(PDF) The Automation of Society is Next: How to Survive the Digital Revolution

The Automation of

Society is Next

How to Survive the Digital Revolution

Version 1.0

DIRK HELBING

i

DEDICATION

I would like to dedicate this book to Dietmar Huber

for the incredible support he has given to me over so many years.

ii

iii

CONTENTS

1

THE DIGITAL SOCIETY

A better future or worse?

1

2

COMPLEXITY TIME BOMB

When systems get out of control

17

3

SOCIAL FORCES

Revealing the causes of success and disaster

35

4

GOOGLE AS GOD?

The dangerous promise of Big Data

61

5

GENIE OUT OF THE BOTTLE

Major socio-economic shifts ahead

83

6

DIGITALLY ASSISTED SELF-ORGANIZATION

Making the invisible hand work

103

7

HOW SOCIETY WORKS

Social order by self-organization

127

8

NETWORKED MINDS

Where human evolution is heading

147

9

ECONOMY 4.0

A participatory market society is born

171

10

THE SELF-ORGANIZING SOCIETY

Taking the future in our hands

199

iv

v

ACKNOWLEDGMENTS

I would like to thank the FuturICT community for the many inspiring

discussions and everyone, who was patient with me, including

my parents and whoever might have reasons for complaints.

I am also very grateful to Philip Ball, Stefano Bennati, Anna Carbone,

Andreas Diekmann, Dietmar Huber, Eoin Jones, Caleb Koch, Richard

Mann, Heinrich Nax, Paul Ormerod, Evangelos Pournaras, Kay-Ti Tan,

and others for their valuable feedback on the manuscript

and the many improvements (but don't hold them responsible

for any contents of this book). Petr Neugebauer and Petra Parikova

have been a great help with figures and formatting.

vi

ABOUT THE AUTHOR

vii

ABOUT THE AUTHOR

Dirk Helbing is one of the most imaginative experts in the world when it

comes to envisioning the opportunities and risks of the digital revolution.

He is an advocate of responsible innovation and strongly contributed to the

public debate around Big Data. He also coordinates the FuturICT initiative,

which built a global interdisciplinary community of experts at the interface

of complexity, computer and data science. These activities, aimed at

confronting global problems and crises were featured by Scientific

American as the number one world-changing idea and earned him an

honorary doctorate from TU Delft, where he is now an affiliate professor.

Dirk Helbing is Professor of Computational Social Science at the

Department of Humanities, Social and Political Sciences and also affiliated

to the Computer Science Department at ETH Zurich. He has a PhD in

physics, was Managing Director of the Institute of Transport & Economics

at Dresden University of Technology in Germany, and Professor of

Sociology at ETH Zurich.

Helbing is an elected member of the German Academy of Sciences

"Leopoldina" and worked for the World Economic Forum's Global

Agenda Council on Complex Systems. He is also co-founder of the Physics

of Socio-Economic Systems Division of the German Physical Society and

ETH Zurich's Risk Center. Furthermore, he is a board member of the

Global Brain Institute in Brussels and the International Centre for Earth

Simulation in Geneva. In addition, he is a member of various high-level

committees to assess the implications of the digital revolution.

The motivation for his research may be summarized by "What can

complexity science and information systems contribute to saving human

lives?" This ranges from avoiding crowd disasters over reducing crime and

conflict to the reduction of epidemic spreading. His work brings theoretical

studies, data analytics, and lab experiments together with agent-based

computer models, where agents might have cognitive features. His recent

publication on globally networked risks establishes the framework of a

Global Systems Science.

Using the emergent "Internet of Things", his team is now engaged in

establishing the core of a decentralized Planetary Nervous System as a

Citizen Web (see nervousnet.info). This will be an open, transparent and

participatory information platform to support real-time measurements of

our world, situational awareness, successful decision-making, and self-

organization. The goal of this system is to open up the new opportunities of

the digital age for everyone.

D. HELBING: THE AUTOMATION OF SOCIETY

viii

Photo: Sabina Bobst

PREFACE

ix

"We hold these truths to be self-evident, that all men are created equal, that they are

endowed by their Creator with certain unalienable Rights, that among these are Life,

Liberty and the pursuit of Happiness."

The United States Declaration of Independence

after Thomas Jefferson

1

"Those who surrender freedom for security will not have, nor do they deserve, either one."

Benjamin Franklin

2

February 23, 2012

"Americans have always cherished our privacy. From the birth of our republic, we

assured ourselves protection against unlawful intrusion into our homes and our personal

papers. At the same time, we set up a postal system to enable citizens all over the new

nation to engage in commerce and political discourse. Soon after, Congress made it a crime

to invade the privacy of the mails. And later we extended privacy protections to new

modes of communications such as the telephone, the computer, and eventually email.

Justice Brandeis taught us that privacy is the "right to be let alone," but we also

know that privacy is about much more than just solitude or secrecy. Citizens who feel

protected from misuse of their personal information feel free to engage in commerce, to

participate in the political process, or to seek needed health care. This is why we have laws

that protect financial privacy and health privacy, and that protect consumers against

unfair and deceptive uses of their information. This is why the Supreme Court has

protected anonymous political speech, the same right exercised by the pamphleteers of the

early Republic and today's bloggers.

Never has privacy been more important than today, in the age of the Internet, the

World Wide Web and smart phones. In just the last decade, the Internet has enabled a

renewal of direct political engagement by citizens around the globe and an explosion of

commerce and innovation creating jobs of the future. Much of this innovation is enabled

by novel uses of personal information. So, it is incumbent on us to do what we have done

throughout history: apply our timeless privacy values to the new technologies and

circumstances of our times.

I am pleased to present this new Consumer Privacy Bill of Rights as a blueprint for

privacy in the information age. These rights give consumers clear guidance on what they

should expect from those who handle their personal information, and set expectations for

companies that use personal data. I call on these companies to begin immediately working

with privacy advocates, consumer protection enforcement agencies, and others to implement

these principles in enforceable codes of conduct. My Administration will work to advance

these principles and work with Congress to put them into law. With this Consumer

D. HELBING: THE AUTOMATION OF SOCIETY

x

Privacy Bill of Rights, we offer to the world a dynamic model of how to offer strong

privacy protection and enable ongoing innovation in new information technologies.

One thing should be clear, even though we live in a world in which we share personal

information more freely than in the past, we must reject the conclusion that privacy is an

outmoded value. It has been at the heart of our democracy from its inception, and we need

it now more than ever..."

Barack Obama, US Consumer Privacy Bill

of Rights

3

1

1: INTRODUCTION: THE DIGITAL SOCIETY

A better future or worse?

Smartphones, tablets and app stores with almost unlimited possibilities have become

symbols of the digital revolution. However, while these innovations make our lives more

comfortable and interesting, they herald a much more fundamental transformation.

Advances in digital technology now affect the way we learn, decide, and interact. By

harnessing "Big Data", the "Internet of Things", and Artificial Intelligence (AI), we

can create smart homes and smart cities. But this is only the tip of the iceberg – our entire

economy and society will also dramatically change. What are the opportunities and risks

related to this? Are we heading towards digital slavery or freedom? What forces are at

work and how can we use them to create a smarter society? This book offers a guided tour

through the new, digital age ahead.

After the automation of factories and the creation of self-driving cars,

the automation of society is next. While we were busy with our

smartphones, the world has secretly changed behind our backs. In fact, our

world is changing with increasing speed, and much of that change is being

driven by developments in Information and Communications Technology

(ICT). These technologies, such as laptop computers, mobile phones,

tablets and smart watches, seemed to be about convenience. They came

along and enabled us to calculate, communicate and archive with greater

speed and efficiency than ever before. However, there was very little

recognition that, one day, they would not only facilitate our cultural

discourse and institutions, but also reshape our entire world. Large-scale

mass surveillance, the global spread of Uber taxis and the BitCoin crypto-

currency are just a few of the irritating symptoms of the digital era to come.

Living in the age of "Big Data"

Suddenly, there is also a great hype about "Big Data". No wonder Dan

Ariely compared the frenzy about Big Data with teenage sex: "everyone talks

D. HELBING: THE AUTOMATION OF SOCIETY

2

about it, nobody really knows how to do it, everyone thinks everyone else is doing it, so

everyone claims they are doing it..."

But some are actually doing it. In fact, "Big Data" has already given rise

to many interesting applications, such as real-time language translation. So,

what is "Big Data"? The term refers to massive amounts of data, which

have been collected about technological, social, economic and

environmental systems and activities. To get an idea of "Big Data," imagine

the digital traces that almost all our activities leave, including the data

created by our consumption and movement patterns. Every single minute,

we produce about 700,000 Google queries and 500,000 Facebook comments.

If you add all of the location data of people using smartphones, the

purchasing data of people who buy things, and cookies which track every

click and tap of our online activity, you will begin to comprehend the

enormity of "Big Data".

All the contents collected in the history of humankind until the year

2003 are estimated to amount to five billion gigabytes – the data volume

that is now produced approximately every day. While we have been

speaking of an "information age" since the middle of last century, the digital

era started only in 2002. Since then, the digital storage capacity has

exceeded the analog one. Today, more than 95 percent of all data are

available in digital form. Even by avoiding credit card transactions, social

media and digital technologies, it is no longer possible to completely avoid

digital footprints on the Internet.

Data sets bigger than the largest library

The availability of Big Data about almost every aspect of our lives,

institutions and cultures has fueled the hope that we could now solve the

world's problems. Every Internet purchase we make generates data about

our preferences, finances and location that will be stored on a server

somewhere and used for various purposes, possibly without our consent.

Cell phones disclose where we are, and private messages and conversations

are being analyzed. It will probably not be long before every newborn baby

is genome-sequenced at birth. Books are being digitized and collated in

immense, searchable databases of words that are being data-mined to

enable "culturomics", a field which puts history, society, art and cultural

trends under the lens. Aggregated data can be used to reveal unexpected

facts in a way that would never have been possible before the digital age.

For example, an analysis of Google searches can reveal an impending flu

epidemic.

This avalanche of data continues to grow. The introduction of

technologies such as Google Glass encourages people to document and

archive almost every aspect of their lives. Further data sets include credit-

card transactions, communication data, Google Earth imagery, public news,

INTRODUCTION: THE DIGITAL SOCIETY

3

comments and blogs. These data sources have been termed "Big Data" and

are creating an increasingly accurate digital picture of our physical and social

world, as well as the global economy.

"Big Data" will certainly change our world. The term was coined more

than 15 years ago to describe data sets so big that they can no longer be

analyzed using standard computational methods. If we are to benefit from

Big Data, we must learn to "drill" and "refine" it into useful information

and knowledge. This is a significant challenge.

The tremendous increase in the volume of data is attributable to four

important technological innovations. First, the Internet enables global

communication between electronic devices. Second, the World Wide Web

(WWW) has created a network of globally accessible websites, which

emerged as a result of the invention of the Hypertext Transfer Protocol

(HTTP). Third, the emergence of social media platforms such as Facebook,

Google+, WhatsApp and Twitter has created social communication networks.

Finally, a wide range of previously offline devices such as TV sets, fridges,

coffee machines, cameras as well as sensors, smart wearable devices (such

as activity trackers) and machines are now connected to the Internet,

creating the "Internet of Things" (IoT) or "Internet of Everything" (IoE).

Meanwhile, the data sets collected by companies such as eBay, Walmart or

Facebook, must be measured in petabytes – 1 million billion bytes. This

amounts to more than 100 times the information stored in the US Library

of Congress, which is the largest physical library in the world.

Mining Big Data offers the potential to create new ways to optimize

processes, identify interdependencies and make informed decisions.

However, Big Data also produces at least four major new challenges (the

"four V's"). First, the unprecedented volume of data means that we need

immense processing power and storage capacity to deal with the huge

amounts of data. Second, the velocity at which data must be processed has

increased: now, continuous data streams must often be analyzed in real-

time. Third, Big Data is mostly unstructured, and the resulting variety of data

is difficult to organize and analyze. Finally, the veracity of the data may be

difficult to handle because Big Data tends to contain errors and is usually

neither representative nor complete.

Will a digital revolution solve our problems?

Let us see what an evidence-based approach building on the wealth of

today's data can do for us. In the past, whenever a problem had to be

solved, the best course of action was to "ask the experts". These experts

would go to the library, collect up-to-date knowledge, and supervise PhD

students who would help to fill gaps in existing knowledge. But this was a

slow process. Nowadays, whenever people have a question, they ask Google

or consult Wikipedia, for example. This might not always give the definitive

D. HELBING: THE AUTOMATION OF SOCIETY

4

or best answer, but it delivers quick answers. On average, decisions taken in

this way may even be better than many decisions made in the past. It is no

wonder, therefore, that policymakers love the Big Data approach, which

seems to provide immediate answers. Business people sensing the immense

commercial opportunities are getting excited too.

Big Data Gold rush for the 21st century's oil

The fact that we have much more information about our world than

ever before is both a blessing and a curse. The accumulation of socio-

economic data often implies a long-term intrusion into personal privacy and

raises a number of important issues. It cannot be denied that Big Data is a

powerful resource that supports evidence-based decision-making and that it

holds unprecedented potential for business, politics, science and citizens.

Recently, the social media portal WhatsApp was sold to Facebook for $19

billion, when it had 450 million users. This sale price implies that each

employee generated almost half a billion dollars in share value.

There is no doubt that Big Data creates tremendous opportunities, not

just because of its application in process optimization and marketing, but

also because the information itself is becoming monetized. As

demonstrated by the virtual currency BitCoin, it is now even possible to turn

bits into monetary value. It can be literally said that data can be mined into

money in a way that would previously have been considered a fairy tale. For

a time, BitCoins were even more valuable than gold.

Therefore, it is no surprise that many experts and technology gurus

claim that Big Data is the "oil of the 21st century", a new way of making

money – big money. Although many Big Data sets are proprietary, the

consultancy company McKinsey recently estimated that the potential value of

Open Data alone is $3-5 trillions per year.

4

If the worth of this publicly

available information were to be evenly distributed among the world’s

population, every person on Earth would receive an additional $700 per

year. Therefore, the potential of Open Data significantly exceeds the value

of the international free trade and service agreements that are currently

under secret negotiation.

i

Given these numbers, are we currently setting the

right political and economic priorities? This is a question we must pay

attention to, because it will determine our future.

