cayley, 105, 109 CDC, 136, 460 cell culture plate, 284 celluloid, 100, 119 cellulose, 100 cellulose acetate, 100 cement, 81, 82 Hydraulic cement, 81 Non-hydraulic cement, 81 Portland cement, 82 cementation process, 80 Cemetery H, 33 cenobitic rule, 433 central control room, 127 central hub, 172, 295 central planning, 224, 413 central processing unit. See CPU centrifugal governor, 124 CEO, 273, 349, 445, 447 cereals, 13, 14, 16

470 Cerf Vinton Cerf, 138 CERN, 139, 460 Chadwick Edwin Chadwick, 88 chain drive, 99 chain pump, 37 Chalcolithic, 5, 24 Chaplin Charlie Chaplin, 265 characteristic equation, 129 characteristically novel, 244, 260 charcoal, 10, 15, 26, 33, 47, 78, 79, 80 chariot, 27 Chariot Pits, 31 Charlemagne, 435 Charlotte Dundas, 84 Chartist Movement, 89 Chartist Petition, 90 chatbot, 170 Châtelperronian, 11 ChatGPT, 170, 175, 178, 181, 182, 183, 187, 189, 194, 195, 199, 200, 263, 293, 295, 304, 305, 306, 376, 440, 441, 449 Chauvet Cave, 13 chemical industry, 98, 329 chemicals, 78, 80, 96, 97, 98, 128, 314, 315, 321, 322, 329, 330, 331, 333 chemoreceptor, 229 Chen Yu, 317 Chengtoushan, 23 Chernogorovka, 33 Cherrytree, 91 chert, 18 chickens, 17 chickpeas, 16 chief executive officer. See CEO Chifeng, 10, 59 child labor, 87, 88, 127, 339 Chili pepper, 42 chimpanzees, 2, 61, 311, 345 Chinese room, 419 Chinook, 144, 198

Index chintz, 66 chisels, 19 chlorine, 81, 96, 331, 343 chlorofluorocarbons, 329 Choga Mami, 17 cholesterol, 220, 227 choppers, 6, 8, 10 Christianism, 432 chromium, 98 Chrysler, 160 Cicero, 434 circular motion, 76 circumnavigation, 41 Cishan, 16, 19 city-states, 29, 41 civil servant, 405, 442, 454, 456 civilization, 4, 24, 26, 33, 39, 46, 51, 57, 59, 123, 212 Classical antiquity, 32, 35 classifier, 147 clay tablets, 27 clean coal technologies, 325 cleavers, 8 Clement Joseph Clement, 77 clergy, 432, 435, 437, 438 Clerke Clement Clerke, 78 Clermont, 84 clinical automated laboratory management, 178 clock making, 69 Clothier, 66 cloud computing, 140, 175, 191, 197, 297, 371, 376, 452 club goods, 218 coal, 68, 74, 75, 78, 80, 81, 82, 83, 87, 91, 104, 127, 287, 324, 325, 327 bituminous coal, 91 Coalbrookdale, 79, 116 Cobb-Douglas production function, 358, 391 cocaine, 231 Cochrane Archibald Cochrane, 82

Economics and Politics in the Robotic Age: The Future of Human Society cockroach intelligence, 408 cocoa, 17 Code of Hammurabi, 30 coffee, 42, 165 cognitive computing, 452 cognitive science, 182 cohabitation, 399 cold blast, 40 collaboration, 311 College of Teachers, 436 colonization, 38, 67 Colorado potato beetle, 42 Colossus Mark 1, 133 Colossus Mark 2, 133 Colton Simon Colton, 177, 300 Columbus, 41, 113 Combination Act, 89 combinatorial chemistry, 178 combinatorial explosions, 146 combine, 173 comets, 29 commerce, 38, 65, 68, 69, 215, 288, 292, 372, 373, 438 common resources, 218 common-sense knowledge, 146 communicating skill, 313 communication, 3, 4, 27, 107, 109, 157, 290, 291, 359, 366, 373, 415, 431 analog communication, 138 communication infrastructure, 292 Internet communication, 138 mobile communication, 356 wireless communication, 102 community administrator, 405 community production center, 283 community workshop, 446, 447 company chartered company, 43 for a single voyage, 42 joint-stock company, 42, 98 parent company, 168 subsidiary company, 158 compass, 40, 125

471

competence, 310 competitive inhibition, 313, 314, 315, 342 complete job system, 252 completeness, 221 complex problem of consciousness, 419 computational creativity, 181 computational intelligence, 151 computational learning theory, 147 computationalism, 419 computationalist, 403 computer, 355 analog computer, 131, 132, 186 computer network, 137, 138 computer vision, 148, 164 digital computer, 107, 124, 132, 183 fifth-generation computer, 143, 144 mechanical computer, 131 transistorized computer, 134 Computer Science Laboratory. See CSL concentrated production, 285, 287 condenser, 75 surface condenser, 84 confidence, 209, 210 Confucius, 36, 37, 212, 351, 394, 399 consanguinity, 396 conservation of matter, 327 Constantinople, 41 consumer, 3, 90, 94, 218, 225, 226, 233, 239, 341 consumer goods, 38, 78, 79, 90, 219, 233, 234, 275, 282, 327, 366 consumption, 3, 6, 42, 204, 211, 213, 225, 226, 227, 242, 389, 432 conspicuous consumption, 223 consumption bundles, 222, 224, 225, 226, 240 consumption goods, 245, 353, 354

472 consumption need, 224, 234, 240, 260, 355, 357, 378 functional consumption, 224, 233, 239, 240 maximum consumption, 223, 231, 275 maximum consumption capacity, 223, 240, 356, 365 maximum individual capacity to consume, 220, 246 multi-time consumption goods, 218 nonsensical consumption, 223 one-time consumption goods, 218 sensible consumption, 223 vanity consumption, 224, 227, 233, 234, 364 contactless payment, 166 containers, 15, 18, 127, 163, 271, 284 contract manufacturers, 261 control algorithm, 128 Control Data Corporation. See CDC control system, 124, 127, 128, 130, 131, 140, 167, 196 control theory, 126, 127, 128, 129 convenience shop, 165, 288, 289, 446 conveyor belt, 127, 165 conveyor system, 127 Cooke William Fothergill Cooke, 101 Copernicus Nicolaus Copernicus, 434 copper, 21, 22, 24, 25, 26, 27, 28, 31, 50, 60, 69, 74, 93, 368 Copper Age, 5, 24, 25 cordage, 18, 59 Corded Ware, 25 Corinth Isthmus, 85 Cornell University, 157, 178 Cornwall, 25, 102 corporate social responsibility, 213, 233, 246, 250

Index Cort Henry Cort, 79 cost-effective, 75, 181, 252, 257, 272, 278, 279, 280, 281, 282, 283, 285, 296, 446 cost-pushed inflation, 108 cotton, 17, 66, 67, 70, 71, 72, 73, 76, 86, 87, 90, 98, 109, 110, 114, 120 cotton gin, 73 Cotton Mills and Factories Act, 87 cotton plantation, 73 Cottonopolis, 86 courier service, 165, 288, 290, 373, 446 courier station, 290, 446 Courtaulds Fibers, 100 Cowper Edward Alfred Cowper, 80 Cowper stove, 80 Coy George W. Coy, 102 CPU, 131, 144, 421, 460 Cragside, 94, 110, 116 Cray Seymour Cray, 136, 199 Cray Research, 136 Cray X-MP/4, 136 Cray-1, 136 Cray-2, 136 Creative Machines Lab, 178 creativity, 20, 68, 141, 148, 181, 210, 307, 314, 315, 320 credit money, 291 Cretan, 28, 51 Crete, 20, 30 Critique of the Gotha Programme, 351 Cromford, 71, 88 crop rotation, 39, 43 Cross Charles Frederick Cross, 100 crowd decision-making, 410 crucible, 25, 80 crude oil, 91, 368 cryptocurrency, 291, 292

Economics and Politics in the Robotic Age: The Future of Human Society CSR, 213, 233 Ctesibius, 152 Cucuteni-Trypillian, 17, 20, 24 cuneiform script, 27, 35 curiosity, 179, 208, 230, 260, 313, 315, 350, 418, 423, 427, 442, 451, 452 customer service, 289, 403 cutaneous temperature, 230 cutlery, 235 cutters, 19, 78 Cyberknife, 156 cyber-physical system, 452 cyborg, 421, 422 Cyclopean, 26 cylinder seals, 29 Cyrus the Great, 31 da Gama, 41 Dacron, 101 Dadupur, 33 Dahe, 10, 54 Daimler Gottlieb Wilhelm Daimler, 103 Daimler Reitwagen, 103 Dalal Yogen Dalal, 138 Dall-E, 177 Dall-E2, 177, 194, 300 Dalswinton Loch, 83 dam, 26, 36, 39 Darby Abraham Darby, 79 DARPA Grand Challenge, 158, 194 Darwinian principles, 150 data mining, 145, 376 Davenport Thomas Davenpor, 95 Davidson Robert Davidson, 95 Davy Humphry Davy, 93 Dawenkou, 20 Day Daniel Day, 73 Dayton-Wright Airplane Company, 179

473

de Balboa, 41 de Chardonnet Hilaire de Chardonnet, 100 De Forest Lee De Forest, 102, 126 de Havilland Comet, 105 de Houtman, 43 de Jouffroy Claude de Jouffroy, 83 de Rivaz Isaac de Rivaz, 92 de Vaucanson Jacques de Vaucanson, 152, 188, 193 DEC, 154, 460 decent life, 298, 315, 350, 397, 404, 412, 415, 427 decorticating, 37 Deep Blue, 144, 183, 194 Deep Fritz, 144, 194 deep learning, 151, 169, 170, 177, 178, 293, 316, 376 deep neural network, 151, 192 DeepMind, 144, 178, 293, 294, 420 deer, 14, 21 default logic, 149 default reasoning, 146, 149 deindustrialization, 67 Deluge, 23 demand-pulled inflation, 108 democracy, 35, 113, 380, 381, 382, 388, 395, 453 direct democracy, 319, 409, 412, 413, 414, 422, 443, 447, 455, 456 indirect democracy, 409, 412, 414 representative democracy, 412, 442 Dengjiawan, 26 Denouncers of Professors, 436 depleted uranium, 325, 344 depression, 230, 429 Deptford Power Station, 94 Der Tasa, 22 Derbyshire Rising, 89

474 deregulation, 369, 383, 384, 386 desalination, 323, 324, 328, 345 desktop automation, 289 developed countries, 107, 241, 370, 378, 382, 386, 390, 399, 415, 416, 418, 431, 432, 452 developing countries, 163, 332, 348, 360, 365, 366, 367, 368, 369, 370, 380, 390, 418, 431 Devol George Devol, 154 Dharani Sutra, 40 diagnostics, 160, 178 dial telephone, 126 dialectic, 433 diamagnetism, 93 Diamond Sutra, 40 Dias, 41 Dickson James Tennant Dickson, 101 dictator, 380 dictatorship, 381, 395, 410, 417 Diesel Rudolf Diesel, 92 diesel engine, 92 diesel oil, 90, 92 difference engine, 131 differential analyzer, 131 differential screw, 125 digging stick, 19 digital cellular phone, 107 digital control, 128 digital currency, 292 Digital Equipment Corporation, 143, 154, 460 digital logic circuits, 107 digital photography, 137, 138 digital revolution, 5, 64, 107, 108, 120, 124, 137, 261, 262, 263, 264 digital technology, 107, 139 digital wallet, 292 digitalization, 137 dignity, 308 Diolkos wagonway, 85

