Beyond the Hole in the Wall

Discover the Power of Self-Organized Learning

By Sugata Mitra
Foreword by Nicholas Negroponte
This book is for Shantibrata Biswas, who made children laugh.
 Table of Contents

Foreword
Introduction
2062: Child’s play
Unequal education
The Hole in the Wall
A laboratory for the world
2062: A boring past
A new idea: Minimally Invasive Education
What did we find?
What will they learn?
Porn patrol
How Minimally Invasive Education works
2062: The cube in the tank
Beyond the Hole in the Wall
Rise of the SOLE
2062: Learning and babble
Learning as emergent phenomena
Connectedness
Conscious systems
2062: Emergence
A step-by-step guide to Method ELSE
Tips from the field
About the author
About TED
 Foreword

Learning without school

By Nicholas Negroponte

Learning and teaching are not symmetrical. They are not the flip sides of the
same coin, in spite of the fact that almost all papers and conversations on
education assume they are.

The working assumption is this: Solve teaching and you will get learning.
Even if true in part, it addresses only some kinds of learning and never really
attempts to understand the learning of learning itself. And, of course, it does
not apply to the 70 million children worldwide who do not even go to first
grade.

I am constantly astonished and disappointed at the assumptions people make

about education: how to achieve it, how to measure it, how to accelerate it
where it exists and how to create it where it does not. The default model that
most people use is called “school,” and that includes yet further assumptions.
The most universal and blatantly false of them is that age segregation is a
good idea. The idea that all 7-year-olds should study together, and then the
next year do the same with 8-year-olds, is just a bad idea. It was designed
only for the convenience of the institution. It is like public toilet stalls not
going from floor to ceiling, for the convenience of the janitors.

Enter Sugata Mitra. He offers a very different view of learning, one which
involves the collective learning of mixed ages, achieved without a teacher.
He has shown that a critical mass of illiterate 10- to 12-year-olds can conduct
exercises of a level of difficulty that would otherwise require an eighth- to
10th-grade education. He lets these children teach each other, self-organize
and explore in a manner more akin to the organization of a sports team than
of a classroom.

There is a great deal to learn from Sugata’s 12 years of experience with the
Hole in the Wall project. To me, the most notable idea is that children are far
better at organizing themselves than we assume. Much of the world is
discoverable, which is how we all learned from the time we were born until
around age 5, when our more formal education began. We interacted with our
environments to acquire language and common sense. We acquired so much
knowledge during those years that we learned many things about
manipulating the world and even some about manipulating our parents.
Suddenly, at age 5, our learning was assumed to be different and was
delivered to us, almost solely through being told by people and soon after by
books.

Sugata’s point is that a great deal of what we call education is now

discoverable. If children can learn to read on their own (the topic of his next
book), they can learn almost everything else on their own, too. Keep in mind,
he does not mean one child by himself or herself; he means by a collective
process. That can include adults, certainly, but in many places on this earth
there are no adults with sufficient literacy to help.

Ask yourself what education means. We measure children on what they

know. By and large, they have to memorize useless content to meet that test.
Because measuring the results of rote learning is easy, rote prevails; and in
developing nations, rote learning is worse than in the United States (believe it
or not) and yet worse than in some nations we admire in Asia. What kids
know is just not important in comparison with whether they can think.

Learning math and spelling is far less important than learning the act of
learning. That is what Sugata’s children are doing, and that is what this book
is about.