The potential of Big Data spans every area of social activity, from

processing human language and managing financial assets, to empowering

i

I recommend the readers to look up Wikipedia to inform themselves about the impending

international agreements coming under the abbreviations TTIP, CETA, TPP, and TISA. It

seems that these would dramatically increase the power of multi-national corporations.

Would this be good or bad?

INTRODUCTION: THE DIGITAL SOCIETY

5

cities to balance energy consumption and production. Big Data also holds

the promise of enabling us to better protect the environment, to detect and

reduce risks, and to discover opportunities that would otherwise have been

missed. In the area of personalized medicine, Big Data will probably make it

possible to tailor medications to patients in order to increase their

effectiveness and reduce their side effects. Big Data will also accelerate the

research and development of new drugs and focus resources on the areas of

greatest need.

It is clear, therefore, that the potential applications of Big Data are

various and rapidly spreading. While it will enable personalized services and

products, optimized production and distribution processes, as well as

"smart cities", it will reveal also unexpected links between our activities. But

beyond this, where are we heading?

Will Artificial Intelligence overtake us?

Today, an average mobile phone is more powerful than the computers

used to send the Apollo rocket to the moon and even the Cray-2

supercomputer thirty years ago, which weighted several tons and had the

size of a building. This amazing progress is a result of "Moore's law", which

posits that computer processing power increases exponentially. But thanks

to powerful "machine learning" methods, information systems are

becoming more intelligent, too. They do calculations faster than us, they

play chess better than us, they remember information longer than us, and

they perform ever more tasks that only humans could do in the past. Will

they soon be smarter than us? Are the days counted when humans were the

"crown of creation"? The famous futurist Ray Kurzweil (*1948), now a

director of engineering at Google, was the first to claim that this moment

(the so-called "singularity") is near.

5

A few years ago, when I read that Artificial Intelligence (AI) might pose

a serious threat to humanity, I found this hard to imagine. However,

experts now predict that computers will be able to perform most tasks

better than humans in 5 to 10 years, and reach brain-like functionality

within 10 to 25 years. The AI systems of today are no longer expert systems

programmed by computer scientists – they are learning and evolving. To

understand the implications, I recommend you to watch some eye-opening

videos on deep learning and artificial intelligence.

6

These videos

demonstrate that most of the activities we earn our money with today (such

as reading and listening to language, distinguishing different patterns, and

performing routines) can now be done by computers almost as well as

humans, if not better. Jim Spohrer's perspective on IBM's cognitive

computing products is as follows:

7

The first Artificial Intelligence

applications will be our tools. As they get smarter, they will become our

"partners", and when they overtake us, they will be our "coaches".

D. HELBING: THE AUTOMATION OF SOCIETY

6

Will algorithms, computers, or robots be our bosses in a few decades

from now? The Massachusetts Institute of Technology has started to study

such scenarios.

8

It is extremely important therefore to realize that the digital

revolution is not just about more powerful computers, better smartphones

or fancier gadgets. The digital revolution will change all our personal lives,

and it will transform entire economies and societies. In fact, in the coming

two or three decades we will see some dramatic changes. A lot of

production and services will become automated, and this will fundamentally

change the way we work in future.

Quite soon, within the next two decades or so, less than 50 percent of

people will have jobs for which they have been trained (i.e. agriculture,

industry or services).

9

Even highly skilled jobs will be at risk. How will the

masses of personal data collected about each of us then be used?

When Big Data starts to steer our lives

It may sound far-fetched at first, but we must ask this question: "Will we

be remotely controlled by personalized information, or is this happening

already?" It is clear that Google and Facebook know very well what we are

interested in when they place individually tailored ads that often match our

interests and tastes. The smart app Google Now is an example of a smart app

that tells you what to do, if you have signed up for it. For instance, if there

is a traffic jam on the way to your next appointment, Google Now may

suggest you to leave 15 minutes earlier in order to be on time. Similarly,

Amazon suggests what we might want to buy, and Trip Advisor suggests what

destinations to visit and what hotels to book. Twitter tells us what others

think – and what we should perhaps think, too. Facebook suggests whom to

be friend with. Apps like OkCupid even suggest whom we might date.

While all these services might certainly be helpful we might ask: what

will be the consequences? Will we end up living in a digital "golden cage" –

a "filter bubble" as Eli Pariser calls it?

10

Will we just execute what our smart

devices tell us to do? Modern learning software already corrects us when we

make mistakes. Smart wristbands tell us how many more steps we should

make in a day. Eye trackers can discover if we are tired or stressed, and

computers can predict when our performance will decrease. In other words,

we are increasingly patronized in our decision making by computer

programs. Will we soon be incompetent to live on our own? And, are we

sliding into a "nanny state", where we don't have a say? Has our decision-

making, has democracy been "hacked"?

Why should we care? Isn't it just great that computers do calculations

for us more quickly than we can do them ourselves? Isn't it fantastic that

our smartphones help us manage our agendas, and that Google Maps tells us

the way to go? Why not ask Apple's Siri to recommend us a restaurant? I

certainly don't object to any of these functions, but it is important to

INTRODUCTION: THE DIGITAL SOCIETY

7

recognize that this is just the beginning of what is to come. Little by little,

our role as self-determined decision-makers is being eroded. The next

logical step will be the automation of society. How might this look like?

The cybernetic society

This question brings us to an old concept that goes back to Norbert

Wiener (1894-1964)

11

who was known as the father of control theory

("cybernetics"), Wiener imagined that our society could be controlled like a

huge clockwork, where every company's and every individual's activity

would be coordinated by a giant plan of how to run a society in an optimal

way.

Many decades ago, Russia and other communist countries ran command

economies. However, they failed to be competitive, while the capitalist

approach based on free entrepreneurship thrived. At that time information

systems were much more limited in power and scope than today. This has

changed. Now there is a third approach besides communism and capitalism:

socio- economic systems that are managed in a data-driven way. In the early

70ies Chile was the first country to attempt a "cybernetic society".

12

It

established a control center, which collected the latest production data of

major companies every week. This was a truly revolutionary approach, but

despite its obvious advantages, the government was unable to stay in power,

and Salvador Allende (1908-1973), the president of the country, had a tragic

end.

ii

Nevertheless, the dream of a cybernetic society has not ceased to

exist.

Today, both Singapore and China are trying to plan social and economic

activities in a top-down way using lots of data, and they enjoy larger growth

rates than Western democracies. Therefore, many economic and political

leaders raise the question: "Is democracy outdated?" Should we run our

societies in a cybernetic way according to a grand plan? Will Big Data allow

us to optimize our future?

Wise kings and benevolent dictators, fueled by Big Data

Given all the data one can now accumulate, is it conceivable that

governments or big companies might try to build "God-like", almost

"omniscient" information systems? Could these systems then make

decisions like a "benevolent dictator" or "wise king"? Will they be able to

avoid coordination failures and irrationality? Would it even become possible

to create the best of all worlds by collecting all data globally and building a

digital "crystal ball" to predict the future, as some people have suggested? If

ii

He committed suicide.

D. HELBING: THE AUTOMATION OF SOCIETY

8

this were possible, and given that "knowledge is power", could a sort of

"magic wand" be created by a government or company to ensure that the

benevolent dictator's master plan remains on course?

What would it take to build such powerful tools? It would require

information systems that knew us so well that they could manipulate our

decisions by stimulating us with the right kind of personalized information.

As I will show in this book, such systems are actually on their way, and to

some extent they already exist.

Do we need to sacrifice our personal freedom?

Establishing a cybernetic society has a number of important

implications. For example, we would need a lot of personal data. In order to

be able to control an entire society, it seems important to understand how

we think, what we feel, and what we plan to do. Large amounts of personal

data are essential to allow artificially intelligent machines to learn what

determines our actions and how to influence them. In fact, while mass

surveillance is surprisingly ineffective in fighting terrorism

13

and child

abuse,

iii

it seems to be very useful to establish a cybernetic society.

But as with every technology, there are serious drawbacks. We would

probably lose some of the most important rights and values that have

formed the bedrock of democracies and their judicial systems since the Age

of Enlightenment. Secrecy and privacy would be eroded by information

technologies, and with this, we would lose our security and human values

such as mercy and forgiveness. With the advent of predictive policing and

other proactive enforcement measures, we could see a deviation from the

"presumption of innocence" principle towards the implementation of an

ominous "public interest" policy at the cost of individual rights. Do we

therefore need to worry about the fact that the leading Big Data nation has

more people per thousand inhabitants in prison than any other country,

including Russia and China?

14

iii

The biggest sexual child abuse scandals revealed in the past years have actually not been

discovered by digital surveillance techniques.

INTRODUCTION: THE DIGITAL SOCIETY

9

FIGURE 1.1: Illustration of how citizens could be punished for any minor transgression

of law, even if it is entirely harmless to society. Note that the traffic authority figured out

my foreign address to send me this ticket for going 1km/h too fast with my car, while I

was actually not even the driver (which they didn't check)...

With the help of mass surveillance, it is now possible to punish even the

smallest mistakes that everyone makes in an overregulated society.

iv

The

speeding ticket above for going 1km/h too fast with my car (see Fig. 1.1) is

perhaps a portent of what could become possible on a much larger scale. In

addition to sanctions by public authorities, will insurance companies punish

us in future for eating unhealthy food? Will banks offer us punitive interest

rates on loans simply because we live in the "wrong neighborhood"? Will

we get restricted offers of products or services if we don't fulfill certain

expectations, or will we have to pay higher prices? While this might sound

like a dystopian science fiction fantasy, much of this is already happening.

China is now even planning to rate the behavior of all its citizens on a one-

dimensional scale, including what they do in the Internet.

15

Opinions that

match the thinking of the communist political party will be rewarded. The

resulting score will be used to determine whether or not a person gets a

particular job or loan.

iv

In this connection I recommend to read J. Schmieder (2013) Mit einem Bein im Knast:

Mein Versuch, ein Jahr lang gesetzestreu zu leben (Bertelsmann).

D. HELBING: THE AUTOMATION OF SOCIETY

10

Can such technology- and data-driven approaches turn a country into a

"perfect clockwork"? And given that every country is exposed to global

competition, will it just be a matter of time until democracies adopt similar

approaches? If you think this is far-fetched, it is probably good to recall that

several influential decision-makers have recently praised China and

Singapore as models for the world.

v

Such thinking could soon end freedom

and democracy as we know it.

vi

That is why we must pay attention to this

now. Of course, some people might ask why we shouldn't do it, if this

increases the efficiency of our lives and of our society, why shouldn't we do

it? Doesn't history teach us that society evolves over time? Why should we

worry, if companies and governments take care of us?

The crucial question is whether they are doing a good job in satisfying

our needs and interests. In view of financial and economic crises,

cybercrime, climate change and many other problems, it seems, however,

that governments have great difficulties to fulfill this promise. Similarly, if

we consider Silicon Valley as a business-driven vision of society, it also

seems far-fetched to claim that everyone there is well taken care of.

Who will rule the world?

There is no doubt that Big Data has the potential to be totalitarian. But,

in principle, there is nothing wrong with Big Data, Artificial Intelligence,

and cybernetics. The question is only, how to use it? For example, who will

rule the world in future: will it be big business, government elites, or

Artificial Intelligence? Will citizens and experts be no longer relevant to

decision-making processes? If powerful information systems knew the

world and each of us, would they vote or take decisions for us? Would they

tell us what to do or steer our behavior through personalized information?

Or will we instead live in a free and democratic society, in which everyone

makes decisions in an autonomous but well- coordinated way, empowered

by personal digital assistants?

Two scenarios: coercion or freedom

Will we need to sacrifice our privacy, freedom, dignity, and

informational self- determination for a more efficient governance of our

world? We must think about these possibilities now. Due to the important

societal, economic, legal and ethical implications, we must take some crucial

v

For example, on March 24, 2015, The Economist featured a comment entitled "China

Model. This house believes China offers a better development model than the West".

vi

If you don't like the concepts of freedom and democracy, replace them by the related ones

of innovation and collective intelligence, which are important to find good solutions to

complex problems (see Appendix 4.1).

INTRODUCTION: THE DIGITAL SOCIETY

11

decisions.

16

In the aftermath of September 11, 2001, we have certainly

witnessed increasing attempts to control citizen activities, including a

massive surveillance of our on-line activities. Does the digital revolution

imply that we will lose our human rights? Will we lose our autonomy and

merely obey what powerful information systems tell us to do? Will we end

up with censorship?

The digital revolution implies great opportunities and risks. Digital

technologies enable different ways of running future economies and

societies. If we don't want to lose our jobs, personal freedom, and

democracy, we must carefully consider how to make digital technologies

work for us, rather than against us. There are at least two possibilities by

automate society: we may either run it in a top-down way, trying to

technocratically controlling citizens decisions and actions, using powerful

information technologies, or we may instead support bottom-up self-

organization based on distributed control. The latter would be compatible

with individual freedom, creativity and innovation (see Fig. 1.2 below).

However, the digital revolution enables both. As I will explain further on,

to support largely autonomous decisions and processes and the

coordination of them, we would have to

1. create participatory opportunities,

2. support informational self-control,

3. increase distributed design and control elements,

4. add transparency for the sake of trust,

5. reduce information biases and noise,

6. enable user-controlled information filters,

7. support socio-economic diversity,

8. increase interoperability and innovation,

9. build coordination tools,

10. create digital assistants,

11. support collective ("swarm") intelligence,

12. measure and consider external effects ("externalities"),

13. enable favorable feedback loops,

14. support a fair and multi-dimensional value exchange,

15. increase digital literacy and awareness.

D. HELBING: THE AUTOMATION OF SOCIETY

12

FIGURE 1.2: Schematic illustration of two evolutionary paths, leading to two different

types of a digital society. Path A would undermine individual freedoms, democracy, and

jobs for most of us. Path B corresponds to a society based on self-control and participatory

information systems supporting creative and innovative activities of everyone. Which one

will we choose?