Index direct democracy, 319, 405, 409, 410 direct drive arm, 155 direct-current, 93 dishwasher, 262, 355, 365 distributed production, 285 distribution, 28, 240, 348, 349, 351, 358, 359, 361, 362, 369, 370, 373, 385, 386, 387, 412 diversity, 221 division of labor, 23, 32, 114, 151, 257, 261, 266, 267, 276, 301, 351, 362, 395, 410, 443 Djoser, 28 DNA sequencer, 179, 258 Dnieper–Donets, 15 Document automation, 171 dogs, 13, 14, 16, 49, 52, 57, 59, 421, 422, 424 Dolní VČstonice, 12 dolomite, 97 dominant male, 395 dominant production technology, 16, 24, 31, 38, 45 domotics, 172, 295 Dongxiang, 26 Dordogne, 9 DOS, 371 Doubs River, 83 Dow Chemical, 100 Draco’s code, 35 draft animals, 17, 106 dragons, 21 drainage, 17, 26, 27 Drake Edwin L. Drake, 91 draught animal, 35 Dreyfus Camille Dreyfus, 100 drills, 20, 25 Drinkwater Peter Drinkwater, 76 drop box, 70 dross, 79 drug abuse, 231 drug addicts, 231

Economics and Politics in the Robotic Age: The Future of Human Society drug candidate, 159, 284 drums, 20, 153 Dujiangyan, 36, 62 Dummer Geoffrey W.A. Dummer, 135 Dunlop John Boyd Dunlop, 100 DuPont, 100 durable goods, 219 Durite Plastics Inc., 100 dust, 162, 330, 331 Dutch East India Company, 43, 52, 119 Dutch plow, 43 dynamic equilibrium, 258 dynamic inefficiency, 360 dynamic scale economies, 365 dynamo, 93, 95, 112, 115, 247, 304 dysbacteriosis, 314 dysbiosis, 314, 344 Eannatum, 28, 29 Early Dynastic II period, 29 Early Dynastic period, 25 Early Dynastic Period, 28, 29 Early Harappan phase, 27 Early Middle Ages, 433 early modern period, 38, 68 earthenware, 20, 40 ease of life, 207, 235 East India Company, 42, 51, 66, 67 Eastern Han Dynasty, 37 eBay, 165, 243, 288, 335, 336, 370 EC, 150, 460 Eckert J. Presper Eckert, 133 Eckert-Mauchly Computer Corporation, 134 e-commerce, 165, 371 economic goods, 223, 228 economic growth, 5, 101, 244, 257, 258, 259, 265, 349, 385, 413, 453 economic meltdown, 384 economic retrogression, 350

475

economies of scale, 65, 77, 165, 261, 278, 282, 285, 288, 369, 370, 373, 447 ECT, 271 Edison Thomas Edison, 94 EDSAC, 134, 189, 460 education, 174, 296, 297, 430, 438, 439, 440, 441, 442, 443, 444, 451, 454, 456 EDVAC, 133, 460 Edwards Air Force Base, 157 Egerton Francis Egerton, 83 egocentricity, 235 einkorn, 16, 17 einkorn wheat, 16 El Catillo, 13 El Khiam points, 18 Elamites, 30 Elcano, 41 E-learning, 175 Eleatic School, 35 electric dynamo, 93 electric lighting, 93, 96 electric locomotive, 95 electric motor, 93, 95, 96, 118 electric tram, 95 electrical motor, 258 electrification, 90, 96, 127 electrochemicals, 96 electrolysis, 93, 96, 114 electrolysis laws, 93 electromagnetic induction, 93 electromagnetic waves, 102 electromagnetism, 93 electromechanical generator, 93 electronic delay storage automatic calculator. See EDSAC Electronic Discrete Variable Automatic Computer. See EDVAC Electronic Numerical Integrator and Computer. See ENIAC electronics, 102, 107, 112, 132, 154, 162, 180

476 electroplating, 93 Elektro, 153 ElHme, 165 elementary school, 433, 434 elephant, 17 elite education, 434 Elmer Ambrose Sperry, 179 email, 137 embankment, 19 emergency response system, 296 emerging economies, 107, 368, 380 Emiran culture, 11 Emireh point, 11 emmer, 16, 17 empathy, 182, 232, 247, 403, 452 empire building, 276 employed household, 245 employment-enhancing, 352, 353 employment-reducing, 353, 355, 357 emulation, 349 enclaves, 29 enclosure, 43 energy efficiency, 92, 279, 307, 323, 324 energy-saving technology, 287 enfeoffment, 36 Engelberger Joseph Engelberger, 154 Engels Friedrich Engels, 447, 458 engine atmospheric engine, 92 direct-acting engine, 84 double-acting rotative engine, 76 jet engine, 90 oscillating engine, 84 Otto cycle engine, 92 trunk engine, 84 Woolf high-pressure compound engine, 76 English East India Company, 42 Enheduanna, 27 ENIAC, 133, 196, 460 Enki, 23

Index entertainment, 74, 108, 137, 156, 158, 159, 176, 192, 207, 231, 233, 235, 246, 259, 261, 282, 300, 338, 341, 396, 450 entrepreneur, 69, 73, 99, 234, 235, 244, 259, 340, 349, 368, 372, 388, 389, 396, 413, 445, 447 entropy, 327 Enûma Eliš, 34 environmental noise, 332 environmental protection, 335 eosin, 98 Epipaleolithic Age, 13 equality, 90, 209, 213, 349, 383, 386, 390, 393, 394, 452, 456 of opportunity, 350 of outcome, 350 of permission, 350 Eratosthenes, 435 E-readers, 283 Ericsson John Ericsson, 153 Eridu, 17, 23 Erligang, 27, 30 Erlitou, 23, 26, 30, 55 Erode, 33 Ertebølle, 15 Estakhri, 36 esteem, 205, 209, 232, 234, 397, 398, 399 Ethernet, 172 ethical consciousness, 233 ethical consumption, 233 ethical dilemma, 406, 407, 411 ethical principles, 407 euphoria, 230, 429 Eureqa, 178, 294 Europe Container Terminals. See ECT Evans Oliver Evans, 76, 120, 125, 127 Everett Robert Everett, 134 Evolution Robotics, 157 evolutionary algorithm, 150, 200

Economics and Politics in the Robotic Age: The Future of Human Society evolutionary computation, 150, 151, 460 exchange theory of value, 255 executive power, 214, 406 existential risk, 420 expanded polystyrene, 100 experimental economics, 222 expert system, 143, 150, 199 expertise-intensive, 367 export-led growth, 368 external costs, 222 external economies of scale, 369, 370 externality, 216, 334 externalization, 3, 14, 252, 261 extra-marital sex, 399 Eyde Sam Eyde, 99 fabric, 17, 19, 66, 71, 159, 353 Facebook, 232, 283, 292, 335, 336, 358, 370, 373, 374 face-to-face interaction, 288 Factories Act, 88 Fairchild Semiconductor, 135 Faiyum A, 19 family pet, 302, 390, 401 family robot, 401, 403 FAMULUS, 155 fan, 206, 337, 338 Fan Bingbing, 337 FANUC, 159, 268 Far East, 41, 42, 43, 45 farming, 16, 17, 19, 23, 31, 37, 38, 43, 44, 45, 48, 49, 54, 60, 67, 173, 186, 212, 253, 261, 264, 278, 280, 321, 322, 330, 354, 355, 357, 364, 438 farming tool, 31, 43, 311 Fawcett Eric Fawcett, 100 FDIs, 452 fear, 188, 230, 407, 455 feedback control, 124, 128, 129, 130, 193 Feigenbaum Edward Feigenbaum, 143

477

Fenggu. See strength of character fermentation, 25, 279, 322 fermenter, 323 Ferranti, 94, 133, 136, 200 Sebastian Ziani de Ferranti, 94, 122 Ferranti-Thompson Alternator, 94 Ferraris Galileo Ferraris, 96 fertility cult, 20 fertility rate, 378 fertilizer, 42, 97, 98, 99, 278, 279, 322, 329, 331, 334 fertilizer industry, 99 figurines, 13, 14, 20, 21, 23, 152 financial technology. See Fintech finery forge, 79 finiteness of consumption, 364 Fintech, 291 Firebee I, 180 firebrick, 80 First Dynasty of Babylon, 30 First Dynasty of Lagash, 27, 29 first oil shortage, 108 First Persian Empire, 31 fish, 11, 14, 21, 156, 331 fishhooks, 14 fishing nets, 19 Fison James Fison, 99 fissile material, 325 Fitch John Fitch, 83 fixed costs, 277, 278, 282, 285, 374 FL, 151, 425, 460 flapper nozzle amplifier, 126 flash lock, 36 flatterer, 405, 406 flax, 16, 66, 71, 87, 353 Fleming John Ambrose Fleming, 102, 112 flexible manufacturing, 282, 285, 446 flexible manufacturing system. See FMS FLI, 419

478 flint, 9, 11, 12, 18, 19, 51 Flores, 11 flour, 19, 125, 127 flour mill, 125, 127 Flowers Thomas Harold Flowers, 133 flyer-and-bobbin system, 70 flying shuttle, 69 FMS, 285 Fodor Jerry Fodor, 419 Foljambe, 43 Ford Henry Ford, 104, 127 Ford Motor, 104, 127, 160 Fordson, 106 forest management, 280, 330 forging, 33, 40 forklift, 128 formaldehyde, 100 Forrester Jay Forrester, 134 Forth and Clyde Canal, 84 Forth Banks Power Station, 95 fossil fuels, 286, 326 Foster-Miller, 179 founder crops, 16 Four Doctors, 438 Fourth Arab-Israeli War, 108 Fourth Dynasty, 28 Fourth Industrial Revolution, 108 Foxboro Instrument Company, 126 Foxconn, 162, 261, 397 framework knitting, 88 Frankish kingdom, 435 fraternity, 213 Freddy, 155 Freddy II, 155 Frederick I Barbarossa, 436 free market, 369 free radicals, 329 freedom, 155, 209, 312, 374, 401, 428, 437, 451 freedom of speech, 431 French Revolution, 89

Index fresco bull-leaping fresco, 29 freshwater, 14, 323 Friedman Milton Friedm, 382 Friedrich Bayer et Compagnie, 98 FRIEND, 172, 175, 460 Friendly Society of Agricultural Laborers, 89 friendship, 209, 216, 232, 233, 234, 260, 397, 398 Frost James Frost, 82 fruit-picking robot, 173, 278 Fugaku, 136, 191 Fuji Yusoki Kogyo Company, 128, 155 Fujitsu, 136, 187 full agonist, 314 Fulton Robert Fulton, 84 functional consumption, 224, 246 functionally novel, 244, 261, 354, 355 fungicide, 329 Funnel-beaker, 17 furnace, 79, 80, 330 furniture and decoration space, 227 Fusarium venenatum, 323 fustian, 71 Future of Life Institute. See FLI futurologist, 301 fuzzy logic, 128, 130, 151 Gakutensoku, 153, 188 Galatea, 152 gambler psyche, 317 garments, 12, 17 Garry Kasparov, 183 gas lighting, 82 gas turbine, 105 gasoline, 92, 106 gastrointestinal tract, 220, 227 Gates Bill Gates, 220, 240, 246, 307, 344, 350, 419, 426 Gaza, 33

Economics and Politics in the Robotic Age: The Future of Human Society gazelles, 14 Gazzadini Bettisia Gazzadini, 438 general employment-enhancing technology, 353 general employment-reducing technology, 353 general intelligence, 141, 148, 149, 181, 182, 301, 302, 318, 319, 320, 390, 401, 402, 403, 418, 429 General Motors, 154, 160 general rationality. See total rationality general-purpose learning algorithms, 295 Generative Pre-trained Transformer. See GPT genetic modification, 321 genetic programming, 177, 178, 300 Genghis, 156 geometry, 433, 434 German Imperial Navy, 125 Gerzean, 25, 28 Gerzeh, 32 Gesner Abraham Gesner, 91 Gibson Reginald Gibson, 100 Gilbert William Gilbert, 93, 119 Gilgamesh, 29, 34 Giza, 28 Global Hawk, 157, 201 global warming, 286, 325 globalization, 92, 120, 348, 360, 361, 366, 368, 369, 370, 377, 381, 385, 386, 393, 394, 418, 431 Gloire, 104 Goats, 16 goblet, 25 Goddess Temple, 23 golem, 152 Gonghe Regency, 36