Massachusetts Institute of Technology

Cambridge, Massachusetts, USA
 Introduction

When wandering down the street, it’s hard to tell, at a glance, whether a
person is speaking on a cellphone through a hands-free headset tucked
discreetly into her ear or simply talking to herself.
At the supermarket, you can’t tell if the cashier knows arithmetic or not.
Your groceries are scanned electronically and the prices tallied automatically.
Yet the cashier probably still performs his work capably. Arithmetic is an
outdated life skill, like swordplay or horse riding. Four hundred years ago,
those were vital life skills; today they are relics of a bygone world and
primarily enjoyed as sports.
Although right now you can tell when a person is online because the
smartphones, computers, and other devices that connect us to the Internet are
big enough to be seen, this too will soon change as the electronic tools that
immerse us in the online world become smaller and more seamlessly
integrated into our products and our lives.
In the same way that it’s getting harder to tell whether a person is talking
on a cellphone, capable of doing basic arithmetic, or surfing the Internet, we
are entering a world in which it will be difficult, maybe impossible, to
determine whether a person is educated or not. Will this matter? Is formal
education, as we know it, an outdated idea? That last question is the central
focus of this book.
This is a story about how different our not-too-distant future may be.
Suspend your disbelief, at least momentarily, and join me on this journey into
the next half century. Remember, 50 years ago there were no computers, no
cellphones, no Internet, no Google, no Facebook and no Prozac. And we have
every reason to expect that 50 years from now there will have been equally
momentous and transformative changes from the way we live today.
A generation of children who are 16 years old or younger have never
known a world without many of the connecting technologies that we have
come to take for granted and rely on heavily. How do these devices affect,
and even improve, how we absorb information?
 This book is about children and learning in a highly connected world. It
was launched through a series of experiments known collectively as the Hole
in the Wall project. The experiments spanned more than a decade and were
spread across five continents, and they revealed a great deal about how
learning happens and how consciousness develops.
I finished this book on a rainy August 15th, India’s Independence Day. But
there is another independence day I would like to talk about here: It is the day
we achieve independence from an education system that is more than 2,500
years old. It’s time to begin that journey.

Sugata Mitra
Cambridge, Massachusetts
 2062: Child’s play

Rita woke up with a start. The volume was too high on the Bluetooth Stereo
Implants (BSIs) in her ears. The implants were new, a gift from her parents
for her ninth birthday, which was just a week ago. Her bio-mom had been
worried the BSIs might damage Rita’s hearing if they were set wrong, but her
father’s partner Joan was a doctor, and Joan convinced Rita’s bio-mom that
BSIs were safe for children. “They don’t have anything to do with the
eardrums,” Joan explained. “BSIs connect directly to the cochlear nerves.”
It took just 15 minutes to get the implants fixed, and it didn’t hurt. Really.
Rita was in love with her BSIs. Every sound was clearer, even the slightest
whisper. Best of all, she liked how her BSIs could read stories to her from the
Internet so softly that no one could tell she was listening.
On the third day after the implants were inserted, Rita learned to do Hi-
Google searches, which were sophisticated ways of doing very deep research
online. Now she could answer almost any question at all. In their groups in
school, Maam had asked them to learn about electricity. One group
announced an early finding: An electric current was a flow of electrons.
“What’s an electron?” asked Maam. The children worked together for a
while in a Self-Organized Learning Environment (SOLE) to find out. “My
goodness, I didn’t know that at all!” Maam exclaimed when the children
reported their findings. The group laughed.
During playtime, Rita’s friend Dev told Rita that electrons could be in two
places at the same time. He said he had asked his BSIs where electrons were,
and the answer came back with information about “uncertainty” and
something called “clouds of probability.” Nearby, other children listened.
They didn’t really understand what the words “probability” and
“uncertainty” meant, but after lunch they told Maam what they had heard.
She looked very surprised and impressed. Sometimes they wondered if Maam
was a real person projecting her image from somewhere on the Granny
Cloud or if she was just a computer program. Rita wanted to ask her, but Dev
said, ‘Don’t ask Maam anything; she will only tell us to find out for
ourselves!’ In the end they decided they didn’t really care if Maam was real
or not. She was funny, sometimes silly, sometimes stern but not really, you
know, a bit stupid when she said things like, ‘I could never have figured that
out by myself!’ They loved their Maam; it was fun growing up with her.
That night, Rita asked her BSIs to tell her everything about electrons, and
she visited websites with easy-to-understand explanations of electrons, clouds
of probability and the uncertainty principle. She even learned about quantum
entanglement, the theory that describes how electrons can be in two places at
the same time. It all made her very, very sleepy though.
The next morning, Rita just barely heard her bio-mom calling out from
downstairs that Rita would be late for school if she didn’t get up “Now!”
Rita hustled out of bed and listened to the rest of the discussion about
quantum entanglement on her BSIs while getting ready for school.
She had to run most of the way to school and got there just in time for the
groups’ weekly test. It was a two-hour history test on the European economic
crisis of 2010-2011. Rita didn’t really care about things that had happened
50 years ago, but she liked the idea of working on the test with her friends.
They formed groups of three, four or five children and got down to work.
Most of the children in Rita’s group had BSIs, but Lee had a state-of-the-art
V-screen implant that could show images in his visual cortex. Also, Scott had
brought in an ancient tablet computer that was as big and thick as a sheet of
paper! The group shared everything as they listened to what their BSIs were
saying and looked at Lee’s images projected on Scott’s tablet.
All the groups created websites on their findings and presented them to the
class. Overall, Maam was pleased with the groups’ explanations about why
the European countries had gone into debt and fallen into recession as well
as with the descriptions of the trouble that ensued. Most of the groups
received high marks on the test, but not Rita’s group. They had gotten
distracted with Scott’s old-fashioned tablet and had slacked off on their work.
Maam instructed them to read the other groups’ websites and to summarize
the findings into a single website of their own.
It was a great day — mostly. Later that afternoon, Rita had to stand alone
outside their group space because she was caught hiding under the tables to
talk to her friend in Chile on her BSIs. Maam would never have been able to
tell Rita was talking on the BSIs if it hadn’t been for Dev, who told on Rita.
She and her friends decided to completely ignore Dev for the rest of the day.
Tattletale.
Unequal education