In this book, I am trying to offer concepts and ideas that can contribute

to a smarter and more resilient digital society. Such a framework is needed,

because in many important respects, the world has become quite

unpredictable and unstable. This is partially due to the increasing level of

interdependency of our systems, often driven by advances in Information

and Communication Technology. Therefore, which approach will be

superior in a world characterized by too much data, too much speed, and

too much connectivity? Will it be top-down governance or bottom-up

participation? Or would it be better to combine both approaches? And how

would we do this?

vii

We will see that this is mainly depends on the

complexity of the systems surrounding us.

vii

In his book on The Third Industrial Revolution, Jeremy Rifkin gives interesting answers to

this for the case of smart (energy) grids and some other systems.

INTRODUCTION: THE DIGITAL SOCIETY

13

A better future ahead of us

Despite the many challenges, I am still optimistic about our long-term

future on the whole. We have already managed societal transitions several

times in human history and I am sure we can manage this one too.

In the following chapters I hope to make a contribution to a necessary

public debate, by proposing two main possible societal frameworks for the

coming digital age. One of these is based upon the concept of a "big

government", which takes decisions like a "benevolent dictator" or a "wise

king". This framework would be empowered by huge masses of data in the

manner of a digital "crystal ball". This might be seen as a futuristic version

of Thomas Hobbes (1588-1679) "Leviathan", the powerful state. His belief

was that social order can not exist without a powerful state, otherwise we

would all behave like wild beasts.

viii

The alternative framework for the digital society is based on the concept

of self-organizing systems. This vision relates to the concept of the

"invisible hand", for which Adam Smith (1723-1790) is known. He assumed

that the best societal outcomes are reached through self-organization based

on market forces. However, financial meltdowns and "tragedies of the

commons" such as environmental pollution or harmful climate change

17

suggest that the "invisible hand" cannot be relied on.

But, what if future information and communication technologies would

allow us to reach desirable systemic outcomes through decentralized

decision-making and self-organization? Can distributed control and

coordination mechanisms, empowered by real-time measurements and

feedback, make the "invisible hand" work? The feasibility of this exciting

vision will be explored in the later chapters of this book, in which I will

describe a new paradigm to achieve success and socio-economic order in

the 21st century. Will this lead us into a new era of creativity, participation,

"collective intelligence", and well- being?

In fact, examples from the spheres of traffic management and

production demonstrate that it is possible to manage complex systems from

the bottom-up and to efficiently produce desirable outcomes in this way. In

the following chapters, I will explain the general principle behind such

"magic self-organization" and how it could help us to navigate our way

through a complex future. I will further explain the role of collective

intelligence and how it can help us to cope with the complexity of our

globalized world.

viii

This has been expressed by the Latin phrase "homo hominis lupus", i.e. humans treat

each other like wolves.

D. HELBING: THE AUTOMATION OF SOCIETY

14

Therefore, rather than trying to control or combat the self-organized

dynamics of complex systems such as our economy, financial system, global

trade, and transportation systems, we could learn to harness their

underlying forces to our benefit. Of course, this would involve to locally

adapt the interactions of the components of these systems. But if we could

achieve this, self-organization could be used to produce desirable outcomes,

and this would enable us to create well-ordered, effective, efficient and

resilient systems!

Critics might argue that, just because self-organization has been shown

to work in complex technological systems (such as traffic control or

industrial production lines), this does not necessarily mean that it would

also work for socio-economic systems. After all, the behaviors of people

can be quite surprising. In the light of this counterargument, we will explore

whether and when self- organization can outperform conventional top-

down control in managing complex dynamical systems. It will discuss

institutional settings and interaction rules that will allow self-organizing

systems to be superior. We can now use real-time data to enable adaptive

feedback mechanisms, so that systems behave favorably and in a stable way.

More specifically, I propose that the "Internet of Things", with its vast

underlying networks of sensors, will make socio- economic self-

organization possible in a distributed and bottom-up way. But the crucial

question is how to make a data-oriented approach based on distributed

control work? The solution, as we will see, requires "complexity science.

On the way to a smarter digital society

In the long run, I am confident about our future, mainly because I

believe in the power of ideas in the digital age. However, we must remain

alert to the possibility that we could make serious mistakes along the way.

The financial system may fail, democracies may – intentionally or

accidentally – turn into surveillance societies, or we may end up fighting

wars. Therefore, with this book, I am trying to explain the opportunities

and risks ahead of us.

Signs of change are everywhere. Information technologies are

transforming the global economy at a rapid pace. In essence, we are

experiencing nothing less than a "third economic revolution",

18

leading to

an "Economy 4.0". Its effects will be at least as profound as those of the

first (agrarian to industrial) and second (industrial to service) revolutions.

The ubiquity of digital technologies – such as social media, smart devices,

the Internet of Things and Artificial Intelligence – is giving rise to a digital

society. We can no longer afford to be passive bystanders to this seismic

societal transition. We must prepare for it. But we should not regard these

changes merely as a threat to social and global stability. In fact, we are faced

with a once-in-a-century opportunity!

INTRODUCTION: THE DIGITAL SOCIETY

15

For example, the digital revolution is not only changing the way we

learn, behave, make decisions and live. It is also altering the way we

produce and consume, and our conception of ownership. Information is a

very interesting resource in that it is the basis of culture, and can be shared

as often as we like. To get more of it, we do not have to take it away from

others. We don't have to fight for it. This of course will depend on how our

economy is organized in future. In particular, in how we reward effort for

the production of data, information, knowledge and creative digital

products. We can either perpetuate the outmoded principles of the 20th

century or open the door to a smarter 21st century society. Why don't we

do the latter?

Many people are now talking about "smart homes", "smart factories",

"smart grids", and "smart cities". It is logical that we will soon have a

"smart economy" and a "smart society". Networked information systems

will enable entirely new solutions to the world's problems. One thing is

therefore clear: the world in the digital age will be very different. But even if

we can't exactly predict what the future will hold, can we at least get a

glimpse of it? I believe we can, at least to some extent, and some trends are

already emerging. Clearly, the characteristics of the future world will result

from the technological, social and evolutionary forces shaping it. The

technological drivers include Big Data, the Internet of Things and Artificial

Intelligence. The social drivers include the increase of information volumes

and networking. In addition, there are evolutionary forces that will lead to

new kinds of incentive systems and decision-making. I will attempt to

evaluate the implications of these forces, and debate the opportunities and

risks they present.

These forces, in turn, are generated by the interactions occurring within

our "anthropogenic systems", i.e. our man-made or human-influenced

techno- socio-economic-environmental systems. In order for interventions

to be beneficial, we must understand how these interactions – and the

forces they are creating – can be harnessed to our advantage in the same

way as we have learned to harness the forces of nature.

How will the digital revolution reshape our socio-economic institutions?

And what preparations can we make? While addressing these and other

questions, I will pursue a non-ideological approach, oriented neither

politicallyleft nor right, but carefully exploring novel opportunities. I will

try to explainhow we can use the digital revolution to make our society

more innovative, successful and resilient, through understanding the new

logic of the digital era to come. I will discuss how we can adapt our systems

in real-time with novel technologies. Furthermore, I will outline how we

can build an information and innovation ecosystem that can create new

jobs and opportunities for everyone.

D. HELBING: THE AUTOMATION OF SOCIETY

16

We are now ready to dive into the details of why our world is troubled

and how we can fix it – by using advanced information and communication

systems in entirely new ways. The following chapters will focus on subjects

such as prediction and control, complexity, self-organization, awareness and

coordination, responsible decision-making, real-time measurement and

feedback, systems design, innovation, reward systems, co-creation, and

collective intelligence. I hope this journey through the opportunities and

risks presented by the emerging digital society will be as exciting for you as

it is for me!

17

2: COMPLEXITY TIME BOMB

When Systems Get Out Of Control

The digital revolution produces more data, more speed, more connectivity, and more

complexity. Besides creating new opportunities, how will this change our economy and our

societies? Will it make our increasingly interdependent systems easier to control? Or are

we heading towards a systemic collapse? In order to figure out what needs to be done to fix

the world's ills, we must explore why things, as they currently stand, go wrong. The

question is, why haven't we learned how to deal with them yet?

These days, we seem to be surrounded by economic and socio-political

crises, by terrorism, conflict and crime. More and more often, the

conventional "medicines" to tackle these sorts of global problems turn out

to be inefficient or even counter-productive. It is increasingly evident that

we approach these problems with an outdated understanding of our world.

While it may still look more or less as it looked for a long time, the world

has changed inconspicuously but fundamentally.

We are used to the idea that societies must be protected from external

threats such as earthquakes, volcanic eruptions, hurricanes and military

attacks by enemies. Increasingly, however, we are threatened by different

kinds of problems that come from within the system, such as financial

instabilities, economic crises, social and political unrest, organized crime

and cybercrime, environmental change, and the spread of diseases. These

problems have become some of the greatest threats to humanity. According

to the "risk interconnection map" published by the World Economic

Forum, the greatest risks faced by our societies today are of socio-economic

and political nature.

19

These risks, including factors such as economic

inequality and governance failure, are 21st century problems which cannot

be solved with 20th century wisdom. They are larger in scale than ever

D. HELBING: THE AUTOMATION OF SOCIETY

18

before and result from the complex interdependencies in today's

anthropogenic systems. As a result, it is of paramount importance that we

develop a better understanding of the characteristics of complex dynamical

systems. To this end, I will discuss the main reasons why things go wrong,

such as unstable dynamics, cascading failures in networks, and systemic

interdependencies. And I will illustrate these problems through a large

variety of examples such as traffic jams, electrical blackouts, financial crises,

crime, wars and revolutions.

Phantom traffic jams

Complex systems can be found all around us and include phenomena

such as turbulent flows in our global weather system, decision-making

processes, opinion formation in groups, financial and economic markets,

and the evolution and spread of languages. However, we must carefully

distinguish complex systems from complicated ones. While a car, which

consists of thousands of parts, is complicated, it is easy to control

nevertheless (when it works properly). Traffic flow, on the other hand,

depends on the dynamical interactions of many cars, and forms a complex

dynamical system. These interactions produce counter-intuitive phenomena

such as "phantom traffic jams" which appear to have no cause. Such

"emergent" phenomena cannot be understood from the properties of the

single parts of the system in isolation, here the driver-vehicle units. While

many traffic jams occur for specific, identifiable reasons, such as accidents

or road works, almost everyone has also encountered situations where a

queue of vehicles seems to form "out of nothing" and where there is no

visible cause.

20

To explore the true reasons for these "phantom traffic jams", Yuki

Sugiyama and his colleagues at Nagoya University in Japan carried out an

experiment in which they asked many people to drive around a circular

track.

21

The task sounds simple, and the vehicles did in fact flow smoothly

for some time. Then, however, one of the cars caused a minor variation in

the traffic flow, which triggered stop-and-go traffic – a traffic jam that

moved backwards around the track.

While we often blame poor driving skills of others for such "phantom

traffic jams", studies in complexity science have shown that they are actually

an emergent collective phenomenon, which is the inevitable result of the

interaction between vehicles. A detailed analysis demonstrates that, if the

traffic density exceeds a certain "critical" threshold – that is, if the average

separation of the vehicles is smaller than a certain value – then even the

slightest variation in the speed of a car can eventually cause a disruption of

the traffic flow through an amplification effect. As the next driver in line

needs some time to adjust to a change in the speed of the vehicle ahead, he

or she will have to brake a bit harder to compensate for the delay. The then

COMPLEXITY TIME BOMB

19

following driver will have to break even harder, and so on. The resulting

chain reaction amplifies the initially small variation in a vehicle's speed and

this eventually produces a traffic jam which, of course, every single driver

tried to avoid.

Recessions - traffic jams in the world economy?

Economic supply chains might exhibit a similar kind of behavior, as

illustrated by John Sterman's "beer distribution game".

22

This simulates

some of the challenges of supply chain management. In this experiment,

even experienced managers will end up ordering too much stock, or will run

out of it.

23

This situation is as difficult to avoid as stop-and-go traffic. In

fact, our scientific research suggests that economic recessions can be

regarded as a kind of traffic jam in the global flow of goods, i.e. the world

economy. This news is actually somewhat heartening, since it implies that

we may be able to engineer solutions to mitigate economic recessions in a

similar way as one can reduce traffic jams by driver assistant systems. The

underlying principle will be discussed later, in the chapter on Guided Self-

Organization. In order to do this, however, one would need have real-time

data detailing the global flow and supply of materials.

Systemic instability

Crowd disasters are another, tragic example of systemic instability. Even

when every individual within a crowd is peacefully minded and tries to

avoid harming others, many people may die nevertheless. Appendix 2.1

outlines why extreme systemic outcomes can even result from normal, non-

aggressive behavior.

What do all these examples tell us? They illustrate that the natural

(re)actions of individuals will often be counterproductive.

24

Our experience

and intuition often fail to account for the complexity of highly interactive

systems, which tend to behave in unexpected ways. Such complex

dynamical systems typically consist of many interacting components which

respond to each other's behaviors. As a consequence of these interactions,

complex dynamical systems tend to self-organize. That is, a collective

dynamics may develop (such as stop-and-go traffic), which is different from

the natural behavior of the system components when they are separated

from each other (such as drivers who don't like to stop). In other words,

the overall system may have new characteristics that are distinct from those

of its components. This can result, for example, in "chaotic" or "turbulent"

dynamics.

Group dynamics and mass psychology may be viewed as typical

examples of spontaneously emerging collective dynamics occurring in

crowds. What is it that makes a crowd turn "mad", violent, or cruel? For

example, after the London riots of 2011, people have asked how it was

D. HELBING: THE AUTOMATION OF SOCIETY

20

possible that teachers and the daughters of millionaires – people you would

not expect to be criminals – were participating in the looting? Did they

suddenly develop criminal minds when the demonstrations against police

violence turned into riots? Possibly, but not necessarily so.

The emergence of new properties requires an interaction-oriented

perspective

When many components of a complex dynamical system interact, they

frequently create new kinds of structures, properties or functions in a self-

organized way. To describe newly resulting characteristics of the system, the

term "emergent phenomena" is often used. For example, water feels wet,

extinguishes fire, and freezes at a particular temperature. But we would not

expect this based on the examination of single water molecules.