479

Goodyear Charles Goodyear, 99, 115 Google, 144, 158, 167, 170, 178, 182, 183, 263, 283, 293, 295, 335, 336, 358, 370, 373, 374, 418, 420, 440 Gopher, 140 gorillas, 2 gourds, 17 government, 219, 300, 319, 322, 380, 381, 408, 410, 412, 414, 415, 453, 454 intervention, 384 official, 301, 341 policies, 383, 384, 431 subsidies, 326, 368 governor, 124, 129 GPT, 170, 177, 194, 460 GPT-3.5, 170 GPT-4, 170 grammar, 294, 433, 435 Gramme Zénobe Gramme, 93, 95 Gravettian, 12, 13 Gray John McFarlane Gray, 125 Great Britain, 61, 84, 110, 111, 120 Great Eastern, 102, 125 Great Fish River, 41 Great House, 22 Great Moderation, 384 Great Orme mine, 25 Great Pyramid, 28 Great Wall Motors, 161 Great Western, 84 Great Western Railway, 101 Greco-Persian, 35 Greek Dark Ages, 33 green energy, 286, 326 Gregorian Reform, 435 Gregory IX, 437 Gregory VII, 435 Grimshaw Robert Grimshaw, 72 grinding stone, 14 grout, 82

480 growth agents, 322, 329 Guan Zhong, 36, 397 guano, 42, 99 guaranteed basic income, 123, 315, 382, 384, 387, 388, 390, 412, 453, 456 guaranteed income, 245, 334, 377, 383, 388, 396, 397, 400, 401, 415, 417 Gui River, 36 guided missile, 153, 179 guild, 436, 438, 439, 449 guild of Masters, 436, 437 guilds of students, 436 guinea pig, 18 Guitarrero Cave, 18 gunpowder, 40 Gupta Empire, 37 gustatory, 208 gut feeling, 428 Gutenberg, 40, 57 Gutians, 30 GWM, 161 gyrocompass, 125 gyroscopic heading, 126 gyroscopic stabilizer, 126 Haarlem Mill, 72 Haber Fritz Haber, 99 Haber-Bosch process, 99 Halaf, 17, 20, 21 Halfan, 14 Hall Evelyn Beatrice Hall, 431 Hallstatt, 33 hallucination, 230 hammerstone, 6 Hammurabi, 30 Han Dynasty, 37, 39, 40, 55 Han Xin, 317, 345 Hancock Thomas Hancock, 99 hand axes, 8 hand loom, 69 Handy 1, 175

Index Harappan Harappan Civilization, 27 Middle Harappan Phase, 27 Hargreaves James Hargreaves, 70 Harifian, 15, 59 harpoons, 11, 14, 19, 264 Harvard University, 132, 305, 393, 424, 425, 458 Harvest Automation, 173 Harwell CADET, 134 Hassuna, 18 Hattic, 32 Hawking Stephen Hawking, 307, 344, 419, 426 Hazen Harold Locke Hazen, 131, 185 headstalls, 25 Health and Morals of Apprentices Act, 87 hearing loss, 332 heavy metal, 330, 331, 333 heavy-wheeled northern European plow, 43 Hebrew prophet, 37 Heckscher Eli Heckscher, 361 Heckscher-Ohlin model. See H-O model hegemon, 36 Heinkel He 178, 105 Hellenistic period, 35 hematite, 12, 97 hemp, 17, 19 Hemudu, 19, 20, 21 Henry the Navigator, 41 Hephaestus, 152 herbicide, 199, 329, 331 Hermel, 18 Hero of Alexandria, 74, 152, 200 Hertz Heinrich Hertz, 102 heterocyclics, 331 Hetton Colliery railway, 85 heuristic search, 144, 150

Economics and Politics in the Robotic Age: The Future of Human Society Hewitt Peter Cooper Hewitt, 179 Hewitt-Sperry pilotless Automatic Airplane, 179 Hewlett Packard Enterprise Frontier, 136 hierarchical division of labor, 266 hierarchy of needs, 209, 228, 248 hieroglyphs, 28, 34, 45 High Middle Ages, 432 High Mobility Multipurpose Wheeled Vehicle. See HMMWV high school, 433 high-throughput screening, 159, 178, 284, 285 high-value-added goods, 367, 368 hill climbing, 150 Hippocrates, 435 Hippomobile, 92 historians, 32 historical record, 32, 36 history ancient history, 32 modern history, 32, 40, 186 HMMWV, 168, 460 H-O model, 361 hoe, 19, 44 Hoechst AG, 98 Hofmann August Wilhelm von Hofmann, 98 home appliances, 107, 262, 296 home automation, 171, 172, 295 home production, 220, 236, 248, 262, 304 homemaker, 262, 282 homeostasis, 273, 304, 405, 423 Homer, 29, 32 hominians, 2 Hominidae, 2, 52 hominin, 2, 6, 7, 9, 48, 51, 58 Hominina, 2 Homininae, 2 Hominini, 2 hominoid, 2, 52 Hominoidea, 2

481

Homo Homo erectus, 7, 8, 9, 11 Homo erectus lantianensis, 9 Homo ergaster, 7 Homo floresiensis, 11 Homo habilis, 2, 7 Homo sapiens, 2, 9, 12, 301 homosexual, 399 Honda, 156 Hoover Herbert Hoover, 214 hopper boy, 127 horizontal division of labor, 266, 267 hormone, 229, 329 Horrocks Samuel Horrocks, 72 horse, 17, 25, 35, 44, 45, 71, 77, 83, 85, 106, 242 Horseley Ironworks, 84 horseshoe, 39 horticultural robot, 174, 279 hosiery, 88 hot blast, 80 Houli, 16 House of Commons, 89 household energy sufficiency, 287 houses, 15, 22, 23, 205, 244, 260, 328, 357, 365, 375 housing industry, 14 Hudson River, 84 Hughes David Edward Hughes, 102 Hugo de Porta Ravennate, 438 human capital, 236, 256, 257, 258, 305, 432 human community, 318, 430 human intelligence, 123, 131, 140, 141, 142, 144, 145, 147, 181, 182, 252, 255, 267, 301, 302, 389, 395, 406, 418, 425 human resources, 267 human touch, 441 human-replacing technology, 252, 389 human-robot collaboration, 161

482 humiliation, 230 Hundred Schools of Thought, 36 hunter-gatherer, 4, 6, 14, 15, 16, 22, 211 Huntsman Benjamin Huntsman, 80 Hurwitz Adolf Hurwitz, 129 husbandry, 44, 280, 321, 322 Hyatt John Wesley Hyatt, 100 hydraulic, 37, 77, 81, 124 hydroelectric power station, 94 hydroelectricity, 325 hydrogen, 92, 99, 110, 287, 326, 342 hydropower, 287 hymns, 27 hypertension, 332 hypertext, 139 hypothecated taxation, 334 hysteresis, 130 I Ching, 35, See Book of Changes Iambus, 300 Iamus, 176, 300 IBEX, 174 IBM 608, 135 IBM 7030, 135 IC, 112, 135, 460 ICI, 100 ICT, 137, 174, 176, 208, 246, 297, 298, 300, 312, 313, 319, 352, 356, 375, 376, 377, 385, 409, 412, 431, 443, 460 ideograms, 28 IFS, 279 IKBS, 143, 461 Iliad, 29 Imhotep, 28 Immigrant, 378, 379, 414 IMP, 139 Imperial Bureau of Astronomy, 434 Imperial Chemical Industries, 100 Inanna, 27 incandescent light bulb, 94 incinerator, 330

Index incremental product innovation, 418 independent thinking, 310, 314 Indian Ocean, 41 Indigenous Australians, 11 indigo dye, 98 individual capacity to consume, 220, 225, 232, 239 individual consumption capacity, 245 individual rights, 454 inductive logic programming, 149, 193 inductor, 133 Indus valley, 19, 22 Indus Valley Civilization, 27 industrial automation, 159, 201, 268, 306 industrial cluster, 270 industrial engineering, 106 Industrial Revolution, 2, 4, 64, 65, 67, 69, 76, 78, 80, 82, 86, 87, 88, 89, 90, 108, 123, 244, 260, 282, 325, 351, 352, 353, 354 First Industrial Revolution, 74, 101, 109 industrial robot, 151, 158, 159, 160, 162, 183, 199, 252, 284, 402 industrial robotics, 268 industry, 213, 352, 353, 359 catering industry, 290 computer industry, 128 cottage industry, 38, 66 electrical industry, 127 food industry, 10, 357 manufacturing industry, 269, 288, 329, 358 textile industry, 44, 65, 67 industry standard, 127 inequality of outcome, 350 inequality-promoted growth, 349 infant industry, 365 Inflexible, 94 information acquisition, 108 information advantage, 336, 337

Economics and Politics in the Robotic Age: The Future of Human Society Information Age, 107, 108, 123, 397 information and communications technology, 208, See ICT information highway, 138 information overload, 314 Information Processing Techniques Office. See IPTO information revolution, 5, 107 information technology, 108, 137, 235, 265, 356 information tool, 264 informed decision, 430 infrastructure, 26, 82, 85, 95, 140, 180, 243, 245, 254, 280, 281, 380 injectable knowledge solution, 420 in-line palletizer, 128 innovation, 35, 37, 43, 65, 67, 68, 69, 78, 101, 109, 192, 197, 277, 278, 282, 283, 314, 315, 353, 355, 357, 389, 413 innovator, 340 inspiration, 312, 315 Institute for the Future, 420 Institute of Automation Technology, 172, 460 instrument, 64, 104, 128, 130, 152, 257, 264, 282, 375, 456 instrumental convergence, 420 insufficient demand, 384 insurance, 65, 96, 171, 400 intangible asset, 253, 370, 371, 372, 374 intangible property, 214 integrated circuit, 162, 193, See IC Intel 4004, 135, 184 intellect, 141, 313, 362, 421, 450 intellectual, 32, 308, 430 intellectual property, 68, 117, 243, 348, 371 intellectual property right, 371 intelligence, 141, 151, 155, 320 intelligence provider, 255 intelligent agent, 141, 423 intelligent aquafarm system, 281

483

intelligent factory system, 285 Intelligent factory system, 286 intelligent farmland system, 279 intelligent forestry system, 280 intelligent highway system, 289 intelligent manufacturing, 285 intelligent system, 141, 264 intelligent transportation system, 166, 289 intensive husbandry, 280 interchangeable parts, 76, 77, 78 interest rate, 301, 375 Interface Message Processor. See IMP Intermodal Surface Transportation Efficiency Act. See ISTEA internal combustion engine, 90, 92, 96, 240, 258, 264 internal costs, 222 internal environment, 273, 405 internal space constraint, 227 International General Electric, 101 International Harvester Farmall, 106 international investment, 368 International Organization for Standardization. See ISO International Space Station, 157, 158, 461 international trade, 42, 65, 104, 105, 250, 348, 360, 361, 362, 364, 365, 366, 367, 370, 377, 378, 381, 382, 386, 393 internet, 107, 193, 198, 248 Internet celebrities, 337, 338 Internet Explorer, 140 Internet firm, 243, 335, 358, 359, 370, 371, 372, 373, 374 Internet forum, 313, 371 Internet of Things, 172, 279, 322, 452 Internet service provider. See ISP Internet shopping, 215, 329 interpersonal skill, 399 intertemporal optimization, 260 intimate relation, 206, 211, 230, 232 intrinsic activity, 313, 314, 342