In 2006, with the help of Google Earth, a few colleagues from NIIT
University’s Centre for Research in Cognitive Systems and I worked out a
route from New Delhi to the heart of rural northeastern India that avoided
major urban areas. Ritu Dangwal and Leher Thadani, two other colleagues,
then rented a car and drove the route. Whenever they encountered a primary
school on their journey, they stopped and administered a set of tests to the
children in order to ascertain the students’ English, math and science
proficiency. They also sought out the students’ teachers and asked them just
one question: “Would you like to work somewhere else?” For the first 50
kilometers of the journey, most of the teachers said “No.” Then something
changed. When the researchers asked the same question of teachers who
worked more than 50 kilometers (31 miles) from New Delhi, they heard a
new answer — an emphatic “Yes!”
When Ritu and Leher brought the data back, we totaled the marks for each
school and plotted this information against the school’s distance from the
greater metropolis of New Delhi. The graph (see Figure 1) showed that the
farther the school was from the urban center, the lower the grades of the
students.
 Figure 1. Northern India: remote areas, lower performance

Remembering the waning long-term commitment of the teachers who were

farther away from New Delhi, we attributed this downward trend in test
performance to the low enthusiasm and general lack of quality among
teachers in remote areas. There are, and always will be, places in the world
where good teachers either don’t want to go or are unable to go, and we
concluded that the quality of education in these areas suffers as a result.
We published the results in the Australasian Journal of Educational
Technology (ascilite.org.au/ajet/ajet24/mitra.html), noting the significant
negative correlation between the quality of education offered at a particular
school and the distance of the school from the nearest urban center.
Remarkably, teacher training does not help. A trained teacher only finds it
easier to migrate away from remote areas to urban centers. When teachers
want to migrate, the better ones succeed, leaving the worse ones behind.
Remote areas continue to offer poorer quality of education.
We stepped back and began to think about the larger issue at hand: How
will learners in these areas ever get equal learning opportunities? We
wondered if more-developed countries had similar problems with education
in remote areas. Probably not, we figured, because the quality of life in most
geographically remote areas of the developed world is comfortable enough to
attract good teachers, and sometimes strongly so. Does this mean that all
schools perform the same across all areas of the developed world? The
answer is clearly no. Here, too, there are areas where good teachers will not
go. Such areas in the developed world are not necessarily remote in a
geographic sense; they may be remote in other ways. There are areas in big
cities that are socio-economically remote and areas that are religiously remote
or ethnically remote.
We decided to expand the scope of our research. My colleague Barrie
Craven gathered data on school performance in the northeastern United
Kingdom and plotted this against the density of subsidized council housing
(sometimes referred to as “public housing” and typically inhabited by the
poorest citizens in the area) in those locations. He compared how U.K.
schools performed on the General Certificate of Secondary Education
(GCSE) exam — the main test taken to complete the first stage of secondary,
or high school, education — with the density of public housing in the areas
where the schools are located. There was a clear decrease in standardized test
performance in the areas with higher subsidized housing. Here, in the heart of
the developed world, was the same problem of inequality of educational
opportunity that we had found in the remote areas of northern India.
The Hole in the Wall