Therefore, complex dynamical systems may show surprising behaviors.

They cannot be steered like a car. In the above traffic flow example,

although the sole aim of participants was to drive continuously at a

reasonably high speed, a phantom traffic jam occurred due to the

interactions between cars. While it is straightforward to control a car, it may

be impossible for individual drivers to control the collective dynamics of

traffic flow, which is the result of the interactions of many cars.

Beware of strongly coupled systems!

Instability is just one possible problem of complex dynamical systems. It

occurs when the characteristic parameters of a system cross certain critical

thresholds. If a system becomes unstable, small deviations from the normal

behavior are amplified. Such amplification is often based on feedback

loops, which cause a mutual reinforcement. If one amplification effect

triggers others, a chain reaction may occur and a minor, random variation

may be enough to trigger an unstoppable domino effect. In case of systemic

instability, as I have demonstrated for the example of phantom traffic jams,

the system will inevitably get out of control sooner or later, no matter how

hard we try to prevent this. Consequently, we need to identify and avoid

conditions under which systems behave in an unstable way.

In many cases, strongly coupled interactions are a recipe for disaster or

other undesirable outcomes.

25

While our intuition usually works well for

problems that are related to weakly coupled systems (in which the overall

system can be understood as being the sum of its parts and their

properties), the behaviors of complex dynamical systems can change

dramatically, if the interactions among their components are strong. In

other words, these systems often behave in counter-intuitive ways, so that

conventional wisdom tends to be ineffective for managing them.

Unintended consequences or side effects are common.

What further differences do strong interactions make? First, they may

cause larger variability and faster changes, particularly if there are "positive

COMPLEXITY TIME BOMB

21

feedbacks" that lead to mutual acceleration. Second, the behavior of the

complex system can be hard to predict, making it difficult to plan for the

future. Third, strongly connected systems tend to show strong correlations

between the behaviors of (some of) its components. Fourth, the

possibilities to control the system from the outside or through the behavior

of single system components are limited, as the system-immanent

interactions may have a stronger influence. Fifth, extreme events occur

more often than expected and they may affect the entire system.

i

In spite of all this, most people still have a component-oriented

worldview, centered on individuals rather than groups and interdependent

events. This often leads to "obvious"

26

but wrong conclusions. For example,

we praise heroes when things go well and search for scapegoats when

something goes wrong. Yet, the discussion above has clearly shown how

difficult it is for individuals to control the outcome of a complex dynamical

system, if the interactions between its components are strong. This fact can

also be illustrated by politics.

Why do politicians, besides managers, have (on average) the worst

reputation of all professions? This is probably because we think they are

hypocritical; we elect them on the basis of the ideas and policies they

publicly voice, but then they often do something else. This apparent

hypocrisy is a consequence of the fact that politicians are subject to many

strong interactions with lobbyists and interest groups, who have diverse

points of view. All of these groups push the politicians into different

directions. In many cases, this forces them to make decisions that are not

compatible with their own point of view – a fact, which is hard to accept

for the voters. However, if we believe that democracy is not just about

elections every few years, but also about a continuous consolidation

between citizens and their elected representatives, it could be argued that it

would be undemocratic for politicians to unreservedly place their own

personal convictions above the concerns of private citizens and businesses

in the systems they represent. Managers of companies find themselves in

similar situations. They are exposed to many different factors they must

consider. This kind of interaction-based decision-making extends far

beyond boardrooms and parliaments. Think of the decision-dynamics in

families: if it were easy to control, there would be probably less divorces...

Crime is another example of social systems getting out of control.

ii

This

can happen both on an individual and collective level. Many crimes,

i

In social systems, this may lead to an erosion of trust in private and public institutions,

which can create social, political or economic instability.

ii

The traditional explanation is that a crime is committed if the expected reward is larger

than the likely punishment, multiplied by the probability of being caught and convicted.

D. HELBING: THE AUTOMATION OF SOCIETY

22

including murders, are committed by average people, rather than career

criminals (and even in countries with death penalty).

27

A closer inspection

shows that many crimes correlate with the circumstances individuals find

themselves in. For example, group dynamics often plays an important role.

Many scientific studies also show that the socio-economic conditions are a

strong determining factor of crime. Therefore, in order to counter crime, it

might be more effective to change these socio-economic conditions rather

than sending more people to jail. I say this with one eye on the price we

have to pay to maintain a large prison population – a single prisoner costs

more than the salary of a postdoctoral researcher with a PhD degree! It

worries even more that prisons and containment camps have apparently

become places where criminal organizations are formed and terrorist plots

are planned.

28

In fact, our own studies suggest that many crimes spread by

imitation.

29

Cascade effects in complex networks

To make matters worse, besides the dynamic instability caused by

amplification effects, complex dynamical systems may produce even bigger

problems. Worldwide trade, air traffic, the Internet, mobile phones, and

social media have made everything much more convenient – and

connected. This has created many new opportunities, but everything now

depends on a lot more things. What are the implications of this increased

level of interdependency? Today, a single tweet can send stock markets

spinning. A controversial Youtube video can trigger a riot that kills dozens of

people.

30

Increasingly, our decisions can have consequences on the other

side of the globe, many of which may be unintended. For example, the

rapid spread of emerging epidemics is largely a result of the scale of global

air traffic today, and this has serious repercussions for global health, social

welfare and economic systems.

It is being said that "the road to hell is paved with good intentions". By

networking our world, have we inadvertently created conditions in which

Theoretically, therefore, it should be possible to eliminate all crime simply by raising the

punishment. If all criminals were rational egoists, who constantly calculated their expected

payoff from crime, this idea should even work. Assuming sufficiently high conviction rates,

the strong punishment would make crime "unattractive" because of the expectation of

making a "loss". However, empirical evidence questions these simple assumptions. On the

one hand, people don't behave entirely rational. For example, they don't usually pick

pockets, even though they could often escape punishment. On the other hand, deterrence

strategies are surprisingly ineffective in most countries and high crime rates are often

recurrent. For example, even though the USA have 10 times more prisoners than most

European countries, the rates of various crimes are still much higher. Therefore, the

conventional understanding of crime is wrong.

COMPLEXITY TIME BOMB

23

disaster is more likely to spread? In 2011 alone, at least three cascading

failures with global impact were happening, thereby changing the face of

the world and the global balance of power: The world economic crisis, the

Arab spring and the combination of an earthquake, a tsunami and a nuclear

disaster in Japan. In 2014, the world was threatened by the spread of Ebola,

the crisis in Ukraine, and the conflict with the Islamic State (IS). In the

following subsections, I will therefore discuss some examples of cascade

effects in more detail.

Large-scale power blackouts

On November 4, 2006, a power line was temporarily turned off in Ems,

Germany, to facilitate the transfer of a Norwegian ship. Within minutes,

this caused a blackout in many regions all over Europe, from Germany to

Portugal! Nobody expected this to happen. Before the line was switched

off, a computer simulation predicted that the power grid would still operate

well without the line. However, the scenario analysis did not account for the

possibility that another line would spontaneously fail. In the end, a local

overload in Northwest Germany triggered emergency switch-offs all over

Europe, creating a cascade effect with pretty astonishing results. Blackouts

occurred in regions thousands of kilometers away.

iii

Can we ever hope to understand such strange behavior? In fact, a

computer-based simulation study of the European power grid recently

managed to reproduce quite similar effects.

31

It demonstrated that the

failure of a few network nodes in Spain could create an unexpected

blackout several thousand kilometers away in Eastern Europe, while the

electricity network in Spain would still work.

32

Furthermore, increasing the

capacity of certain parts of the power grid could worsen the situation and

cause an even greater blackout. Therefore, the weak elements of the system

serve an important function: they are "circuit breakers", which can interrupt

the failure cascade. This is an important fact to remember.

From bankruptcy cascades to financial crisis

The sudden financial meltdown in 2008 is another example of a crisis

that hit many companies and people by surprise. In a presidential address to

the American Economic Association in 2003, Robert Lucas said:

"[The] central problem of depression-prevention has been solved."

Similarly, Ben Bernanke, the former chairman of the Federal Reserve

Board, held a longstanding belief that the economy was both well

understood and in sound financial shape. In September 2007, Ric Mishkin

iii

while other areas close to the region of initial overload were not affected at all

D. HELBING: THE AUTOMATION OF SOCIETY

24

(*1951), a professor at Columbia Business School and then a member of

the Board of Governors of the US Federal Reserve System, made another

interesting statement, reflecting widespread beliefs at this time:

"Fortunately, the overall financial system appears to be in good health,

and the U.S. banking system is well positioned to withstand stressful market

conditions."

As we all know with the benefit of hindsight, things turned out very

differently. A banking crisis occurred only shortly later. It started locally,

when a real estate bubble burst, which had formed in the West of the USA.

As it was a regional problem, most people thought it could be easily

contained. But the mortgage crises had spillover effects on stock markets.

Certain financial derivatives could not be sold anymore and became "toxic

assets". Eventually, hundreds of banks all over the US went bankrupt. How

could this happen? A video produced by professor Frank Schweitzer and

others presents an impressive visualization of the chronology of

bankruptcies in the USA after Lehman Brothers collapsed.

33

Apparently,

the default of a single bank triggered a massive cascading failure in the

financial sector. In the end, hundreds of billions of dollars were lost.

The video mentioned above looks surprisingly similar to another one,

which I often use to illustrate cascade effects.

34

It shows an experiment in

which many table tennis balls are placed on top of mousetraps. The

experiment impressively demonstrates that a single local disruption can

mess up an entire system. The video illustrates chain reactions, which are

the basis of atomic bombs or nuclear fission reactors. As we know, such

cascade effects can be technologically controlled in principle, if a certain

critical mass (or "critical interaction strength") is not exceeded.

Nevertheless, these processes can sometimes get out of control, mostly in

unexpected ways. The nuclear disasters in Chernobyl and Fukushima are

well-known examples of this. We must, therefore, be extremely careful with

large-scale systems featuring cascade effects.

A world economic crisis results

As we know, the cascading failure of banks mentioned above was just

the beginning of an even bigger problem. It subsequently triggered a global

economic and public spending crisis. Eventually, the financial crisis caused

a worldwide damage of $15 trillion

35

– an amount a hundred times as big as

the initial real estate problem. The events even threaten the stability of the

Euro currency and the EU. Several countries including Greece, Ireland,

Portugal, Spain, Italy and the US were on the verge of bankruptcy. As a

consequence, many countries are suffering from unprecedented

unemployment rates, amounting to more than 24 million unemployed

people in Europe. In some countries, more than 50 percent of young

COMPLEXITY TIME BOMB

25

people do not have a job. In many regions, this has caused social unrest,

political extremism and increased rates of suicide, crime and violence.

Unfortunately, the failure cascade hasn't been stopped, yet. There is a

long way to go until we fully recover from the financial crisis and from the

public and private debts accumulated in the past years. If we can't

overcome this problem soon, it even has the potential to endanger peace,

democratic principles and cultural values, as I pointed out in a letter to

George Soros in 2010.

36

Looking at the situation in Ukraine, we are perhaps

seeing this scenario already.

While all of this is now plausible with the benefit of hindsight, the

failure of conventional wisdom to provide an advanced understanding of

events is reflected by the following quote, made by the former president of

the European Central Bank, Jean-Claude Trichet in November 2010:

"When the crisis came, the serious limitations of existing economic and financial

models immediately became apparent. Arbitrage broke down in many market segments,

as markets froze and market participants were gripped by panic. Macro models failed to

predict the crisis and seemed incapable of explaining what was happening to the economy

in a convincing manner. As a policy-maker during the crisis, I found the available models

of limited help. In fact, I would go further: in the face of the crisis, we felt abandoned by

conventional tools." Similarly, Ben Bernanke summarized in May 2010: “The brief

market plunge was just an example of how complex and chaotic, in a formal sense, these

systems have become… What happened in the stock market is just a little example of

how things can cascade, or how technology can interact with market panic.”

Even leading scientists found it difficult to make sense of the crisis. In a

letter on July 22, 2009 to the Queen of England, the British Academy came

to the conclusion:

37

"When Your Majesty visited the London School of Economics last November, you

quite rightly asked: why had nobody noticed that the credit crunch was on its way? ... So

where was the problem? Everyone seemed to be doing their own job properly on its own

merit. And according to standard measures of success, they were often doing it well. The

failure was to see how collectively this added up to a series of interconnected imbalances

over which no single authority had jurisdiction. ... Individual risks may rightly have been

viewed as small, but the risk to the system as a whole was vast. ... So in summary ... the

failure to foresee the timing, extent and severity of the crisis … was principally the failure

of the collective imagination of many bright people to understand the risks to the systems

as a whole."

Thus, was nobody responsible for the financial crisis in the end? Or do

we all have to accept some responsibility, given that these problems are

collective outcomes of a huge number of individual (inter)actions? And

D. HELBING: THE AUTOMATION OF SOCIETY

26

how can we differentiate the degree of responsibility of different individuals

or firms? This is certainly an important question worth thinking about.

It is also interesting to ask, whether complexity science could have

forecasted the financial crisis? In fact, I followed the stock markets closely

before the crash and noticed strong price fluctuations, which I interpreted

as advanced warning signals of an impending financial crash. For this

reason, I sold my stocks in late 2007, while I was sitting in an airport

lounge, waiting for my connection flight. In spring 2008, about half a year

before the collapse of Lehman brothers, James Breiding, Markus Christen

and I wrote an article taking a complexity science view on the financial

system. We came to the conclusion that the financial system was in the

process of destabilization. We believed that the increased level of

complexity in the financial system was a major problem and that it made

the financial system more vulnerable to cascade effects, as was later also

stressed by Andrew Haldane, the Chief Economist and Executive Director

at the Bank of England.

In spring 2008, we were so worried about these trends that we felt we

had to alert the public. At that time, however, none of the newspapers we

contacted were ready to publish our essay. "It's too complicated for our readers"

was the response. We responded that "nothing can prevent a financial crisis, if you cannot

make this understandable to your readers". With depressing inevitability, the

financial crisis came. Although it gave us no pleasure to be proved right

when the consequences were so dire, a manager from McKinsey's UK

office commented six months later that our analysis was the best he had

seen.