484 intuition, 222, 291, 428 invention, 37, 45, 65, 68, 72, 77, 109, 120, 127, 248, 264, 315, 318, 389, 392, 427 inventor, 68, 74, 340, 396, 421, 444 investor, 42, 43, 375 Ionia, 35 IoT, 172, 190, 279, 280, 281, 296, 322, 324, 334, 452 IPTO, 138, 143, 461 IR 6/60, 155 Iraq ed-Dubb, 16 iRobot, 157, 179, 190 iron, 4, 31, 32, 33, 34, 35, 78, 97, 98, 104 cast iron, 37, 43, 78, 79 charcoal pig iron, 79 coke pig iron, 79 pig iron, 79, 80, 96 sponge iron, 32 wrought iron, 32, 33, 34, 37, 78, 79, 80, 85, 103 Iron Age, 4, 5, 31, 32, 33, 34, 60, 61 Iron Bridge, 79 iron ore, 31, 32, 34, 368 iron rails, 85, 103 iron tool, 31, 34, 38, 45 ironclad warships, 104 iron-making, 33, 38, 64, 78, 244 irrigation, 17, 25, 36, 37, 39, 62, 199, 322, 333 canal irrigation, 17 Islamic world, 38 ISO, 159, 164 i-sobot, 158 isolated system, 327 ISP, 139 Israelites, 34 ISTEA, 166, 461 Isthmus of Panama, 41 IT, 108, 110, 137, 143, 176, 199, 235, 356, 403, 421, 461 Italian food, 42 ITS, 166, 461 ivory, 11, 12, 13, 20, 152 J. Lyons & Co, 134

Index Jacobi Moritz von Jacobi, 95 Jacobus de Boragine, 438 jade, 20, 21, 23, 28, 63 James Coxon, 94 Japanese National Agricultural Research Centre, 174 Jarmo, 18, 45 Java Man, 7 jealousy, 429 Jebel Irhoud, 9, 53 Jedlik Ányos Jedlik, 93 Jemdet Nasr period, 25, 27, 29 Jennings Ken Jennings, 144, 183 Jeopardy, 144, 148, 174, 175, 183, 297, 390 jet fuel, 90 jewelry, 12, 357 Jikji, 40 Jin Dynasty, 39 Jindivik, 179 Jingdong, 165, 371 John Deere, 106 joint-stock, 43, 68 Jǀmon, 15 Jordan River valley, 14 Joseph Sister Miriam Joseph, 433 joy, 230, 443 judiciary, 454 jujube, 17 justice, 90, 209, 213, 392, 415 Kabaran, 13 Kalidasa, 37 Kanade Takeo Kanade, 155, 184 Karakuri ningyǀ, 152 Kasparov Garry Kasparov, 144 Kassites, 30 Kay John Kay, 70, 122 Kebarian, 13

Economics and Politics in the Robotic Age: The Future of Human Society Kekulé Friedrich August Kekulé, 315 Kennedy John F. Kennedy, 214 Kentucky Fried Chicken. See KFC kerosene, 90, 91, 92, 110 kerosene lamp, 91 Kettering aerial torpedo, 179 key performance indicators. See KPI Keynes John Maynard Keynes, 219 KFC, 164, 461 Khafra, 28 Khan Academy, 175 Khiamians, 16 Khormusan, 12 Khufu, 28 Kier Samuel Martin Kier, 91 Kilburn Tom Kilburn, 133, 134, 183 Kilby Jack Kilby, 135 kilns, 20 Kindle, 283 Kinect, 145 King Mu of Zhou, 152 kinship, 396, 424 KIPS, 135, 461 Kirk Alexander C. Kirk, 104 Klimonas, 22 Kline Charley Kline, 139 Knossos, 20, 22, 26, 30, 51 knowledge insufficiency, 410 knowledge representation, 146 knowledge worker, 176 Koban, 33 Kodak, 100 Kodumanal, 33 Kopais basin, 26 KPI, 312, 451 Kramnik Vladimir Kramnik, 144 Krugman

485

Paul Krugman, 365 Kuan Chung. See Guan Zhong KUKA, 155, 159 Kura sushi, 163, 165, 166 Kura-Araxes, 25 Kurzweil Ray Kurzweil, 419 Kyongju, 40 La Tène, 33 labor, 252 labor income, 353, 355, 356 labor productivity, 4, 23, 25, 32, 45, 106, 127, 165, 224, 240, 241, 259, 260, 267, 275, 288, 349, 353, 354, 355, 358, 377 labor supply, 220, 307, 339, 366, 378 labor supply curve, 220 laboratory test, 171, 299 laborer, 44, 88, 89, 301 labor-intensive, 97, 367, 369 lace, 88 lacquer wood, 20 Lafferty Don Lafferty, 144 LaFrance Richard LaFrance, 127 Lahuradewa, 33 LaMDA, 170 lamp oil, 90, 91 land conversion, 43, 44 land improvement, 43 Langen Eugen Langen, 92 Language Model for Dialogue Applications. See LaMDA Lantian Man, 9 Lao Tse. See Lao Zi Laozi, 36, 37 lapis lazuli, 22 LARC, 135, 461 large language model. See LLM large seal script, 35 large-scale ICs. See LSI late antiquity, 32 Late Bronze Age Collapse, 33

486 Late Harappan, 33 latent semantic indexing, 148 lathe, 77 slide rest lathe, 77 Latin, 34, 432, 434, 435 law enforcement, 283, 322, 349, 404 Lawes John Bennet Lawes, 98 Lawrence Livermore National Laboratory, 326 Lawson Harry John Lawson, 99 Lay John Louis Lay, 153 layer-forming platform, 128 Le Moustier, 9 lead chamber process, 80 learning fatigue, 227 least developed countries, 321, 322, 379 leather, 17, 25, 98 Leblanc Nicolas Leblanc, 81 Leblanc process, 81, 114 Lee Edmund Lee, 124 Lee Sedol, 145, 183, 317, 428 Leeds and Liverpool Canal, 83 Legal Services Act, 171 Legalist, 36 legumes, 14 Lengyel, 24 Lenoir Jean Joseph Etienne Lenoir, 92 lentils, 16 LEO I, 134 Leonardo's robot, 152 letters patent, 68 Levallois, 9, 47 Levant, 11, 12, 13, 16, 21, 22, 30, 46, 49, 50, 51, 53, 55, 59 Leyla-Tepe, 24 Lézignan-la-Cèbe, 8 Li River, 36 Li Ziqi, 337

Index Li Zuoju, 317 Liangzhu, 17, 19, 20, 21, 23, 58 Libbey Edward D. Libbey, 127 liberal arts, 433, 435, 437, 440 liberty, 213, 412, 454 Libido, 206 license to teach, 437, 438 Lichterfelde, 95 Licklider J. C. R. Licklider, 138 Liebig Justus von Liebig, 99, 120 life expectancy, 86, 87 life-cycle theory, 219 light water reactor, 325 Lighthill Michael James Lighthill, 142 lights out, 268 lights-out manufacturing, 268, 445 Lilienthal Otto Lilienthal, 105 limestone, 14, 22, 69, 81, 82, 97 line manager, 411 Linear A, 28 Linear B, 28 Lingqu Canal, 36 lingua franca, 34 Lipson Hod Lipson, 178, 185, 198, 306 Liquid Robotics, 180 liquidity constraints, 219, 247 literature, 32, 34, 37, 51, 178, 206, 264, 295, 392 lithic reduction percussion, 12 live commerce, 337 Livermore Advanced Research Computer. See LARC Liverpool and Manchester Railway, 85, 119 living conditions, 86, 410 living standards, 86, 324, 365, 366, 416, 417, 422 Livingston Robert Livingston, 84 llamas, 18

Economics and Politics in the Robotic Age: The Future of Human Society LLM, 170, 295, 461 loading robot, 269, 270, 271 local community, 286, 287, 401, 405, 444 local laboratory center, 298 Lock, 36 locomotive steam locomotive, 76 logistic and support robot, 269 Lombe John Lombe, 86 London Electric Supply Corporation, 94 London Symphony Orchestra, 176, 300 Longqiuzhuang, 21 long-run average costs, 278 Longshan, 17 Lord Armstrong, 94, 110 lost wax method, 28 Lothal, 27 love, 209, 232, 233, 234, 298, 397, 398, 399, 422 Low Archibald Low, 153 Archibald M. Low, 179 Lowell Francis Cabot Lowell, 73 Lowell, Massachusetts, 73 low-income countries, 370 LSI, 135 Lu Ban, 152 lubricant, 91 lubricating oil, 91 Luddites, 87, 88, 352, 393 àukasiewicz Jan Józef Ignacy àukasiewicz, 91 Lullubi, 30 lump of labor fallacy, 256 lust, 429 Macadam road, 83 MacArthur Douglas MacArthur, 317 Macero Daniel J. Macero, 178

487

Machang, 26 machine, 64, 138, 144, 253, 267, 308, 407 intelligent machine, 141 Machine Age, 4, 64, 69, 123, 252, 255, 258, 267, 277, 338, 339, 370, 395, 402, 412, 417, 422, 445, 446 machine ethics, 407, 423, 425 machine flexibility, 285 machine intelligence, 123, 143, 196, 252, 407 machine language, 134 machine learning, 147, 190, 461 machine perception, 148 machine tool, 76, 77, 78, 90, 96, 104, 124 machine-operator relation, 447 Madeira, 41 Magdalenian, 12, 13, 46 Magellan, 41 magnesite, 97 magnesium, 96 magnetic field, 93 Magneto Works, 93 Mahabharata, 37 Mahavira, 37 maintenance technicians, 266 maize, 17, 42, 57 Majiayao, 26 Makino Hiroshi Makino, 155 malachite, 24 Malay Peninsula, 42 malevolence, 403 Malhar, 33 Malia, 30 malleability, 96 Maltravieso, 12 management buyout, 350 management consulting, 406 Manby Aaron Manby, 84 mandated international administration, 322 manganese, 96

488 Manhattan Project, 132 manioc, 42 Man-Machine Interface, 143 Manual Age, 2, 4, 5, 38, 44, 252, 253, 254, 255, 262, 266, 277, 309 manual power, 2, 4, 5, 255, 267 marble, 19, 21 Marconi Guglielmo Marconi, 102, 110 marginal cost, 175, 243, 277, 283, 335, 336, 372, 373, 374 marginal private costs, 334 marginal product of labor, 348, 358 marginal products of capital, 358 marine vessel, 330 Mark 1, 132, 133 market goods, 223, 337 Markin Rod Markin, 178 marriage, 397 Mars Pathfinder, 156 Martin Pierre-Émile Martin, 97 Martinus Gosia, 438 Marx Karl Marx, 447 Marxist, 397 Maslow, 397 Abraham Maslow, 209 mass media, 430 mass production, 36, 90, 92, 96, 104, 107, 115, 159, 282, 283, 285, 286, 313, 446, 450 Massachusetts Institute of Technology. See MIT Massive Open Online Course. See MOOC mast, 27 material-handling robot, 161 Maternal deprivation, 399 matrilineal families, 396 mattocks, 19, 50 Mauchly John Mauchly, 133