Can technology play a role in improving the quality of education? We all

have heard the phrase “educational technology,” but what is it really? Are
LCD projectors, laptops and PowerPoint slides education technology? They
weren’t designed as such. Laptops were created for rich company executives,
Microsoft wrote PowerPoint for corporate presentations, and LCD projectors
were invented for corporate boardrooms. Teachers were convinced to adopt
this technology at atrocious prices by salespeople from the corporate world
who had discovered a new market in schools. Initially, the salespeople
focused on selling this technology to the richest schools in the world. But the
richest schools in the biggest cities in the most economically privileged
countries already had good teachers and, for the most part, good students, and
they found the new corporate technology to be overhyped and
underperforming.
Can educational technology be done in a different way? As I was thinking
about this scenario, I wanted to develop, modify and take technology to some
of the most remote locations I could find in order to see if it would survive
and, if it did survive, how it would affect educational outcomes in those
locations.
Backing up, a decade earlier, in 1988, I had written a paper for the annual
conference of the All India Association for Educational Research. It proposed
a simple idea: Give PCs to small children and leave them alone to learn the
technology for themselves. No teachers. No lessons. Just the tech. Surely,
some smart kids will teach themselves to use it. The delegates to the
convention thought my idea was preposterous. Ten years later, however, my
colleagues and I were going to take the idea to the community.
On January 26, 1999, my friend Vivek sunk a computer into the opening of
a wall near our office in Kalkaji, New Delhi. The screen was visible from the
street, and the PC was available to anyone who passed by. The computer had
online access and a number of programs that could be used, but no
instructions were given for its use.
What happened next astonished us. Children came running out of the
nearest slum and glued themselves to the computer. They couldn’t get
enough. They began to click and explore. They began to learn how to use this
strange thing. A few hours later, a visibly surprised Vivek said the children
were actually surfing the Web.