Of course, some far more prominent public figures also saw the

financial crisis coming. The legendary investor Warren Buffet, for example,

warned of the catastrophic risks created by large-scale investment in

financial derivatives. Back in 2002 he wrote:

"Many people argue that derivatives reduce systemic problems, in that

participants who can't bear certain risks are able to transfer them to

stronger hands. These people believe that derivatives act to stabilize the

economy, facilitate trade, and eliminate bumps for individual participants.

On a micro level, what they say is often true. I believe, however, that the

macro picture is dangerous and getting more so. ... The derivatives genie is

now well out of the bottle, and these instruments will almost certainly

multiply in variety and number until some event makes their toxicity clear.

Central banks and governments have so far found no effective way to

control, or even monitor, the risks posed by these contracts. In my view,

derivatives are financial weapons of mass destruction, carrying dangers that,

while now latent, are potentially lethal."

As we know, it still took five years until the "investment time bomb"

exploded, but then it caused trillions of dollars of losses to our economy.

COMPLEXITY TIME BOMB

27

Fundamental ("radical") uncertainty

In liquid financial markets and many other systems, which are difficult

to predict such as our weather, we can still determine the probability of

close enough events, at least approximately. Thus, we can make

probabilistic forecasts such as "there is a 5 percent chance to lose more

than half of my money when selling my stocks in 6 months, but a 70

percent chance that I will make a good profit..." It is then possible to

determine the expected loss (or gain) implied by likely actions and events.

For this purpose, the damage or gain of each possible event is multiplied by

its probability, and the numbers are added together to predict the expected

damage or gain. In principle, we could do this for all the actions we might

take, in order to determine the one that minimizes damage or maximizes

gain. The problem is that it is often impractical to calculate these

probabilities. With the increasing availability of data, this problem may

recede, but it will remain difficult or impossible to determine the

probabilities of "extreme events". By their very nature, there tends to be

little precedent for such rare events, which means that the empirical basis is

too small to determine their probabilities.

In addition, it may be completely impossible to calculate the expected

damage incurred by a problem in a large (e.g. global) system. Such

"fundamental" or "radical" uncertainty can result from cascade effects,

whereby one problem is likely to trigger other problems, leading to a

progressive increase in the overall damage. In principle, the overall losses

may not be quantifiable in such situations at all. This means in practice that

the actual damage might be either insignificant (in the best case) or

unbounded (in the worst case), or anything in between. In an extreme case,

this might lead to a collapse of the entire system, as we know it from the

collapse of historical empires and civilizations.

38

Explosive epidemics

When studying the spread of diseases, the outcome is highly dependent

on the degree of physical interactions between people who may infect each

other. A few additional airline routes might make the difference between a

case in which a disease is contained, and a case which develops into a

devastating global pandemic.

39

The threat of epidemic cascade effects might

be even worse if damage which occurs early on reduces the ability of the

system to withstand problems later. For example, assume a health system in

which the financial or medical resources are limited by the number of

healthy individuals who produce them. In such a case, it can happen that

the resources needed to heal the disease are increasingly used up, such that

the epidemic finally spreads explosively. A computer-based study, which

Lucas Böttcher, Olivia Woolley-Meza, Nuno Araujo, Hans Hermann and I

D. HELBING: THE AUTOMATION OF SOCIETY

28

performed, shows that the dynamics in such a system can change

dramatically and unexpectedly.

40

Thus, have we built global networks which

we can neither predict nor control?

Systemic interdependence

Recently, Shlomo Havlin and others made a further important

discovery. They revealed that networks of networks can be particularly

vulnerable to disruptions.

41

A typical example of this is the interdependence

between electrical and communication networks. Another example, which

illustrates the global interdependence between natural, energy, climate,

financial and political systems is provided by the Tohoku earthquake in

Japan in 2011. The earthquake caused a tsunami which triggered a chain

reaction and a nuclear disaster in several reactors at Fukushima. Soon

afterwards, Germany and Switzerland decided to exit nuclear power

production over the next decade(s). However, alternative energy sources are

also problematic, as European gas supply depends on geopolitical regions

which might not be fully reliable.

42

Likewise, Europe’s DESERTEC project – a planned €1000 billion

investment in solar energy infrastructure – has been practically given up due

to another unexpected event, the Arab Spring. This uprising was triggered

by high food prices, which in turn was partially caused by biofuel

production. While biofuels were intended to improve the global CO2

balance, their production decreased the production of food, making it more

expensive. The increased food prices were further amplified by financial

speculation. Hence, the energy system, the political system, the social

system, the food system and the financial system have all become closely

interdependent, making our world increasingly vulnerable to disruption.

Have humans created a "complexity time bomb"?

For a long time, problems such as crowd disasters and financial crashes

have puzzled humanity. Sometimes, they have even been regarded as "acts

of God" or "black swans"

43

that we had to endure. But problems like these

should not be simply put down to "bad luck". They are often the

consequence of a flawed understanding of the counter-intuitive way in

which complex systems behave. Fatal errors and the repetition of previous

mistakes are frequently the result of an outdated way of thinking. However,

complexity science allows us to understand how and when complex

dynamical systems get out of control.

If a system is unstable, we will see amplification effects, such that a local

problem can lead to a cascading failure, which creates many further

problems down the line. For this reason, the degree of interaction between

the system components is crucial. Overall, complex dynamical systems

become unstable, if the interactions between their components get stronger

COMPLEXITY TIME BOMB

29

than frictional effects, or if the damage resulting from the degradation of

system components occurs faster than they can recover. As a result, the

timing of processes can also play a key role in determining whether the

overall system will remain stable. This means that delays in adaptation

processes can often lead to systemic instabilities and loss of control (see

Appendix 2.2).

We have further seen that an unstable complex system will sooner or

later get out of control, even if everyone is well-informed and well-trained,

uses advanced technology and has the best intentions. And finally, we have

learned that complex dynamical systems with strong internal interactions or

a high level of connectivity tend to be unstable. As our increasingly

interdependent world is characterized by a myriad of globalized links, it is

necessary to discuss the potential consequences. We must raise a

fundamental question which has mammoth implications for the viability of

our current economic and political systems: have humans inadvertently

produced a "complexity time bomb", i.e. a global system which will

inevitably get out of control?25

In fact, for certain kinds of networks, the potential chain reaction of

cascading failures bears a disturbing and ominous resemblance to that of

nuclear fission. Such processes are difficult to control and catastrophic

damage is a realistic scenario. Given this similarity with explosive processes,

is it possible that our global anthropogenic systems will similarly get out of

control at some point? When considering this possibility, we need to bear in

mind that the speed of a destructive cascade effect might be slow, and the

process may not remind of an explosion. Nevertheless, the process may be

hard to stop and may ultimately lead to a systemic failure.

iv

So, what kinds of global catastrophes might complex societies face? A

collapse of the global information and communication system or of the

world economy? Global pandemics? Unsustainable growth, demographic or

environmental change? A global food or energy crisis? A clash of cultures?

Another world war? A societal shift, triggered by technological innovation?

In the most likely scenario, we will witness a combination of several of

these contagious phenomena. The World Economic Forum calls this the

"perfect storm"19, and the OECD has expressed similar concerns.

Unintended wars and revolutions

It is important to realize that large-scale conflicts, revolutions and wars

can also be unintended outcomes of systemic interdependencies and

iv

For example, the underlying processes which cause crowd disasters are slow, but deadly

nonetheless.

D. HELBING: THE AUTOMATION OF SOCIETY

30

instabilities. Remember that phantom traffic jams were unintended

consequences of interactions. Similarly, wars and revolutions can happen

even if nobody wants them. There is a tendency to characterize these events

as the deeds of particular historical figures. But this trivializes and

personalizes such phenomena in a way which distracts from their true,

systemic nature.

It is essential to recognize that complex dynamical systems usually resist

change if they are close to a stable equilibrium. This effect is known as

Goodhart's law (1975), Le Chatelier’s principle (1850-1936) or "illusion of

control". Individual factors and randomness only affect complex dynamical

systems if they are driven to a tipping point,

v

where they become unstable.

In other words, the much-heralded individuals to whom so much

attention is afforded in our history books were only able to influence

history because much bigger systems beyond their control had already

become critically unstable.

44

For example, historians now increasingly

recognize that World War I was a largely unintended consequence of a

chain of events. Moreover, World War II was preceded by a financial crisis

and recession, which destabilized the German economic, social and political

system. Ultimately, this made it possible for an individual to become

influential enough to drive the world to the brink of extinction.

Unfortunately, civilization is still vulnerable today and a large-scale war may

happen again. This is even likely if we don't quickly change the way we

manage our world.

Typically, the unintended path towards war is as follows. Initially,

resources become scarce due to a disruption such as a serious economic

crisis. Then, the resulting competition for limited resources leads to an

increase in conflict, violence, crime and corruption. Human solidarity and

mutual tolerance are eroded, creating a polarized society. This causes

further dissatisfaction and social turmoil. People get frustrated with the

system, calling for leadership and order. Political extremists emerge, who

scapegoat minorities for social and economic problems. This decreases

socio-economic diversity, which reduces innovation and further hinders the

economy. Eventually, the well-balanced "socio-economic ecosystem"

collapses, such that an orderly system of resource allocation becomes

impossible. As resources become scarcer, this creates an increasing "need"

for nationalism or even for an external enemy to unify the fractured society.

In the end, as a result of further escalation, war seems to be the only viable

"solution" to overcome the crisis, but instead, it mostly leads to large-scale

destruction.

v

Such tipping points are also often called "critical points".

COMPLEXITY TIME BOMB

31

Revolutionary systemic shifts

Note that a revolution, too, can be the result of systemic instability.

Hence, a revolution is not necessarily set in motion by a "revolutionary

leader", who challenges the political establishment. The breakdown of the

former German Democratic Republic (GDR) and some Arab Spring

revolutions have shown that uprisings may even start in the absence of a

clearly identifiable political opponent. On the one hand, this is the reason

why such revolutions cannot be stopped by killing or imprisoning a few

individuals. On the other hand, the Arab Spring took secret services

throughout the world by surprise precisely because there were no

revolutionary leaders. This created complications for countries which

wished to assist these uprisings, as they did not know whom to interact with

for international support.

It is more instructive to imagine such revolutions as a result of situations

in which the interests of government representatives and those of the

people (or particular societal groups) have drifted away from each other.

Similarly to the tensions created by a drift of the Earth's tectonic plates,

such an unstable situation is sooner or later followed by an "earthquake-

like" release of tension (the "revolution"), resulting in a re-balancing of

forces. To reiterate, contrary to the conventional wisdom which assumes

that revolutionary leaders are responsible for political instability, it is due to

an existing systemic instability that these individuals can become influential.

To put it succinctly, in most cases, revolutionaries don't create revolutions,

but systemic instabilities do. These instabilities are typically created by the

politics of the old regime. Therefore, we must ask ourselves how well our

society balances the interests of different groups today and how well it

manages to adapt to a world, which is rapidly changing due to demographic,

environmental and technological change?

Conclusion

It is obvious that there are many problems ahead of us. Most of them

result from the complexity of the systems humans have created. But how

can we master all these problems? Is it a lost battle against complexity, or

do we have to pursue a new, entirely different strategy? Do we perhaps

even need to change our way of thinking? And how can we innovate,

before it is too late? The next chapters will try to answer these questions...

APPENDIX 2.1: How harmless behavior can become critical

In the case of traffic flow, we have seen that a system can get out of

control when the interaction strength (e.g. the density) is too high. Why can

a change in the density make normal and "harmless" behavior become

D. HELBING: THE AUTOMATION OF SOCIETY

32

uncontrollable? To understand this better, Roman Mani, Lucas Böttcher,

Hans J. Herrmann, and I studied collisions in a system of equally sized

particles moving in one dimension,

45

which is similar to Newton's Cradle.

46

We assumed that the particles tended to oscillate elastically around equally

spaced equilibrium points, while being exposed to random forces generated

by the environment.

If the distance between the equilibrium points of neighboring particles is

large enough, each particle oscillates around its equilibrium point with

normally distributed velocities, and all particles have the same small

variance in speed. However, when the separation between the equilibrium

points reaches the diameter of the particles, we find a cascade-like

transmission of momentum between particles.

47

Surprisingly, the variance

of particle speeds rapidly increases towards the boundaries — it could even

go to infinity with increasing system size. Due to cascading particle

interactions, this makes their speeds unpredictable and uncontrollable.

While every particle in separation performs a normal dynamics, which is not

excessive at all, their interactions can cause an extreme behavior of the

system.

APPENDIX 2.2: Loss of synchronization in hierarchical systems

When many socio-economic processes are happening simultaneously

while having feedbacks on each other, a puzzling kind of systemic instability

can occur, which is highly relevant for our complex societies, since many

socio-economic processes happen at an increasing pace.

For the sake of illustration, let us first discuss hierarchically organized

systems in physics. There, elementary particles form atoms, atoms form

chemical compounds, these form solid bodies, and together they may form

a planet, which is part of a planetary system, and a galaxy. Similarly, we

know from biology and the social sciences that organs are constituted of

cells, which collectively may form a human body. In turn, humans tend to

organize themselves into groups, cities, organizations and nations.

Importantly, the stability of such hierarchies is based on two important

principles. First, the forces are strongest at the bottom, and second, the

changes are slowest at the top. In other words, adjustment processes in

these systems are faster at lower hierarchical levels (such as the atoms) as

compared to higher ones (such as planetary systems). This means that lower

level variables can adjust quickly to the constraints set by the higher level

variables. As a result, the higher levels basically control the lower levels and

the system remains stable. Similarly, social groups tend to take decisions

more slowly than the individuals who form them. Likewise, organizations

and states tend to change more slowly than the individuals who form them

(at least this is how it used to be in the past).

COMPLEXITY TIME BOMB

33

Such "time-scale separation" implies that the dynamics of a system are

determined only by relatively few variables, which are typically located at

the higher levels of the hierarchy. Monarchies and oligarchies are good

examples for this. In current-day socio-political and economic systems,

however, the higher hierarchical levels sometimes change so fast that the

lower levels have difficulties to keep pace. Laws are now often enacted

more quickly than companies and people can adapt. In the long run, this is

likely to cause systemic instability, as time-scale separation is destroyed, so

that many more variables begin to influence the dynamics of the system.