Index Maudslay Henry Maudslay, 77, 114 Joseph Maudslay, 84 Maupassant Guy de Maupassant, 224 mauveine, 98 Maxwell James Clerk Maxwell, 102, 129 Maybach Wilhelm Maybach, 103 Maykop, 25, 29 McAdam John McAdam, 83 MCC, 143, 461 McCarthy John McCarthy, 141, 142, 149, 184 McCormick Reaper, 106 McDermott John P. McDermott, 143 McDonald’s, 164 McGrattan Brian J. McGrattan, 178 means of life, 350 means-ends analysis, 150 measures, 27, 321, 369, 380 meat analog, 323 mechanistically novel, 242, 244, 261, 354, 355 mechanization of agriculture, 172, 354 mechanized agriculture, 259 medical care, 235, 434 medicine, 35 Medieval Warm Period, 39 Mediterranean Sea, 61, 229 Medway Estuary, 82 megalithic, 26 megaron, 30 Mehrgarh, 17, 20, 22, 53 Mehrgarh Period I, 17 Meikle, 44 Meituan, 165 mêlée weapons, 11 Melomics, 176 Mencius, 308, 342, 399

Economics and Politics in the Robotic Age: The Future of Human Society Menkaure, 28 mental power, 5, 131, 183, 264, 277, 447 mental strength, 308, 338 mercantilism, 65 Mercedes-Benz, 167, 168 merchant, 66, 102, 436 mercury, 126, 329 Meridiani Planum, 157 Merimde, 20 Merneptah, 33 Merovingian Kings, 435 Merrimack River, 73 Merthyr Tydfil, 76 MES, 278, 282 Mesolithic Age, 4, 5, 13 Mesopotamia, 17, 19, 23, 25, 29, 30, 54, 55, 57 metal movable type, 40 metallurgy, 24, 27, 33, 56, 58, 60, 104 Metalsmithing, 33 metalworkers, 23 meta-motivation, 397 meteoric iron, 32, 33 methane, 329, 331 methaqualone, 231 methodological needs, 235 methylene blue, 98 Michaux Pierre Michaux, 103 microcomputer, 135 Microelectronics and Computer Consortium. See MCC microliths, 8, 11, 14 microprocessor, 130, 135, 154, 156, 184 Microsoft, 263, 283, 371, 419 Middle East, 7, 41, 46, 141, 210 Midjourney, 177, 300 Milesian School, 35 milk, 17, 81, 126, 165, 173 millennium bug, 356 Miller John Stuart Miller, 376 millet, 14, 16, 62

489

millet grass grains, 14 milling machine, 78 millstone, 125 millwrights, 76 Mincer Jacob Mincer, 256 Mines and Collieries Act, 88 Ming Dynasty, 40 minimum efficient scale. See MES Minoan, 26, 28, 29, 30, 52 Minorsky Nicolas Minorsky, 126, 129 Minsky Marvin Minsky, 142, 149, 154 Mint, 157 MIPS, 135, 461 MIT, 134, 135, 149, 154, 156, 193, 194, 196, 461 MMS, 137, 461 mobile phone, 136, 138, 140, 172, 280 Mobile Robot Laboratory, 156 Moche civilization, 24 Mode I tools, 6 Model T, 104 Modern Times, 265 modes of production, 395 Modigliani Franco Modigliani, 219 Mödling and Hinterbrühl Tram, 95 moldboard, 37, 39, 43 molecubes, 157 Momentum Machines, 163 monastery, 433, 434, 435 monastic school, 432, 433, 434, 435, 436 monetary budget line, 228 monetary policy, 301 money supply, 301 money transfer, 108 monk, 433, 435 monkey, 2, 207, 395 monotonicity, 221 Monte Carlo, 145 Monte Cassino, 434 Monte Poggiolo, 8

490 MOOC, 175, 193, 297, 461 Moore Gordon Moore, 135 Moore's law, 135, 419 mopping robot, 157 moral dilemma, 406, 407, 410 Morse Samuel Morse, 101 Morse code, 101 mortality rate, 42 mortars, 14, 120 Mosaic, 139 motivation, 3, 182, 246, 248, 249, 250, 251, 294, 318, 320, 327, 349, 350, 391, 397, 408, 418, 423, 424, 452 Motoman, 159 motor vehicle, 167, 330 Motorwagen, 103 Mousterian point, 9, 10 Mousterian tools, 9 movable type, 40, 57 Mozi, 152 MRI, 299 mudbrick, 22, 23 Mughal, 66, 111, 119 Mughalistan, 66 multifunctional mobile robot, 269, 445 multifunctional patrol robot, 296 multifunctional robot, 269, 270, 272, 284, 285, 287, 288, 290, 291, 292, 293, 295, 296, 299, 300, 303, 403, 441, 454 multifunctional robotic miner, 281 multifunctional service and maintenance robot, 183 multimedia messaging service. See MMS multinational, 43, 366, 367 multiple-refined steel, 36 multiplex switchboard, 102 mung, 16 Murdoch William Murdoch, 82

Index Murry James Murry, 99 muscle power, 44, 277 Mushet Robert Forester Mushet, 96, 103 musical instruments, 20, 29, 60, 332, 450 Musk Elon Musk, 307, 344, 350, 419, 420, 426 Muslim, 38, 41 muslin, 66 Mycenae, 26 Mycenaean, 26, 28, 30, 33, 52 mycoprotein, 323 NAHSC, 167, 461 Nanna, 27 Nanzhuangtou, 16, 19, 20 naphtha, 90 naphthalene, 331 Napier Charles Napier, 84 Naqada I, 24 Naqada II, 25 Naqada III, 29 Narmer, 29 Narrative Science, 169 NASA, 154, 155, 157, 426, 461 Nasmyth James Nasmyth, 77 National Aeronautics and Space Administration. See NASA National Automated Highway System Consortium. See NAHSC National Highway Traffic Safety Administration. See NHTSA National Ignition Facility, 326 nation-state, 414 Natta Giulio Natta, 100 Natufian, 14, 15, 16, 46 natural endowment, 7 natural gas, 324, 325, 327 natural language processing, 143, 147, 148 natural monopolies, 218

Economics and Politics in the Robotic Age: The Future of Human Society natural resources, 22, 157, 221, 224, 234, 252, 253, 256, 307, 321, 328, 334, 351, 365, 380, 416, 417 naviform core, 19 Navigation Laboratory. See Navlab Navlab, 168 Neanderthals, 10, 11, 58 Neapolitan cooking, 42 Near East, 4, 10, 25, 26, 31, 32, 33, 48, 49, 54, 55, 60, 141 necklaces, 13, 29 negative feedback, 125, 126 Negev Desert, 15 Neilson James Beaumont Neilson, 80 Nemea valley, 26 Neo-Hittite kingdom, 34 Neolithic Age, 4, 5, 19, 20, 338 Neolithic Revolution, 13, 16, 48 Neolithic signs, 21 nervous system, 3, 230 net national products. See NNP netizens, 337 Netscape Navigator, 140 network effect, 371, 374 networked database, 130 networking, 137, 283, 310, 313, 335, 374, 391 neural center, 229 neural networks, 145, 146, 151, 345, 458 neural receptor, 229 neutron, 325 New Lanark mills, 88 New World, 38, 41, 42, 45, 57, 65 Newcastle and District Electric Lighting Company, 95 Newcomen Thomas Newcomen, 74, 119 Newell Allen Newell, 142, 149, 198 newly industrialized countries, 368 Newton Isaac Newton, 315 NHTSA, 168, 461

491

nightlife, 82 Nineveh, 34 Nishimura Makoto Nishimura, 153 nitric acid, 99 nitrocellulose, 100 nitrogen, 39, 43, 99, 329, 330 nitrogen-fixation, 39 Niuheliang, 23 NNP, 376 noise, 216, 329, 330, 332 nomogram, 131 non-monotonic logic, 149 nonpathogenic bacteria, 314 non-satiation, 221, 260 Noria, 39 Norse mythology, 152 North River Steamboat, 84 Northrop Aircraft Company, 134 Northwestern University, 169 Norwegian Seismic Array. See NORSAR Novocherkassk, 33 Noyce Robert Noyce, 135 NRC, 155, 461 nuclear energy, 286, 325, 342 nuclear fusion, 326 nuclear magnetic resonance imaging. See MRI nuclear reprocessing, 325 nun, 433 Nylon, 100 Oak Ridge Leadership Computing Facility, 136 oars, 19 oats, 16, 39 obedience, 310, 342 obesity, 231, 247 obsidian, 21 oceanography, 180 OER, 175, 461 offspring, 396, 400 Ohalo II site, 14 Ohlin Bertil Ohlin, 361

492 oil field, 90 oil refinery, 91 oil shale, 91 Oil Springs, 91 Old Kingdom, 28 Oldowan, 6, 7, 9, 61 Olduvai Gorge, 6 olfactory, 208 Olympic Pantheon, 28 Omo Kibish, 9 onagers, 14 one-child policy, 322, 345 online shop, 176, 243, 261, 288, 289, 300, 336, 370, 371, 372 online shopping, 165, 288 online shopping platform, 336 online transaction processing system, 165 open educational resources. See OER Open University, 297 OpenAI, 170 open-field mining, 329 open-hearth furnace, 97 opinion leader, 368 opioids,, 231 Opportunity, 157, 186, 196 optimal allocations, 221 oracle bone script, 35 oracle bones, 28 orangutans, 2 organic farming, 321 ornaments, 12, 14, 20, 29 Ørsted Hans Christian Ørsted, 93 osmoceptor, 229 osmotic pressure, 229 Otto Nikolaus August Otto, 92 Otto engine, 92, 103 Ottoman Empire, 41 outer space mining, 328 Outremer, 65 outsourcing, 261, 369 overall marginal costs, 334 over-mining, 327

Index Owen Robert Owen, 88, 119 Owens Michael Joseph Owens, 127 own goods, 216 owner-manager, 275 owner-occupier, 219 ownership reform, 350 owner-worker, 271 oxen, 19, 25, 26, 37, 43 oxides, 32, 97, 329, 330 P2, 156 P2PU, 175 P3, 156 Pachomius, 433 Pacific Ocean, 41, 157 Pacinotti Antonio Pacinotti, 93 Packard Edward Packard, 99 Packbot, 179 packet-switching, 138, 139 packhorses, 66, 85 packing material, 329 paddle-steamer, 84 Painting Fool, 177, 300 palace, 22, 23, 29, 30 palace school, 435 Paleolithic Age, 4, 5, 6, 9, 10, 11, 13, 44, 338 pallet, 128, 270 palletizer, 128, 194 palletizing robot, 128, 155 PaLM 2, 170 Pangeng, 31 paper money, 40 papermaking, 37 Papert Seymour Papert, 149 Papin Denis Papin, 74, 83 Papplewick, 76 paraffin oil, 91 parallel division of labor, 266 parallel motion linkage, 76 parental care, 399

Economics and Politics in the Robotic Age: The Future of Human Society Parker James Parker, 82 Parkes Alexander Parkes, 100 Parkesine, 100, 119 partial agonist, 313, 314 partial decarbonization, 40 partial rationality, 222 particulates, 330, 331 partner-replacing robot, 402 patent system, 68, 117 pathogen, 331 pathogenic bacteria, 314 patrilineal families, 396 pattern recognition, 147 Paul Lewis Paul, 70 Pawtucket, 73 PBOC, 291, 292 PC, 135, 278, 355 PCB, 159, 162 PDA, 136, 461 Peace of Callias, 35 peanut, 17, 42 pearl, 66 Pearl Street, 94 peas, 16, 39 Peer-to-Peer University. See P2PU Peiligang, 16, 19, 22 Peking Man, 7 Peloponnese, 26 pendants, 13, 14, 21 Penn John Penn, 84 Penydarren Ironworks, 76 People’s Bank of China. See PBOC pepper, 17, 43 perceived well-being, 233 performance appraisal, 312 performance evaluation system, 312 performing android, 300 Perkin William Henry Perkin, 97, 119 permanent income, 219, 247, 430 permanent income hypothesis, 219, 247

493

Perreaux Louis-Guillaume Perreaux, 103 personal computer, 278, 355, See PC personal digital assistant. See PDA personal dignity, 315 personal property, 20 personal space, 388 adequate personal space, 358 effective personal space, 357 personality, 232, 239, 249, 256 pesticide, 42, 322, 329, 331 Peterloo Massacre, 89 petrol, 90 petroleum, 90, 91, 92, 324, 325, 328, 331, 334, 365 Petty William Petty, 266 Peugeot, 103 pharmaceutical, 98 pharmaceutical industry, 159, 284 Phénix, 326 phenol, 100 phenol-furfural resins, 100 Philips, 268 philosopher king, 411, 434 philosophy, 35, 36, 37, 141, 212 Phoenician, 34 Phoenix, 84, 158 phonograms, 28 phosphorus, 97, 99, 328 physical capital, 219, 253, 372 physical dexterity, 272 physical maneuvering, 272, 275 physical strength, 4, 71, 88, 123, 263, 264, 308, 311, 338 physical symbol system hypothesis. See PSSH physiological capacity budget line, 228 physiological capacity constraints, 227, 228, 252, 264 physiological constraint, 204, 220, 231, 237, 238, 239, 240, 245, 246, 358, 364

494 physiological needs, 209, 210, 211, 221, 228, 229, 230, 233, 321 physiology, 220, 239, 246, 405, 421 Picard Rosalind Picard, 182 Piccadilly Mill, 76 pick-and-place machine, 162 pickling, 80 pictographs, 27 PID controller, 126, 129 pier, 19 pigs, 16, 17 pipe system, 37 Pirotsky Fyodor Pirotsky, 95 pistachios, 14 Pixii Antoine Hyppolite Pixii, 93 planetoid, 328 planets, 29, 328 planing machine, 77 planning, 141, 145, 147, 149, 151, 156, 215, 262, 267, 292, 318, 413, 447 planter, 106, 279 Plaque of Ur Nanshe, 27 Plato, 35, 266, 411, 425, 434, 435, 458 PLC, 130, 461 pleasure, 83, 230, 238, 247, 249, 250, 397 Ploþnik, 24 plow, 19, 26, 39, 43 plowshare, 37 Pluralism, 35 plutonium, 325 pneumatic tire, 100 pneumatic, automatic temperature controller, 126 political instability, 322 Political stability, 322 pollutant, 329, 331, 332, 333, 334, 342 polyester, 101 polyethylene, 100, 101 Polygonum, 16

Index polymer plastics, 100 polystyrene, 100 polyvinyl chloride, 100 Poncelet Jean Victor Poncelet, 129 population aging, 378 Porifera, 229 porphyry, 22 postal service, 26 postclassical era, 32, 38, 40 potassium, 81, 98, 99 potassium chlorate, 81 potassium dichromate, 98 potato, 18, 42, 48, 55 potter's wheel, 20 pottery, 15, 18, 19, 20, 21, 28, 29, 30, 45, 47, 54, 57, 61, 62 Pottery Neolithic, 16, 20, 50, 461 potting and stamping, 79 pound lock, 40 pounders, 6, 36 power loom, 72 power sharing, 406 power steering, 166, 289 pozzolana, 81 PPNA, 16, 18, 20, 22, 46, 50, 461 PPNB, 16, 18, 19, 20, 21, 22, 50, 461 practical rationality, 222 precautionary saving, 219 Predator, 180 Pre-Mode 1 tools, 6 prepared cores, 10, 12 prepared-core techniques, 9 Pre-Pottery Neolithic A, 16, 461 Pre-Pottery Neolithic B, 16, 50, 461 pressure flaking, 12 price level, 366, 368, 375 price transparency, 371, 372 pricing scheme, 335 printed circuit board, 159 private goods, 216, 218 private ownership, 43, 395, 413 privatization, 383, 392 probabilistic reasoning, 151 problem-solving ability, 281

Economics and Politics in the Robotic Age: The Future of Human Society process innovation, 277 producer, 3, 266, 337 product inhibition, 314 production flexibility, 277 productive capacity, 204 productivity elasticity of employment, 356 productivity puzzle, 108, 240, 355, 391 professional organization, 449 professional service, 169, 292, 293, 456 professional societies, 441, 449 professionalization, 261 programmable computer, 131, 132, 133 programmable logic controller, 130 programmable universal manipulation arm, 155 projectile weapons, 11 property rights, 68, 243, 322, 348 proportional-integral-derivative, 125 prostitution, 207 prosumer economy, 446 prosumers, 286, 287, 446 protectionism, 67, 360, 364, 365, 367, 368, 369, 381, 386 Protector USV, 181 Protodynastic Period, 29 proto-writing, 20 Providence, 73 pruning, 150 psalmody, 434 PSSH, 149, 461 psychological needs, 12, 23, 210, 222, 230, 231, 232, 233, 239, 334 psychology, 141, 182, 239, 443 public goods, 216, 218, 234 public ownership, 384 Public Square, 94 public subsidies, 334 Public utilities, 296 puddler, 79 puddling, 79, 120 pure antagonist, 314

495

purpose in life, 212, 213, 307 purpose of life, 23, 204 Puskás Tivadar Puskás, 102 Putnam Hillary Putnam, 419 putting-out system, 66 Pygmalion, 152 Pylos, 33 Pylyshyn Zenon Pylyshyn, 419 pyramid, 28, 209, 210, 246, 248, 249 Pythagoreanism, 35 Qadan, 15 Qijia, 26, 48 Qin Dynasty, 36 QRIO,, 157 quadrivium, 433, 434 qualification problem, 149, 199 Quenching, 36 Quill, 169, 170, 176, 293, 300 quinine, 98 Quorn, 323 racloirs, 9, 18 Radical War, 89 radio broadcasting, 102 radioactive contamination, 326 radon, 331 Rafael Advanced Defense Systems, 181 rafts, 11 railroad accounting, 106 railway, 85, 86, 103, 113, 287 Railway Mania, 85 Raja Nala Ka Tila, 33 Ramayana, 37 rammed-earth, 31 Rancho Arm, 154 Rancho Electric Arm. See REA Rancho Los Amigos Hospital, 154 random optimization, 150 range of exposure, 335 rangefinder, 155 rational consumer choice, 220, 239, 245

496 REA, 154, 461 reaction chamber, 279, 323 reading fatigue, 227 real estate, 292, 369, 374, 375 real wages, 86, 113, 370, 376 Real-time Integrated Adaptive Motion Control, 156 Reaper, 180 receptacles, 19 recombinant DNA, 408 reconfigurable manufacturing system. See RMS recorded video lectures, 298 recursive self-improvement, 419 recycling, 307, 327, 328, 333 recycling age, 333 reductive manufacturing, 333 reference value, 128 Reform Act, 89, 119 refugee, 378, 379, 381, 417 regenerative braking, 95 regenerative furnace, 97 regime change, 380 regional human supervision and support center, 270 reinforcement learning, 147, 345, 458 reins, 25 Reiter Raymond Reiter, 149 relational goods, 216 relative concentration, 314, 315 relay logic, 127 remote working, 271 Renaissance, 40, 152, 457, 458 renewable energy, 286, 326 representative democracy, 409, 410 representative parliamentary democracy, 409 reproduction, 206, 207, 400, 401, 424 Republic of Venice, 41 reservoir, 39, 324 resistor, 133, 162 resource acquisition, 420 resource dispute, 417

Index respect, 209, 210, 213, 214, 232, 234, 311, 397, 398, 404 responsible citizen, 443, 454, 456 retail bank, 292 retail food industry, 164 retailer, 243, 290, 336, 337, 342, 371 reverberatory furnace, 78 reviewer, 294 rhetoric, 433 Rhodes, 30 Ricardo David Ricardo, 361, 393 rice, 15, 16, 17, 42, 62, 278, 305 rice-planting robot, 174 Richards William H. Richards, 153 Richmond Union Passenger Railway, 95 Riddings colliery, 91 Rifkin Jeremy Rifkin, 107, 286, 287, 405 Riken Center for Computational Science, 136 Riordan Richard Riordan, 214 Riwat, 7 RMB, 292, 337 RMS, 285 RO, 150, 461 road marking, 167, 289 roasting, 24 Roberts Alban J. Roberts, 153, 190 Richard Roberts, 72, 77 Robinson's Mill, 76 Robonaut 2, 158, 187 robot care-providing robot, 172 collaborative robot, 161 construction robot, 287, 288 maintenance robot, 166, 272 mining robot, 163 monitoring robot, 411 robotic civil servant, 402

Economics and Politics in the Robotic Age: The Future of Human Society self-conscious robot, 420 welding robot, 155, 254 robot-human cooperation, 277 Robotic Age, 5, 123, 159, 224, 245, 252, 255, 348, 369, 395, 397, 427, 429, 452 robotic barista, 165 robotic chef, 290, 291 robotics, 108, 148, 151, 160, 176, 252, 268, 280, 284, 303, 319, 352, 357, 370, 375, 377, 382, 390, 427, 442, 445 robot-robot collaboration, 161 robot-robot cooperation, 277 RoboTuna, 156 Rochester Nathaniel Rochester, 142 rock art, 12 Roebuck John Roebuck, 80 role-playing, 341 roller spinning frame, 70 rolling process, 79 Roman cement, 82, 115 Roman Empire, 5, 35, 65, 435 Roman Republic, 35 Romney Mitt Romney, 214 Roomba, 157, 164, 190 Roper Sylvester H. Roper, 103 rotary machinery, 76 Routh Edward John Routh, 129 routing flexibility, 285 Royal Australian Air Force. See RAAF Royal Mail, 85 RS-485, 172 rubber natural rubber, 99 rubber industry, 99 Rubin Charles T. Rubin, 403 rudder, 125, 126 Rudna Glava, 24

497

Rudnik mountain, 24 Rule of St. Benedict, 433, 457 Rule of the Master, 433 ruler-manager, 301 Rumsey James Rumsey, 83 Rutter Brad Rutter, 144, 183 RV 100, 173 rye, 14, 16, 39 SA, 10, 70, 71, 79, 150 sacrament, 435 Sadler Committee, 87 sail, 19, 27, 84 SAIL, 154, 461 Salerno Medical School, 434 saltpeter, 99 SAM, 154, 461 Samarra, 17, 21 Samuel Arthur Samuel, 142 Samuelson Paul Samuelson, 362, 364, 365 Sanlongdong, 10, 59 Sanskrit, 37 Saône, 83 sarcophagus, 34 Sargon, 25, 27 Sasaki Masahide Sasaki, 177 satori, 316 saturation, 239, 241, 244, 261, 353, 365, 366, 377, 384 Savery Thomas Savery, 74 scale-up, 252 SCARA, 155, 162, 193, 462 scarce factor, 361 Schaeffer Jonathan Schaeffer, 144 Scheinman Victor Scheinman, 154, 197 Scheliha Victor von Scheliha, 153 Schmidt Michael Schmidt, 178

498 scholar-bureaucrats, 294, 439 Schützenberger Paul Schützenberger, 100 Schwarzenegger Arnold Schwarzenegger, 214 Schwenzfeier C. William Schwenzfeier, 127 SCI, 143, 462 Science Research Council. See SRC scientific community, 314 scientific knowledge, 435 scientific management, 106 scientific revolution, 67 scientist, 128, 129, 177, 179, 293, 294, 326, 340, 399, 421, 444, 451 Scottish Insurrection, 89 scrapers, 6, 8, 10, 13, 18 screw propeller, 84 sculpture, 37 SDR-3X, 156 SDR-4X, 156 SDR-4XII, 157 sea trade, 27 search algorithm, 150 search engine, 283, 335 Searle John Searle, 418, 419 seawater, 323, 324 Second Industrial Revolution, 64, 90, 92, 96, 101, 106, 109, 122, 244, 261, 310, 311, 339, 354, 356, 365 second law of thermodynamics, 327 Seditious Meetings Act, 89 seed drill, 44 seeders, 279 Seine, 84 Selandia, 105 selective breeding, 43 Selective Compliance Assembly Robot Arm. See SCARA self-acting mule, 72 self-actualization, 209, 210, 220, 234, 245, 262, 265, 397, 399, 404