Such attempts to make mutual adjustments on different hierarchical levels

could potentially lead to turbulence, "chaos", breakdown of

synchronization, or fragmentation of the system. In fact, while it is known

that delays in adaptation can destabilize a system, we are putting many of

our problems on the long finger (e.g. public debts, implications of

demographic change, nuclear waste, or climate change). This creates a

concrete danger that our society will eventually get out of control.

34

35

3: SOCIAL FORCES

Revealing the causes of success and disaster

Complex systems can behave in unpredictable ways and cause a lot of trouble. But it

doesn't have to be like this. Their behavior depends on the interactions between the system

components, the strength of these interactions, and the institutional settings. Consequently,

for a complex system to work well, it is important to understand the factors that drive its

dynamics. In physics, many phenomena have been understood in terms of forces, which can

be measured by suitable procedures. In a similar way, the success or failure of socio-

economic systems depends on hidden forces, too. Thanks to new data about our world we

can now measure the forces driving socio-economic change. This will allow us to act more

successfully in future.

Societies around the world are suffering from financial crises, crime,

conflicts, wars, and revolutions. These "societal ills" do not occur by

chance, but for a reason. The fact that they are happening time and again

proves that there are hidden causes, which we haven't understood

sufficiently well. This is why we keep failing to cope with these problems.

In future, however, we will be able to understand societal problems and

cure them. This will be akin to the discovery of the x-ray by Wilhelm

Conrad Röntgen (1845-1923), which helped to reveal the causes of many

diseases and to cure billions of people.

i

Given that we are now living in a Big Data age, will we soon be able to

answer all our questions and find the best possible course of action in every

situation? Of course, it's far from clear that this dream will ever come true.

i

Nevertheless, it is important to avoid an overdose of radiation, which can be very harmful.

In a similar way, mass surveillance can be harmful to society, as I will show later. When

measuring what's going on in society, it is therefore, important to respect privacy and the

fundamental right of informational self-determination. The Nervousnet platform featured in

this chapter is taking this into account.

D. HELBING: THE AUTOMATION OF SOCIETY

36

However, the growing amount of data about our world will certainly allow

us to measure the hidden forces behind our global technological, social,

economic and environmental systems, just as microscopes and telescopes

enabled us to discover and understand the micro- and macro-cosmos –

from cells to stars – in the past. Similarly to how we have built elementary

particle accelerators to discover the forces that keep our world together, we

can now create "socioscopes" to reveal the principles that make our society

succeed or fail.

Measuring the world 2.0

It's a sad but well-known fact that the loss of control over a system

often results from a lack of knowledge about the rules governing it.

Therefore, it is important that we learn to measure and understand the

hidden "forces" determining changes in the world around us. This will

eventually put us in a position where we can harness these forces to

overcome systemic instability and create complex dynamical systems with

particular structures, properties and functions.

Remember that some of the greatest discoveries in human history were

made by measuring the world. We have discovered new continents and

cultures. We have reached out to the skies and explored our universe to

discover black holes, dark matter, and new worlds. Now, the Internet is

offering entirely new ways to quantify what's happening on our Earth. By

analyzing the sentiment of blogs, Facebook posts, or tweets, we can visualize

human emotions such as happiness.

48

Furthermore, it is possible to get a

picture of the social, economic, and political "climate", by identifying the

subjects that people publicly discuss.

49

By mining data on the Web, we can

also map social and economic indicators. This includes quantities such as

the gross domestic product per capita

50

or the levels of violence and crime,

51

highly resolved according to geographic location and time.

52

It is even

feasible to digitally re-construct our three-dimensional world based on the

photos that people upload on platforms such as flicker.

53

We can now create "Financial Crisis Observatories" to detect the

likelihood of financial bubbles and crashes.

54

We can map crises and risks to

help first-aid teams in regions struck by disasters.

55

We can analyze and

visualize the production of knowledge and the spread of scientific concepts,

as I did it with Amin Mazloumian, Katy Börner, Tobias Kuhn, Christian

Schulz and others

56

,

57

(see Fig. 3.1). Furthermore, it is fascinating to examine

the way culture has spread across the world over the centuries, as

Maximilian Schich Laszlo Barabasi and I did it together with others.

58

Inspired by Wikipedia and OpenStreetMap, we might now create an

OpenResourcesMap to visualize the resources of the world and who uses

them. This could help, for example, to reduce undesirable shortages. In

addition, we could produce an OpenEcosystemsMap to depict

SOCIAL FORCES

37

environmental change and who causes it. What else can we do? Let me

elaborate a health-related example.

FIGURE 3.1: Illustration of scientific productivity and impact.

59

Monitoring the flu and other diseases

Pandemics are a major threat to humanity. Some of them have killed

millions of people. The Spanish flu in 1918 was a shocking example of this.

In fact, such pandemics are expected to happen time and again because

viruses keep mutating, such that our immune systems might be unprepared.

For instance, the world was caught by surprise by the recent Ebola

outbreak.

To contain the spread of epidemics, the World Health Organization

(WHO) is continuously monitoring emerging diseases. It takes about two

weeks to collect the data from all the hospitals in the world, meaning that

each overview of the current situation is two weeks out of date. However,

Google Flu Trends pioneered an approach called "nowcasting", which was

celebrated as major success of Big Data analytics. It was claimed that it is

possible to estimate the number of infections in real-time, based on the

search queries of Google users. The underlying idea was that queries such as

"I have a headache" or "I don't feel well" or "I have a fever", and so on,

might indicate that the user has the flu. While this makes a lot of sense, the

Google Flu approach was recently found to be unreliable, partly because

Google constantly changes its search algorithms and also because

advertisements bias people's behavior.

60

D. HELBING: THE AUTOMATION OF SOCIETY

38

Flu prediction better than Google

Fortunately, a model which uses much less data can be used to analyze

how a disease spreads, namely by augmenting data of infections with a

model based on air travel data. Dirk Brockmann and I found this approach

in 2012/13. About ten years back, Dirk started to investigate the spread of

disease by analyzing the time and geographic location of infections using

computer simulations. He also analyzed the paths of dollar bills in his

famous "Where is George?" study.

61

But when visualizing the spatio-

temporal spread of epidemics, the patterns looked frustratingly chaotic and

unpredictable. The relationship between the arrival time of a new disease as

a function of the distance from the place where it originated was so

scattered that it was hard to make much sense of the data. Eventually,

however, it became clear that this problem resulted from the high volume

of passenger air travel. Thus, Dirk had the idea to define an "effective

distance", based on the volume of travel between all airports in the world,

and to study the spread of disease as a function of this alternative measure

of distance.

62

In effective distance, two airports such as New York City and

Frankfurt are close to each other because of the large passenger flows

connecting them, while two nearby cities without any direct flights between

them might be largely separated.

ii

Dirk Brockmann and I started to collaborate in 2011, when Germany

was witnessing the spread of the deadly, food-borne EHEC epidemic. I got

in touch with Dirk and suggested that we could combine a model of the

spread of epidemics with a model of food supply chains. In this way, we

wanted to identify the location where the disease originated, which was

unknown at that time. Unfortunately, we could not obtain proper supply

chain data at that time. But our discussion triggered a number of important

ideas. In particular, the research activities shifted from predicting the spread

of diseases toward detecting the locations where they originate.

In fact, when analyzing empirical data of infections as a function of

effective distance from the perspective of all airports worldwide, we found

that the most circular spreading pattern identifies the most likely origin of

ii

Independently of Dirk Brockmann's activities, I became interested in the modeling of

epidemic spread back in 2002. In the wake of the September 11 attacks the year before, there

were fears that terrorists could use anthrax or other deadly germs to threaten the USA and

the rest of the world. At this time, I proposed to Otto Schily, the then German Minister of

Internal Affairs, to build a self-calibrating epidemic simulator to predict the spread of

pandemics. Directly after the outbreak of a disease, accurate data about infection and

recovery rates is often not available. Thus, the idea was that a self-adaptive calibration model

could produce increasingly accurate predictions, as more data became available. At that time,

I received a letter stating that such an approach was not feasible. But of course, it was!

SOCIAL FORCES

39

the disease. More importantly, however, once the location of origin of a

disease is known, one can use the circular spreading dynamics as a function

of effective distance to predict the order in which cities will be hit by a

pandemic.

iii

This helps to put medical drugs (such as immunization shots)

and doctors in place where they are most effective in countering the impact

and spread of the disease.

When Ebola broke out, Dirk furthermore used the method discussed

above to make early predictions about possible cases in other countries.

This helped to inform international preparations to contain the virus.

63

However, I would also like to highlight here the fantastic research teams of

Alessandro Vespignani and Vittoria Colizza, both partners of the FuturICT

initiative. To predict the spread of diseases, they have built a very detailed

and sophisticated simulator. Whenever a disease breaks out, this simulator

can be used to test the effectiveness of countermeasures and inform policy-

makers around the world.

64

It was found, for example, that closing down

some airline connections can only delay the spread of the disease, while the

best way for industrialized countries to protect themselves from diseases

such as Ebola is to spend their money on fighting the disease in those

countries that are suffering from the disease first.

65

Creating a Planetary Nervous System as a Citizen Web

Very soon, we will not only have maps, which aggregate data from the

past and represent them as a function of space, time or network

interdependencies. We will also have systems, which deliver real-time

answers. We will be able to ask questions, which trigger tailored

measurements to answer them. "How is the traffic on Oxford Street in

London?" "How is the weather in Moscow?" "How could investment

decisions and consumer choices be affected?" "How happy are people in

Sydney today, and how much money will they spend in shops?" "What

worries people in Paris at the moment?" "How many people are up

between 3am and 4am on Sunday nights around Manhattan's Central

Square, and is it worth selling pizza at that time?" "How noisy is it in the

part of town I am considering to move to?" "What's the rate of flu

infections in the region where I wish to spend my holidays?" "Where are

the road holes in my city located?" "When did we have the last significant

iii

see the movie http://www.youtube.com/watch?v=ECJ2DdPhMxI . It turns out that this

technique can be successfully applied even in cases where certain key information (such as

the infectiousness of the disease and recovery rate) is not well-known, which is typical after

the outbreak of a new disease. The only data besides the outbreak location which is

important for our analysis is the volume of passenger air traffic between all airports. This is

needed to specify the effective distance.

D. HELBING: THE AUTOMATION OF SOCIETY

40

earthquake within a range of 500 kilometers?" Answers to questions like

these would help us to be more aware of the world around us, to make

better decisions, and act more effectively. But how will we get all this real-

time information?

The sensor networks, on which the "Internet of Things" is based, will

enable us to perform real-time measurements of almost everything. They

can be used to build a "Planetary Nervous System" (PNS), an intelligent

information platform proposed by the FuturICT project

(http://www.futurict.eu).

66

The In fact, my team has started to develop

such an information platform, called Nervousnet.

67

Nervousnet will harness the

power of the Internet of Things for everyone's benefit and will be built and

managed in a participatory way, as a "Citizen Web".

68

We Similar to

OpenStreetMap, we will develop this system together with an emerging

network of volunteers, who are committed to developing the project

further.

This collaborative approach will give citizens control over their personal

data, in accordance with their right of informational self-determination, and

create new opportunities for everyone. Nervousnet will not only offer the

possibility to contribute to the measurement of our world, in order to

jointly create something like a real-time data Wikipedia. Nervousnet will also

establish a social mining paradigm, where users are given freedom and

incentives to collect, share and use data in ways that do not aim to

undermine privacy. Appendix 3.1 provides further information on the

platform. With your help, it may very well become a cornerstone of the

public information infrastructure of the emergent digital society. So why

don't you join us in building the Nervousnet platform or in measuring the

world around us?

iv

Sociophysics: Revealing the hidden forces governing our society

But it takes more than data to understand the world and its problems.

Measuring, analyzing and visualizing data is just the first step, because data

mining alone may not lead to a good understanding of unstable system

dynamics, which produces most of the good and bad surprises in the world.

In order to help us when we really need it, we must find explanatory

models, which can predict situations that haven't occurred before and

cannot be understood by extrapolation.

In fact, some "social mechanisms" influence human behavior in a similar

way to how the gravitational force determines planetary motion

v

. Just think

iv

You can get in touch with us through nervousnet@ethz.ch

v

even though there is more diversity and randomness to human interaction (which mostly

SOCIAL FORCES

41

of the way social norms determine our roles and behaviors.

69

The scientific

approach of "agent-based simulations" aims to formalize these rules and

turn them into computer codes. Complementary, the research field of

sociophysics tries to express the corresponding interactions and their

outcomes using mathematical formulas.

70

In this chapter, I will particularly

discuss the powerful concept of "social forces", which enables researchers

to understand the link between micro-level interactions of individuals and

the often unexpected macro-level outcomes in socio-economic systems.

The concept of forces is one of the main pillars of physics. In order to

discover it, the old geocentric worldview, in which the Earth was assumed

to be the center of the universe, first had to be replaced by the heliocentric

worldview, which recognized that all the planets in our solar system revolve

around the sun. Later, the new understanding of the planetary system

allowed Isaac Newton (1642-1727) to interpret measurement data about the

planets in a new way, which led him to formulate a simple and plausible

model of planetary motion based on the concept of gravitational forces.

Now, most concepts used in modern physics are formulated in terms of

forces and the way they influence the world. The predictive power of these

models is striking and has been impressively demonstrated by the Apollo

moon shot and every single satellite launch.

Another foundation of the success of physics is the tradition of building

instruments to measure the forces that would otherwise be imperceptible to

our senses. This has enabled physicists to explore a vast array of questions,

spanning from the early stages of our universe to the exploration of

elementary particles and the study of fundamental processes in biological

cells. Therefore, the next logical frontier of science is to build

"Socioscopes" that can reveal the hidden forces behind the dynamics of

socio-economic systems. In this way, we will eventually learn to understand

the counterintuitive behaviors of complex dynamical systems.

71

I believe

that we will soon be able to diagnose emergent societal problems such as

financial crashes, crime, or wars before they happen.

72

This will empower us

to avoid or mitigate these problems similarly to the way medical diagnostic

instruments have helped us to prevent or cure diseases. Isn't that an

exciting prospect?

Social forces between pedestrians

To demonstrate the feasibility of this vision, let me first discuss the

example of pedestrian and crowd dynamics to underline the usefulness of

force models in the social sciences. Starting in 1990, when I wrote my

benefits our economy and society)

D. HELBING: THE AUTOMATION OF SOCIETY

42

diploma thesis,

73

I noticed that pedestrian paths around obstacles looked

similar to the streamlines of fluids. So, I decided to formulate a fluid-

dynamic theory of pedestrian flows. I derived it from a model for the

motion of individual pedestrians, which was inspired by Newton's force

model.

FIGURE 3.2: Illustration of the various "social forces" acting upon a pedestrian.

vi

This "social force model" assumes that the acceleration, deceleration an

directional changes of pedestrians can be approximated by the sum of a

number of different forces, each of which captures a specific desire or

"interaction effect". For example, each pedestrian likes to move with a

certain desired speed into a preferred direction of motion. This can be

represented by a simple "driving force", which captures how the person's

velocity is gradually adapted. Moreover, each pedestrian seeks to avoid

collisions and to respect a certain personal "territory" of others. This is

reflected by a "repulsive interaction force" between pedestrians which

increases with proximity. Repulsive interactions with walls or streets can be

described by similar forces. The attraction of tourist sites and the tendency

for friends and family members to stay together can be represented by

"attractive forces".

74

Finally, a random force may be used to reflect the

individual behavioral variability.

Despite its simplicity, computer simulations of this model match many

empirically observed phenomena surprisingly well. For example, it is

possible to understand the emergence of river-like flow patterns through a

standing crowd of people, the wave-like progression of individuals waiting

in queues, or the lower density of people on a dance floor compared to the

surrounding spectators watching them.

75

vi

I would like to thank Mehdi Moussaid for providing this graphic.

SOCIAL FORCES

43

Self-organization of unidirectional lanes in pedestrian counter-

flows

There are also various self-organization phenomena that lead to

fascinating collective patterns of motion. For example, when people enter a

corridor on two sides, we observe the formation of lanes of unidirectional

flow

76

(see Fig. 3.3). That is, people walking in opposite directions

automatically coordinate each other so that they are hardly obstructed by

the respective counter-flow. This makes transit more efficient for everyone.

While it seems as if the "invisible hand" is at work here, we can actually

explain how social order is created and how collectively desirable outcomes

occur from local interactions: whenever an encounter between two

pedestrians occurs, the repulsive interaction force between them pushes the

pedestrians a bit to the side. These interactions are more frequent between

opposite directions of motion, due to the higher relative velocity. This is the

main reason why people walking in opposing directions tend to separate

into lanes of unidirectional flow. To explain the phenomenon, we don't

need to assume that pedestrians prefer to walk on a certain side of the

street.

77

In conclusion, complexity science can explain the formation of

lanes of pedestrians as a so called "symmetry breaking" phenomenon,

which occurs when a mixture of different directions of motion becomes

unstable.

FIGURE 3.3: Illustration of the formation of lanes of uniform walking direction in

pedestrian counterflows.

78

Walking through a "wall" of people without stopping

Surprisingly, the very same force model also reproduces a number of

other interesting findings in pedestrian crowds, such as oscillatory changes

in the flow direction of pedestrians at bottlenecks. Therefore, when a crowd

of people builds up at a junction, the pressure tends to be relieved

alternately on either side of the bottleneck. Another example of self-

organization is the amazing phenomenon of "stripe formation", which

allows pedestrians to cross another pedestrian stream without having to

stop (see Fig. 3.4). This is almost akin to walking through a wall! Using the

D. HELBING: THE AUTOMATION OF SOCIETY

44

social force model, it's possible to understand how this is possible. The

formation of stripes – which occur for similar reasons as the lanes discussed

before – allows pedestrians to move forward with the stripes and sideways

within stripes that are forming in the intersecting pedestrian flow. In

combination, this enables a continuous collective motion through a

pedestrian stream moving in another direction.

79

FIGURE 3.4: Illustration of the phenomenon of "stripe formation" in two crossing

pedestrian flows.

80

Measuring forces

In physics, forces are experimentally determined by measuring the

trajectories of particles, especially changes in their speed and direction of

motion. It would be natural to do this for pedestrians, too. At the time

when we developed the social force model for pedestrians, I could not

imagine that it would ever be possible to measure social forces

experimentally. But a few years later, we actually managed to do this. In

around 2006, the advent of powerful video camera and processing

technologies put my former PhD student, Anders Johansson, into the

position to detect and analyze the trajectories of pedestrians from filmed

footage. Using this data, we adapted the parameters of the social force

model in such a way that it optimally reproduced the trajectories of the

observed pedestrians.

81

In 2006/07, similar tracking methods became

important for the analysis of dense pedestrian flows and the avoidance of

crowd disasters.

82

Later, in 2008, Mehdi Moussaid and Guy Theraulaz set up a pedestrian

experiment in Toulouse, France, under well-controlled lab conditions.

83

This finally allowed us to perform data-driven modeling. While before, we

had to make assumptions about the functional form of pedestrian

interactions, it then became possible to determine the functional

dependencies directly from the wealth of tracking data generated by the

SOCIAL FORCES

45

pedestrian experiment. After fitting the social force model to individual

pedestrian data, it was finally used to simulate the flows of many

pedestrians. To our excitement, the computer simulations yielded a

surprisingly accurate prediction of the pedestrian flows observed in a wide

pedestrian walkway.

So, pedestrian modeling can be considered a great success of

sociophysics. One can say that, over time, pedestrian studies have turned

from a social to a natural science, bringing theoretical, computational,

experimental and datadriven approaches together. This has even led to

practical and surprising lessons for the design of pedestrian facilities and for

the planning of largescale public events such as the annual pilgrimage in and

around Mecca, as we will discuss now.

Most pedestrian facilities are inefficient

Back in 1994/95, Peter Molnar and I compared a range of different

designs of pedestrian facilities. Surprisingly, we found that obstacles, if

properly placed, can make pedestrian counter-flows more efficient (see Fig.

3.5). In fact, all of the conventional design elements of pedestrian facilities

such as corridors, bottlenecks, and intersections turn out to be ill-designed

and can be considerably improved! In many cases, "less is more" in that

providing less space for pedestrians can produce a better flow. This

surprising discovery can be best understood for bottlenecks such as doors.

Here, a funnel-shaped design can reduce disturbances in the pedestrian

flow, which otherwise result when the directions of motion are not well

enough aligned (e.g. when some people approach the door from the front

and others from the side).

In the case of busy bi-directional pedestrian flows, the efficiency of

motion can be improved by a series of pillars in the middle. These pillars

help to stabilize the interface between the opposite flow directions, thereby

reducing disturbances. The effectiveness of the design becomes particularly

clear in subway tunnels, where pedestrians move both ways and pillars exist

for static reasons.

D. HELBING: THE AUTOMATION OF SOCIETY

46

FIGURE 3.5: Illustration of conventional and improved elements of pedestrian

facilities.

84

Finally, an obstacle in the middle of a pedestrian intersection may also

improve the flow. When Peter Molnar and I discovered this, it took us

some time to understand this unexpected finding. Eventually we noticed

that, at intersections, many different collective patterns of motion can

emerge, for example, clockwise or counter-clockwise rotary flows, or

oscillatory patterns of the crossing flows. The problem is that the different

collective patterns of motion conflict with each other, so that none of them

are stable. Putting a column in the center increases the likelihood of rotary

flows and thereby increases the overall efficiency of pedestrian traffic. The

flow can be improved even more by replacing a four-way intersection with

four bidirectional intersections, which can be achieved by placing railings in

suitable locations. This encourages a rotary flow pattern, which greatly

reduces disturbances.

Crowd disasters

Unfortunately, pedestrian flow doesn't always self-organize in an

efficient way. Sometimes, terrible crowd disasters happen, and hundreds of

people may be injured or killed, even when everyone has peaceful

intentions and does not behave aggressively. How is this possible?

When I got interested in the problem in 1999, crowd disasters were

often regarded as "acts of God" similar to natural disasters that are beyond

human control. However, the root cause of the breakdown of social order

SOCIAL FORCES

47

in crowds is similar to the reason behind "phantom traffic jams". If the

density of people gets too high, the flow of pedestrians becomes unstable.

The resulting crowd dynamics can be uncontrollable for individual people

and even for hundreds of security guards. Nevertheless, crowd disasters can

be avoided, if their causes are well enough understood and proper

preparations are made.

Crowd disasters have happened since at least Roman times. That's why

building codes were developed for stadiums, as exemplified by the

Coliseum in Rome. The Coliseum had 76 numbered entrances and could

accommodate between 50,000 and 73,000 visitors, who would exit through

the same gate through which they had entered. These rules and the

generous provision of exits meant that the Coliseum could be evacuated

within five minutes. Modern stadiums, which generally have a smaller

number of exits, can rarely match this performance.

Despite the frequent and tragic occurrences of crowd disasters in the

past, they continue to happen due to common misperceptions. Media

reports often suggest that crowd disasters occur when a crowd panics,

causing a stampede in which people are crushed or trampled. Therefore,

crowd disasters are claimed to be the result of unreasonable or aggressive

behavior, with some individuals pushing others relentlessly as they try to

escape. But why would people panic? My colleague Keith Still put it like

this: "People don't die because they panic, they panic because they die."

In fact, studies that I conducted with Illes Farkas, Tamas Vicsek, Mehdi

Moussaid, Guy Theraulaz and others revealed that many crowd disasters

have physical rather than psychological causes.

85

They may occur even if

everybody behaves reasonably and tries not to harm anyone else. Therefore,

the common view that crowd disasters are mostly a result of panic is

outdated. I don't negate that people are in a danger to be crushed when the

inflow of people into a spatially constrained area exceeds the outflow for an

extended period of time. Certainly, a high density crowd can become life-

threatening under such conditions, as more and more people accumulate in

too little space. However, most of the time crowd disasters occur for a

different reason.

For example, during the annual Muslim pilgrimage around Mecca in

2006, a crowd disaster occurred on a large plaza. Anders Johansson and I

were asked to evaluate video footage of this accident. Initially, due to the

high density of pilgrims, we could not see anything else than very slow

motion, a few centimeters per second. However, when I asked Anders to

play the videos 10 or even 100 times faster, we made some surprising

discoveries!

86

D. HELBING: THE AUTOMATION OF SOCIETY

48

FIGURE 3.6: Time-lapse photograph of stop-and-go flows in dense pedestrian crowds.

87

The accelerated videos showed some striking phenomena. First, we

observed an unexpected, sudden transition from smooth pedestrian flows

to stop-and-go flows (see the long-term photograph above).

88

In contrast to

highway traffic, however, these stop-and-go waves were previously

unknown and unlikely to result from delayed adaptation. Eventually, we

discovered that these waves were caused by a competition of too many

pedestrians for too few gaps in the crowd, i.e. by a coordination problem.

89

The stop-and-go movement emerged when the overall flow suddenly

dropped to lower values, similar to the capacity drop phenomenon in

vehicle traffic. As a consequence, the outflow from the area drastically

decreased, while the inflow stayed the same. Thus, the density increased

quickly, but while this certainly created a dangerous situation, it was not the

ultimate cause of the tragedy!

FIGURE 3.7: Illustration of the phenomenon of crowd turbulence under

extremely crowded conditions.87

SOCIAL FORCES

49

To our further surprise, some minutes later we witnessed another

unexpected transition – from stop-and-go flows to a phenomenon that we

call "crowd turbulence" (see Fig. 3.7).

90

In this situation, people were

pushed around in random ways. So, Anders Johansson and I discovered

that it was not the density, but the density multiplied by the variability of

speed – the so-called "crowd pressure" – which triggered the disaster at a

certain time and location. Moreover, we found that any body movement –

even unintentional – could exert a force on the bodies of pedestrians in the

immediate neighborhood, when the crowd density exceeded a certain

critical threshold. These forces could add up from one body to the next,

meaning that the resulting force quickly changed in strength and direction.

As a consequence, people were pushed around in unpredictable and

uncontrollable ways.

It was just a matter of time until someone lost balance, stumbled, and

fell to the ground. This produced a "hole" in the crowd, which unbalanced

the forces acting on the surrounding people, because the counter-force

from where the person stood before was missing. Therefore, the

surrounding people tended to fall on top of those that had previously fallen

or they were forced to step on them. The situation ended with many people

piled up on top of each other, suffocating the people on the ground. Similar

observations were made in other crowd incidents, such as the Love Parade

disaster in Duisburg, Germany, for example.

91

Countering crowd disasters

Can we use the above knowledge to avoid crowd disasters in future?

The answer is yes! Some years back, together with several colleagues, I

became involved in a project aiming to improve the flow of pedestrians

during the annual Muslim pilgrimage to Mecca. We were asked to find a

better way of organizing the crowd movements around the Jamarat Bridge,

a focal point of the pilgrimage, where thousands of pilgrims had died in the

past due to a number of tragic crowd disasters. How could we avoid them?

This was a challenge that was not simply related to technical matters

such as crowd densities. We also needed to take dozens of religious,

political, historical, cultural, financial and ethical factors into account. Our

previous experience of modeling crowds led us to propose a range of

measures including the counting of crowds through a newly developed

video analysis tool, the implementation of time schedules for groups of

pilgrims, re-routing strategies for crowded situations, contingency plans for

possible incidents, an awareness program to inform pilgrims in advance

about the procedures during the Hajj, and an improved information system

to guide millions of pilgrims who spoke 200 different languages.

92

After

implementing many of our proposals, the Hajj (in 1427H) was indeed

D. HELBING: THE AUTOMATION OF SOCIETY

50

performed safely without any incident in 2007.

vii

The principle which

underpinned the success of this approach was to gather real-time

information and suitably adapt to it.

Since then, the principle of providing real-time feedback to crowds has

become a trend. An interesting example is an app to improve the safety of

mass events, which was developed by Paul Lukowicz, a member of the

FuturICT project, and a number of other scientists such as Ulf Blanke.