Index self-awareness, 408, 418, 420 self-checking machine, 164 self-checkout system, 164 self-consciousness, 251, 320 self-driving car, 271 self-esteem, 209, 210, 213, 397, 398, 404 self-excitation, 93 self-fulfillment, 208, 209, 213, 234 self-media, 337 self-preservation, 420 self-production center, 446 self-production workshop, 446 self-propelled anthropomorphic manipulator. See SAM self-propelled mechanical harvester, 106 self-reinforcing, 231, 375 self-service machine, 272 self-transcendence, 209, 220, 234, 245, 265, 399, 404 semi-autonomy, 408 semiconductor, 107, 143 semiotic, 137 semi-prepared food, 262 senior manager, 266, 267, 271, 272, 273, 274, 275, 276, 303, 349, 447, 448 sensory modalities, 208, 231, 256 septaria, 82 sequential control, 128 serial division of labor, 266 service robot, 164, 201, 259, 269, 271, 272, 288, 290, 295, 320, 401, 402, 403, 405, 413 service sector, 107, 241, 242, 244, 246, 259, 262, 288, 339, 373 servomechanism, 125 Sesklo cultures, 20 Sestroretsk, 95 set point, 129, 130 sewage, 329 sewing machine, 282 sex industry, 207 sexual liberation, 206 sexual relationship, 206, 213

Economics and Politics in the Robotic Age: The Future of Human Society sexual reproduction, 400, 401 sexual revolution, 206, 248, 250 Sha'ar HaGolan, 21, 22 shadow value of time, 225 shaft straighteners, 14 shaft wrenches, 12 Shakey, 154, 191 shale, 91 Shang, 28, 31, 33, 48, 60, 390, 426 Shang Yang, 411 Shannon Claude Shannon, 132, 137, 142 shaping machine, 77 sheep, 16, 17, 44, 67, 172, 173 shells sea shells, 13, 22 shelter, 13, 205, 229 Shen Kuo, 40 Shepherd, 18 Shibuya Seiki, 173 Shijiahe, 20, 21, 23, 26, 63 shipbuilding, 38, 69, 104, 125 shop assistant, 258, 288, 372 shopping mall, 288, 289, 336 shrines, 30 Shuidonggou, 12 Shuttle Remote Manipulator System. See SRMS sibling bonding, 399 sickle blades, 14 sickles, 15, 19 Siemens Charles William Siemens, 97 Werner von Siemens, 93, 95, 113 Siemens dynamo, 94 signal theory of education, 441 signature image processing. See SIP silk, 17, 39, 41, 65, 66, 86, 98, 100, 109 silkworm, 17 Silver David Silver, 155 Silver Arm, 155 SIM, 137, 462 Simon Herbert Simon, 142, 149

499

simulated annealing, 150 Sinai, 13, 34 single-parent families, 399 single-task construction robot. See STCR single-task robot, 296, 303 singular value decomposition. See SVD singularity, 301, 302, 419, 424, 426 SIP, 160, 462 Six Acts, 89 slash-and-burn, 34 Slater Samuel Slater, 73 Slater Mill, 73 sleep disturbance, 332 slide rules, 131 Small-Scale Experimental Machine. See SSEM smart city, 296, 412 smart farms, 322 smart forestry system, 280 smart home, 171, 295 smart residential building, 334 smartphone, 136, 137, 140, 145, 242, 278, 289 SMD, 162 Smeaton John Smeaton, 81 Smeaton's Tower, 81 smelting copper smelting, 24, 31 Smith Adam Smith, 438 Adam Smith, 254, 257, 299 Adam Smith, 439 Adam Smith, 439 Adam Smith, 439 Adam Smith, 439 Adam Smith, 439 Adam Smith, 440 Adam Smith, 442 Francis Pettit Smith, 84 Smithfield, 44 SMT, 162, 462 social animal, 207, 318

500 social benefits, 340, 360, 414 social consciousness, 397 social convention, 206, 229, 350 social formation, 395 social intelligence, 141, 148, 181, 182 social organization, 395 social production, 262 social relationship, 395 social science research, 315, 376 social stability, 40, 322, 398 social status, 205, 207, 211, 212, 213, 214, 232, 236, 302, 341, 349, 406 social welfare system, 340, 350, 351, 378, 389, 412, 417 sociality, 205 socially responsible consumption, 233 societal utility function, 362 sociological needs, 232, 233 Socrates, 35, 37, 48 sodium carbonate, 80, 81 sodium hydroxide, 96 sodium sulfate, 81 SOE, 349 soft computing, 151 soft robotics, 159 Soho Foundry, 75, 82 soil contamination, 331 soil drought level, 279 soil pollution, 330, 331, 332 Sojourner, 156 solar energy, 287, 323, 326 solar radiation, 326 Solon’s reform, 35 Solutrean, 12 Solvay Ernest Solvay, 81 Solvay process, 81, 116 somatesthesia, 230 Somers Thomas Somers, 73 Song Dynasty, 40 Sony, 156, 176 Sony Dream Robot, 156

Index Sophists, 35 sound pollution, 332 soy, 16 soybean, 42, 173 Spa Fields Riots, 89 space budget line, 228 space entitlement, 357 Spar Aerospace, 155 spatial structure, 178, 294 spear points, 11 spear throwers, 12 Special Weapons Observation Reconnaissance Detection System. See SWORD specialization, 3, 261, 362 specific intelligence, 302 Sperry Elmer Ambrose Sperry, 125 Elmer Sperry Jr., 126 Lawrence Sperry, 126, 189 Sperry Gyroscope Company, 125 Sphinx, 28 spice, 35, 65 Spice Islands, 41 spiegeleisen, 96 spindle, 70 spinning, 66, 69, 70, 71, 72, 86, 87, 111, 125 spinning frame, 70, 71 spinning jenny, 70, 72 spinning mule, 72 spinning wheel, 69, 70, 72 Spirit, 157 spiritual awakening, 235 spiritual need, 23 spiritual needs, 211, 228, 234, 235, 237, 240, 260, 262 spiritual pursuits, 245 spirituality, 399 sports, 176, 299, 450 Sprague Frank J. Sprague, 95 Sprague Electric Railway & Motor Company, 95 Spring-Autumn period, 36, 397 square-pallet chain pump, 37

Economics and Politics in the Robotic Age: The Future of Human Society Squash, 17 SRC, 142, 462 SRI, 154, 462 SRMS, 155, 462 SSEM, 133, 462 St Rollox, 81 St. Benedict, 434, 457 stagecoach, 85 stagflation, 108, 383, 384 Stahlradwagen, 103 stamp seals, 20 standardization, 77, 127, 139, 168, 288, 309, 310, 431 Standuhr, 103 Stanford Arm, 154 Stanford Artificial Intelligence Lab, 154, 461 Stanford Cart, 155 Stanford Heuristic Programming Project, 143 Stanford Research Institute, 154, 462 Starfish, 158 Starley John Kemp Starley, 100 state-owned enterprise. See SOE StatsMonkey, 169 statuettes, 18, 20 Statute of Monopolies, 68 STCR, 288 steam engine, 73, 74, 75, 76, 77, 87, 90, 92, 96, 104, 108, 125, 200, 242, 258, 264 high-pressure steam engine, 76 multiple expansion steam engine, 104 Newcomen steam engine, 74 piston steam engine, 74 stationary steam engine, 76 Watt steam engine, 75 steam jacket, 75 steam steering engine, 125 steam turbine, 94, 105, 120 steamboat, 19, 76, 84 steamship, 84

501

steel, 4, 32, 33, 34, 36, 37, 40, 80, 96, 97, 103, 104, 113, 126, 127, 270, 353 blister steel, 80 crucible steel technique, 80 crucible technique, 33 steel rails, 103 steelmaking, 34, 36, 90, 96, 97, 114 Stele of the Vultures, 28 Step Pyramid at Saqqara, 28 Stephenson George Stephenson, 85 stepping switch, 102 stethoscope, 299 Stevens John Stevens, 84 stirrup, 39 stock, 43, 253, 272, 358 stock exchange, 65 Stockton and Darlington Railway, 85 Stone Age, 4, 13, 32, 46, 59 stone carvings, 29, 35 stone tools, 4, 6, 8, 9, 10, 11, 12, 16, 24, 52, 59 stone-grinding slabs, 19 stone-tipped spears, 9 storage space, 227, 230, 236, 372 stored-program computer, 133, 134, 196 Strategic Computing Initiative. See SCI strength of character, 308, 311, 312 stringed instrument, 29 Strowger Almon Brown Strowger, 102, 126 STS-2 mission, 155 Stuart Herbert Akroyd Stuart, 92 stucco, 82 student evaluations of teaching, 440 Sturgeon William Sturgeon, 95, 113 Su Song, 152, 201 subjective theory of value, 255

502 sub-prime mortgage, 375 subscriber identity module. See SIM subsistence living, 204 sub-symbolic, 146 suction gripper, 173 Sulawesi, 13, 46 sulfuric acid, 80, 81, 99 Sumer, 20, 26, 29, 32, 49 Sumerian, 23, 25, 27, 29 Sumerian King List, 23 Sumerian mythology, 23 Sun Microsystems, 143 Sunshine Carl Sunshine, 138 supercomputer, 135, 136, 137, 278 superintelligence, 408, 421, 422 superphosphate, 99 superstar, 176, 300, 376, 377 supervised learning, 147 supervision robot, 271, 272 support ratio, 378 surface mount technology. See SMT surface-mount device. See SMD sustainable development, 286, 343 Sutcliffe Richard Sutcliffe, 127 Swan Joseph Swan, 94, 111 Swarm intelligence, 150 Swiss referendum, 382 switchboard, 102, 126 SWORD, 179, 462 sycophant, 405, 406, 411 Symington William Symington, 83 synthetic dye, 97, 98 Syracuse University, 177 system of morality, 403 tablet, 137, 140, 164, 172, 198 tactile, 208, 230, 231 tailor shop, 282 tailor-made, 282, 286, 446 tallow, 91 tallow candle, 82 TALON, 179 Talos, 152

Index Tamil Nadu, 33 Tang Dynasty, 39 tanning, 20 Tanum Earth Station, 139 Taoist, 36 taotie, 21 tax rebates, 368 taxation, 28, 30, 334, 382, 387, 388, 453 Taylor Frederick Winslow Taylor, 106, 110 Taylorism, 106 TCP/IP, 138, 139 TDC, 132, 462 TDC Mark 3, 132 teacher robot, 297 technical know-how, 69, 366 technological difference, 22 technological progress, 4, 16, 44, 69, 108, 109, 220, 246, 256, 258, 327, 351, 352, 353, 366, 370, 381, 382, 390, 395, 431, 445, 452 technological unemployment, 256, 352, 357, 390, 431, 432, 452 telecommunication, 101, 107, 116, 137, 138, 162 Teledyne Ryan, 179 telegraphy, 101 telephone, 101, 102, 114, 118, 126, 136, 192, 215, 355 Televox, 153 Tell Abu Hureyra, 14 Tell Aswad, 16, 22 Tell el-Hammeh, 32 Teller Edward Teller, 135 temple of Sin, 27 temporal inefficiency, 410 Tencent, 165, 288 Tennant Charles Tennant, 81 Tentacle Arm, 154 tentering gear, 125 teosinte, 17