93

By

using this app, festivalgoers at a number of festivals in London, Vienna and

Zurich voluntarily provided GPS data about their locations, which was used

to determine their speeds and directions. This data was then returned to the

festivalgoers to give them a picture of the areas which they should better

avoid due to over-crowding.

Forces describing opinion formation and other behaviors

Is the usefulness of the concept of social forces restricted to pedestrian

flows (and traffic flows

94

), or can it also be applied to various other kinds of

social phenomena such as crime and conflict as well? The success of force

models in describing pedestrian flows is related to the fact that pedestrians

are moving continuously in space. Therefore, the dynamics of a pedestrian

can be represented by an equation of motion, which states that the change

of his/her spatial position over time is given by their velocity. An additional

equation expresses that the change in velocity over time (i.e. the

acceleration) can be modeled by a sum of forces. But can we also

understand how people form opinions or other behavioral changes based

on social forces? Surprisingly, the answer is "yes", if the changes in opinion

are more or less gradual on a continuous opinion scale or in a continuous

opinion space.

95

Otherwise, generalized models would have to be used,

which exist as well.

96

After formulating the social force model for pedestrians in Göttingen,

Germany, in 1990, I joined Professor Wolfgang Weidlich's team at the

University of Stuttgart. He was probably the only physicists working on

socio-economic modeling at that time. So, Professor Weidlich might be

seen as grandfather of sociophysics.

97

When I joined his team, my plan at

this time was to learn how to model opinion formation and decision-

making. Since my work on pedestrians, I had the idea that both individual

vii

In the meantime, three further levels of the new Jamarat Bridge were completed, and the

organization of the pilgrim flows was changed several times. However, since 2007 I haven't

been involved in this anymore. To the best of my knowledge, the Hajj was safe as long as the

recommendations of the international expert panel were approximately implemented. It also

seems that the international expert panel, which is now in charge, has not been responsible

for the sad crowd disaster in 2015.

SOCIAL FORCES

51

and collective human behavior could be understood as a result of social

forces, and I formulated a corresponding theory (see Appendix 3.1).

Interestingly, it is possible to develop social force models for migration,

too, if one assumes that people relocate within a certain (not too large)

geographic range. A model that I formulated in 2009 examines "success-

driven migration".

98

According to this, individuals try to avoid locations

where they expect bad outcomes and are attracted to locations that appear

to be favorable. Bad neighborhoods (in which people were uncooperative)

were found to have a repulsive effect, whereas good neighborhoods (where

people were cooperative) attracted migrants. It is even possible to calculate

the direction and strength of this repulsion and attraction effects, i.e. the

forces which imply the average direction and speed of motion in a certain

location.

In general, a great advantage of using the concept of "social forces" is

that it can help us to develop a better idea of the complex processes

underlying social change. Movements towards a subject or object are

reflected by attractive forces, while movements away from a subject or

object are reflected by repulsive forces. It is also important to recognize that

such forces may not be attributable to individuals, but rather to groups of

individuals, companies or institutions. In other words, social forces may be

a collective effect. Group dynamics or "group think", as a result of the

emergence of a particular group identity is probably a good example of this.

individuals creates a collective "group" perspective, which in turn changes

the opinion and behavior of individuals. In fact, the theory of social milieus

posits that the behavior of individuals is largely influenced by their social

environment. Very soon, it will be possible to quantify the underlying forces

and to derive mathematical formulas for them. But what is more powerful,

physical or social forces?

Culture: More persistent than steel

It has often been claimed that civilizations were born out of war, and

that the world is ultimately ruled by those with the greatest military power.

However, I don't buy this. Even though war certainly played a role in

establishing the modern world, I believe the main mechanisms underlying

the spreading of civilization are migration and the exchange of goods and

ideas. Today, the Internet can certainly advance civilizations in ways that

don't need to be paid by human lives.

But what is the basis of civilizations? It's culture, and to a large extent,

culture is the result of numerous sets of rules, such as social conventions,

values, norms, roles, and routines.

99

These rules determine the success or

failure of societies and guide their evolution. Just take religious values for

instance, which can determine the behavior of millions of people for

thousands of years. It is not an exaggeration, therefore, to say that culture is

D. HELBING: THE AUTOMATION OF SOCIETY

52

more persistent than steel

viii

and probably also more relevant to the success

of civilizations than weapons.

ix

In other words, social forces can be stronger

than physical forces. A good example of this is ancient Greek culture,

which spread to the Roman occupants because it was more advanced.

Reducing conflict

While we all learned about physical forces at school, very few people

have an explicit understanding of the social forces, which determine the

behavior of socio-economic systems. This has to change, if we want to

overcome or at least mitigate the problems we are faced with. Conflicts,

wars and revolutions can be understood as a result of, too. Certain forces

can destabilize systems and cause them to disintegrate. There are at least

three types of conflict situations: (1) An encounter (say, between two

countries) causes losses on both sides. This might be avoided by increasing

the awareness of the likely outcomes of such an encounter in advance. (2)

The encounter is beneficial to one party but unfavorable to the other, and

causes damage overall. Here, the second party needs to be protected from

exploitation (e.g. through solidarity from third parties, or by separating the

disputing parties). (3) The encounter is advantageous for one side and

undesirable for the other, but the overall outcome is positive. In such

situations, the benefits can be redistributed to make the interaction

beneficial for both sides, i.e. it's possible to align interests to create a win-

win situation.

x

Would it also be possible to actually measure the forces creating

conflict? Yes, I think one can build a ConflictMap, which illustrates regional

and international tensions and explains how they come about. In fact, when

working in my team, Thomas Chadefaux mined millions of news articles

over a period of more than 100 years and performed a sentiment analysis

for words indicating conflict. This allowed him to quantify the level of

tension between countries in the world. Moreover, he could show that the

level of tension could be used to predict the likelihood of outbreaks of war

within a 6-12 months time period.

100

Such advance warning signals might

give politicians enough time to engage in diplomacy to peacefully resolve

the tensions before it's too late. Our analyses also revealed how tension

spreads from one country to the next, destabilizing a huge region, as it

viii

It is surprising enough that one can cut steel with light (i.e. a mighty laser beam), but the

persistence of culture is even more astonishing.

ix

even though protection from aggression and exploitation is needed, of course.

x

I have recently proposed "Social (Information) Technologies" that aim to reduce the

occurrence of conflict between people and companies.

SOCIAL FORCES

53

happened after the war in Iraq. This might also have produced fertile

ground for the rise of the Islamic State (IS).

What we can learn from Jerusalem

Another data-driven study analyzed a problem that has worried the

world for many decades: the conflict in the Middle East. Why haven't we

been able to end this conflict yet? A classical Big Data approach, even if we

knew the trajectories of all the bullets shot, couldn't really reveal the causes

of this conflict. Nevertheless, it is possible to understand the roots of the

conflict. A few years ago, I got involved in a study with Ravi Bhavnani, Dan

Miodownik, Maayan Mor, and Karsten Donnay, which lead to an

empirically grounded, agent-based model.

101

Our model suggests that

intercultural distance is the main driver of the conflict.

A further analysis reveals that violent events are correlated with each

other.

102

So, there is a responsive dynamic, whereby each side retaliates for

previous attacks by the other side.

103

For example, Palestinians retaliate to

Israeli violence and vice versa. What does this tell us? Basically, both sides

punish each other for violence that they suffered before. It seems that each

party tries to send the message: "Stop being violent to us or you will have to

pay a high price!" From the point of view of rational choice theory, this

should stop the chain of violence. As one event triggers another, usually

bigger one, or even several, the conflict becomes increasingly costly for

both sides over time. However, rather than creating peace, a deadly spiral of

violence sets in. An Israeli documentary film entitled "The Gatekeepers",

104

which interviewed previous chiefs of the secret service, came to a

remarkable conclusion: "We have won every battle, but we are losing the

war." In other words, it doesn't pay to be violent, quite the opposite!

So why does such retaliation cause an escalation rather than a calming of

hostilities? This occurs because both sides think their actions are right. In

fact, they are applying sanctioning mechanisms that are intended to create

social order, but these mechanisms are only suited for a mono-cultural

context.

Punishment doesn't always work

To understand the problem better, we must ask the question: "Why do

we punish others?" This has a simple reason: we have learned that

punishment can establish and stabilize social norms. Therefore, who doesn't

follow our norms is usually sanctioned. But such punishment is only

effective, if it is accepted by the punished party. Otherwise this party will

strike back and inflict revenge, which escalates the conflict. Therefore, it is

important to recognize that punishment is only effective if people share the

same values, norms and culture.

D. HELBING: THE AUTOMATION OF SOCIETY

54

In multi-cultural settings, punishment is often not suited to create social

order. Under such circumstances, however, it might be possible to reduce

the level of conflict by physically separating the opposing parties so that

they live in different areas.101 Another option is to develop a culture of

tolerance, understanding and respect. In fact, as we will see later, there are

many social mechanisms that foster social order, such as reputation

systems, for example. I am confident, therefore, that a deeper

understanding of the mechanisms and forces producing conflicts will

eventually help us to overcome or mitigate them. In a multi-cultural world,

I would strongly recommend to move away from a punitive culture.

Instead, it seems more promising to engage with each other in a

differentiated, reputation-based culture in which diversity is welcomed.

xi

This brings us to another important set of invisible factors, which

determines the success of societies, namely "social capital".

Why "social capital" is so terribly important

Most of us have probably heard the proverb that "money makes the

world go round", but there are other, intangible factors that matter. For

example, human capital (like education) can boost individual careers, and

social capital can act like a catalyst of socio-economic success. But what is

social capital? I define it as everything that results from interactions within a

social network which could potentially (be used to) create benefits.

Examples include cooperativeness, public safety, a culture of punctuality,

reputation, trust, power, and respect. However, while our own actions

influence our social capital, we can't fully control it. This is in contrast to

money. In many cases, we can't buy social capital (or only to a limited

extent), but social capital creates added value.

Moreover, we cannot automatically generate a certain amount of social

capital by doing certain things. Similarly to reputation and respect, these

things are given to us by others. They depend on the effects of social

interactions. Note that the amount of social capital within a system also

determines how resilient the system is, and how likely it is to fail. Social

capital influences both the probability and extent of damage. This became

clear to me at a seminar at ETH Zurich's Risk Center,

xii

when we discussed

the disproportional effect of large disasters on public opinion. Plane crashes

and terror attacks, for example, matter a lot to people, while they seem to

feel less threatened by everyday risks such as car accidents or fatalities

xi

We will discuss this further in Chapter 8: How Society Works

xii

The Risk Center brings together experts in probability theory and experts in complexity

and network theory.

SOCIAL FORCES

55

caused by smoking. Therefore, it is often believed that "size matters", in the

sense that large disasters make people respond irrationally or even in panic.

However, having studied the phenomenon of panic for some time, I

came to a different conclusion. People realize that the damage is not just

physical in nature. Social capital can be damaged, too. For example, a large-

scale disaster often reduces public trust in the risk management of

companies or public authorities, particularly if it was caused by

unprofessional conduct or corruption. While people care about such things,

no insurance company covers damage to social capital.

Hence, we must protect social capital similarly to how we protect

economic capital or our environment. Social capital can be destroyed or

exploited, but this should be prevented. In order to do this, we must learn

to measure social capital and to quantify its value. Quantifying the value of

our environment also helped to protect it.

Trust and power

To stress the importance of social capital, it is important to acknowledge

that the financial crisis resulted from a loss of trust. Banks did not trust

other banks anymore and did not want to lend out money; customers did

not trust their banks anymore and emptied their bank accounts; banks did

not want to give loans to companies anymore; people did not want to invest

in financial derivatives anymore – the list goes on. In the end, the resulting

financial meltdown cost an estimated $15 trillion at least.

105

So, trust is

highly valuable and when it erodes, the economic losses are tremendous. To

give another example, the recent loss of trust in US cloud storage

companies due to the revelations about mass surveillance by the National

Security Agency (NSA) has substantially reduced their business volume.

106

Trust is also the basis of power and legitimacy. When I studied in

Göttingen, Germany, a deadly car accident caused by a mistake of the

police triggered a large public outcry and massive demonstrations. This was

the first time when I noticed that public institutions can easily lose public

support. In other words, they can easily lose social capital. Trust is eroded

whenever authorities do something that is contrary to the moral beliefs of

the public. I made a similar observation in Zurich, Switzerland, when many

people complained about the policies of the migration office. During this

time, the windows of the migration office were repeatedly smashed in, but

when the director was replaced, the problem disappeared.

D. HELBING: THE AUTOMATION OF SOCIETY

56

FIGURE 3.8: Illustration of how power results from trust and transparency.

It is very interesting that "soft" factors such as credibility and trust are

the basis of power! For example, the London Riots of 2011 occurred after a

person was shot by the police, but no sufficient justification was given to

the public. Something similar happened in 2014 in the city of Ferguson,

Missouri, and 2015 in the city of Baltimore, Maryland. In fact, riots in many

other countries were triggered by events where public authorities lost

legitimacy in the eyes of the public. The Arab spring, for example, started in

Tunisia, after Mohamed Bouazizi burnt himself to death to protest against

police corruption and brutality.

In other words, both legitimacy and power are contingent on the

authorities doing what the people regard to be "the right thing". If people

withdraw their support, authority and power vanish. While the state can

purchase weapons and, with this, acquire destructive power, constructive

power depends on the trust and support of the people. While weapons

might create fear, this is not a good substitute for genuine legitimacy. As the

situation becomes increasingly unacceptable, more and more people will

lose their fear, and start to resist the previously respected authorities. Some

may even be willing to sacrifice their own lives. Remember that many

extremists and terrorists previously led normal family lives. But even the

passiveness of citizens can make a country fail within just a few years. This

could be observed, for example, in the former German Democratic

Republic, which had a horrible surveillance system. For these reasons, I am

convinced that trust is the only sustainable basis of power and social

order.

107

SOCIAL FORCES

57

APPENDIX 3.1: Nervousnet: A decentralized digital nervous

system

The open standards of the World Wide Web (WWW) have unleashed a

digital economy worth many billion dollars, and participatory projects such

as Linux

The Automation of Society is Next: How to Survive the Digital Revolution

Abstract and Figures