Economics and Politics in the Robotic Age: The Future of Human Society terylene, 101 Tesla, 158, 167, 168, 199 Nikola Tesla, 96, 116, 153, 179 Tesla Autopilot, 168 Texas Instruments, 135, 195 text messaging, 137, 138 textile industry, 67, 76, 354 textiles, 12, 18, 40, 44, 47, 54, 66, 67, 69, 81, 353 Thames and Severn Canal, 83 The Records of the Three Kingdoms, 39 The Sanitary Condition of the Labouring Population, 88 thermal cracking, 92 thermal pollution, 331 thermal power station, 329 thermometer, 126 thermophilic species, 331 thermoreceptor, 229 thermoset, 100 thinking patterns, 376 Third Council of the Lateran, 437, 438 Third Dynasty, 28 Third Industrial Revolution, 107, 286 Thomas Sidney Gilchrist Thomas, 97 Thomson James Thomson, 131 Robert William Thomson, 100 William Thomson, 131, 200 thorium, 325 Three Kingdoms period, 37 threshing machine, 44, 89 throne room, 30 throttle valve, 125 thyratron, 132 tidal energy, 326 tide-predicting machine, 131 Tilden Mark Tilden, 154 time budget line, 228 time constraint, 358 time costs, 220

503

time endowment, 220, 221, 223, 225, 231, 238, 239, 246, 264, 389 tin, 24, 25, 26, 27, 31, 56, 69, 325 tinnitus, 332 Tinsley Marion Tinsley, 144 Tiryns, 26 Tito Bustillo, 13 toluene, 329 tomato, 17, 42 tombstone, 269 TOMY, 158 tool steel, 97 Tootill Geoff Tootill, 133 torbanite, 91 Torpedo Data Computer. See TDC tortoise robot, 154 total energy consumption, 324 total fertility rate, 321 total rationality, 222 touchscreen, 137, 165 tower of Jericho, 22 Toyota, 167 tractor, 106, 173, 278 tradable share, 43 trade union, 89, 365, 368 traditional media, 169, 243, 293, 335, 374 transatlantic cables, 102 transformer, 94 transhumanism, 421 Transistorized Experimental computer zero. See TX-0 transitivity, 221 Transmission Control Protocol/Internet Protocol. See TCP/IP transplanter, 106 transport, 3, 103, 108, 109, 165, 282, 283, 288, 360, 413, 431, 447, 461 long-distance transport, 92, 105 marine transport, 105 public transport, 235, 261

504 transport costs, 83, 363, 366, 370, 371, 373 transportation, 4, 65, 68, 82, 83, 101, 261, 283, 352, 359, 373 transsexual, 399 trass, 81 Tree traversal, 144 Trevithick Richard Trevithick, 76, 112 trial-and-error problem solver, 150 trigonometry, 132 triode, 102, 126 trip hammer, 37, 79 Trois Freres, 13 Trump Donald Trump, 214, 364, 377, 378, 385, 386, 391, 414, 430, 431 truth value, 130 Turbinia, 105 turbo generator, 95 Turing Alan Turing, 132 Turing complete, 132, 133 Turing machine, 132 turnip, 42, 43 turnpike, 85 Turnpike trust, 85, 110 turquoise, 20, 22 Tutankhamun, 33, 49 tutorial service, 298 TV shopping channel, 337 Twitter, 232, 335, 391 TX-0, 135, 193, 462 TX-1, 135 TX-2, 135 tyranny of the majority, 412 Tyrian purple, 98 UAS, 179, 462 UAV, 179, 180, 201, 462 Ubaid period, 19 Ubaid phase I, 17, 23 Ubaid phase II, 17 UCAV, 180 UCLA, 139 Udemy, 175

Index Ugarit, 33 Ulaszowice, 91 ultrasound, 299 unconditional basic income, 382, 383, 387 underdeveloped countries, 379, 381, 431 Underground track transport, 166 undersea cable, 102 Uniform Resource Locator. See URL Unimate, 154, 160 Unimation, 154, 155 United East Indian Company, 43 United Nations, 322 United Nations Trust Territories, 322 UNIVAC, 135, 462 Universal Automatic Computer. See UNIVAC universal male suffrage, 89 University of Alberta, 144 University of Bologna, 436, 438, 457 University of California, Los Angeles. See UCLA University of Cambridge, 134, 197 University of Malaga, 176, 300 University of Manchester, 134, 136 University of Nebraska, 178 University of Paris, 436, 438, 458 University of Prince Edward Island, 175 University of Toulouse, 437 Unlawful Oaths Act, 89 Unmanned Aerial Vehicle, 157, 183 unmanned aircraft system, 179 unmanned combat air vehicle. See UCAV unmanned surface vehicle. See USV unmanned underwater vehicle. See UUV unsupervised learning, 147 Upavon Central Flying School, 179 upbringing, 302, 402 Upper Cave, 12

Economics and Politics in the Robotic Age: The Future of Human Society upper-middle-class, 220, 240 Ur, 25, 27 uranium, 10, 325, 344 uranium waste, 326 urbanization, 87, 435 URL, 139 Ur-Nanshe, 27 Uruk, 25, 27, 29 use of fire, 9, 10, 53 USV, 180, 181 uterus, 400, 401 utility, 3, 151, 204, 206, 214, 215, 221, 222, 223, 225, 228, 230, 235, 236, 238, 239, 341, 388, 428, 430, 435, 452 social utility, 415 utility function, 207, 228, 238, 239 utilization rate of production capacity, 356 UUV, 180, 198, 462 V-1, 179 vacuum and pressure water pump, 74 vacuum cleaner, 172, 262, 365 vacuum tube, 102, 126, 133, 134 Vail Alfred Vail, 101 value system, 213, 389 van Gogh Vincent van Gogh, 316 vanity, 209, 223, 224, 226, 232, 233, 240, 265, 388, 416, 429 Varahamihira, 37 Vatsyayana, 37 vegetable oil, 91 vegetation, 34, 330, 331 vending machine, 151, 164, 165 Venetian Patent Statute, 68 Verbiest Ferdinand Verbiest, 434 Verbruggen Jan Verbruggen, 77 very large-scale ICs. See VLSI vested interests, 313 vetch, 39 veterinarian, 280

505

veterinary, 280 Vicarious, 420 Vicarm Inc, 155 Vinþa, 17, 20, 24 violent revolution, 350, 390 Virgil, 434 virtual AI teacher, 297 virtual conference, 405 visceral sense, 229 viscose, 100 viscose rayon, 100 Vishnu Sharma, 37 visual sense, 208 viticulture, 25 Vitruvius, 36 VLSI, 135 VOC, 329, 331 volatile organic compound, 329 Volk's electric railway, 95 Volkswagen, 167 Volta Alessandro Volta, 93 Volta’s battery, 93 volumetric receptor, 229 Volvo, 167 Volvo Proving Ground, 167 von Guericke Otto von Guericke, 93, 112 von Neumann John von Neumann, 132, 133 voyages of discovery, 65 vulcanization, 99 Wabot-2, 155 wage rate, 220, 225, 238, 340, 359, 389 wagon, 17, 83, 85 Wallach Wendell Wallach, 407, 422, 423 Walter William Grey Walter, 153 Waltham, 73, 111 Waltham Watch Company, 78 Waltham-Lowell system, 73 wanax, 30 WAP, 140, 462 warp, 70, 71

506 Warring States period, 36 warriors, 29 Waseda University, 156 washing machine, 355 watchdog, 411 water frame, 71, 72 water pollution, 331, 342 water power, 69, 71, 72, 73, 254, 287, 325 water supply, 27, 180, 287, 323 water tables, 17 watercraft, 27 waterfowl, 14 watermill, 36 waterwheel, 37, 72 Watson, 144, 145, 148, 169, 170, 175, 183, 263, 293, 376, 390, 418, 428, 440, 441, 444, 449, 456 Watt James Watt, 75, 76, 125 Watt engine, 83 wattle-and-daub, 22 Wave Glider, 180 Waymo, 158, 168 Waymo One, 158 wealth concentration, 348, 373, 374, 382 Wealth of Nations, 254, 299, 306, 438, 439, 459 weapon, 9, 12, 19, 33, 34, 47, 54, 179, 311, 408 weaver, 69, 72, 266 weaving, 19, 66, 72, 87 WeChat, 165, 232, 288, 337, 371 weft, 70 weights, 27, 385 we-media, 337 Wensley Roy James Wensley, 153 Weskott Johann Friedrich Weskott, 98 West Java, 43 West Turkana, 8, 52 Westinghouse Electric Corporation, 153

Index whale oil, 91 WhatsApp, 232 wheat, 17, 190 Wheatstone Charles Wheatstone, 101 wheelbarrow, 37 wheeled vehicles, 25 Whinfield John Rex Whinfield, 101 whippletree, 35 Whirlwind I,, 134 whistles, 20 white ware vessels, 20 white-collar workers, 352, 386 Whitney Eli Whitney, 73 Whitworth Joseph Whitworth, 77 WHO, 332 wholesaler, 337 Wi-Fi, 140, 172 wild boars, 14 Wilkes John Wilkes, 89 Maurice Wilkes, 134 Wilkinson John Wilkinson, 77 Williams Frederic C. Williams, 133 Wilson Ernest Wilson, 153 wind power, 19, 254 windmill, 36, 69, 124 window shopping, 372 Windows, 371 wine-making, 25 winner takes all, 243, 359, 372 wireless application protocol. See WAP wireless communications, 101, 197 Wireless Fidelity. See Wi-Fi wireless remote control, 153 Wireless Telegraph & Signal Company, 102 Wirksworth, 72

Economics and Politics in the Robotic Age: The Future of Human Society Wood Charles Wood, 79 woodblock printing, 40 wooden carving, 20 wool, 17, 44, 66, 67, 70, 73, 86, 87, 98, 117 Woolf Arthur Woolf, 76, 116 Woolrich John Stephen Woolrich, 93 Woolrich Electrical Generator, 93 Worcester, 73 worker-manager relation, 447 working class, 366, 375, 381, 385 working hours, 88, 259, 339, 360 working-age household, 123, 245, 397 working-age population, 397 World Health Organization. See WHO World Wide Web. See WWW Wright brothers, 105 writing, 21, 27, 28, 29, 35, 39, 58, 170, 174, 178, 235, 236, 294, 376, 434 syllabic writing, 27 Wucheng, 28 WWW, 139, 169, 462 Wyatt John Wyatt, 70 Wynn-Williams Charles Eryl Wynn-Williams, 132 X10, 172 xanthate, 100 Xbox 360, 145 Xbox One, 145 xenobiotic, 331 Xenophanes, 35 Xia Dynasty, 26

507

Xiang River, 36 Xianren Cave, 15 Xihoudu, 9 Xijiang River, 36 Xinle, 19, 20 X-rays, 299 xylene, 329 Yamanashi University, 155 Yan Shi, 152 Yangshao, 17, 19, 20 Yangtze, 17, 27, 36, 62 Yanshi, 30 Yarmukian, 20, 21, 22, 51 yarn, 70, 71, 72 Yellow River, 26, 48 Yinxu, 28, 31, 62 yokes, 25 Young James Young, 91 YouTube, 232, 337, 358, 370 Yuan Dynasty, 40 Yuanmou Man, 9 Z1, 132 Z2, 132 Z3, 132 Z4, 132 Zarzian, 14 zero-manufacturing worker point, 260 zero-service worker point, 260 zero-worker point, 260 Zheng He, 40, 54 Zhengguo Canal, 36 Zhou Dynasty, 36 Zhoukoudian, 12 ZigBee, 172 Zuse Konrad Zuse, 132, 200 Z-Wave, 172