Kalkaji, 1999: the original Hole in the Wall

We were delighted, but we also wondered how this magical computer

literacy had happened. After the heady rush of the initial success of the
Kalkaji experiment, it was time for some thoughtful scientific introspection.
We left the PC where it was, available to everyone on the street, and within
six months the children of the neighborhood had learned all the mouse
operations, could open and close programs, and were going online to
download games, music and videos. We asked them how they had learned all
of these sophisticated maneuvers, and each time they told us they had taught
themselves.
Interestingly, they described the computer in their own terms, often
coining words to indicate what they saw on the screen. For instance, the
children’s word for the hourglass symbol that appears when a program is
“thinking” was “damru,” the name of a small wooden drum shaped like an
hourglass that is a symbol of the Hindu god Shiva. The mouse cursor was
called “sui,” a Hindi word for needle, or “teer,” which means arrow.
We repeated the experiment in two other locations: in the city of Shivpuri
in Madhya Pradesh (Digvijay Singh, a prominent politician, was interested in
our research), and in a village called Madantusi in Uttar Pradesh. Both of
these experiments showed the same result as the Kalkaji experiment: The
children seemed to learn to use the computer without any assistance.
Language did not matter, and neither did education.
Had we stumbled upon something universal about children and computers?
I desperately needed to find research funding.
We were in luck. The social initiatives department at the ICICI Bank (the
second-largest financial services company in India) wanted to put 10
computers in villages in the Sindhudurg area of Maharashtra. A young banker
named Bikram Duggal was the driving force behind the idea, and he has my
gratitude and the gratitude of thousands of rural children.
Wheels were in motion. Sheila Dixit, the chief minister of Delhi, visited
Kalkaji. The principal secretary of Delhi at the time, Mr. S. Regunathan, a
longtime educationist, took what he felt was a calculated risk and funded 30
computers in the Ambedkar Nagar area of Delhi, a decision that has
benefitted more than 10,000 children in that area. Finally, James Wolfensohn,
president of the World Bank, came to Kalkaji. He had heard of the Hole in
the Wall project from Peter Woicke, director of the International Finance
Corp. Together, they decided to fund a three-year research project to find out
if what had happened at Kalkaji was a freak accident or a universal learning
mechanism. That decision has affected the futures of a million children so far,
and it might yet come to affect the lives of all the children on the planet.
A laboratory for the world

India is a great laboratory for determining if any child in the world will
respond to a Hole in the Wall computer in the same way as the Kalkaji
children. India has a great diversity of children belonging to every economic,
social, genetic and cultural category. India is also a good place to test the
survival of outdoor computers that are built into walls and buildings. We
have all the weather conditions experienced by the rest of the world.
The research design was simple. We would install computers in 22
locations across India that had been chosen for their diverse populations and
climatic conditions. We would then choose 15 random children in each of the
locations as a focus group and measure their progress over nine months. We
would compare them with other children who had no exposure to computers.
Our ultimate goal was to gain some insight into what we called the “Hole in
the Wall effect.”
First, we needed some tools for gathering data. How does one measure
computer literacy? We found tests on the Internet that took hours to
administer — much too long for our purpose. As the funding clock was
ticking, we had to find a solution fast. Desperation can often do wonders in
science; in this case, it produced an idea.
Two colleagues from our center, Parimala (Pari) and Ritu, proposed a test
in which a subject is shown the 77 common icons in a Windows environment
and is then asked to describe what they mean. It seemed like a simple but, to
be honest, stupid test. We all use computers quite well, but do we know all
the icons? Of course not. Some of us don’t use the icons at all.
Pari and Ritu administered the test to people in our own office. The results
were interesting. Our employees correctly identified around 70 percent of the
icons, whether or not they used them. Not surprisingly, those who worked
with a lot of spreadsheets knew the Microsoft Excel icons, and those who did
a lot of word processing knew the Microsoft Word icons. It seemed as though
familiarity with an application would give you the ability to guess the icons
of that application correctly, even if you had never used them.
Monica, another colleague of mine, and I did a validation of Pari and
Ritu’s Icon Association Inventory, the official name for the icon test. We
gave two sample groups the icon test and also administered a much longer
American test of computing literacy. The results matched to an incredible 98
percent correlation. The icon test would measure in 20 minutes what the other
test measured in two hours. We had found our measuring instrument.
Ritu added a battery of intelligence and creativity tests to the icon test.
While we were sharpening our measuring instruments, Ravi Bisht, the creator
of India’s first shopping catalog applications, was traveling from village to
village all across India. Wherever he found an appropriate place and a
friendly “panchayat” (local government) or school, he built a little structure
with three computers facing the road — a design that became the standard
design for the Hole in the Wall experimental sites.
Sanjay Gupta, now head of our center, had another set of problems to
solve. He was concerned that the computer touchpads would fail within
weeks of being installed outdoors, that the key tops would vanish, that the
power conditioning would cost more than the computers and that the dust
would get into everything.
Between 1999 and 2002, Sanjay and Ravi made a series of inventions that
solved all these problems. They invented new solid-state, touch-sensitive
mice, and covers for keyboards; they reversed exhaust fans and added a
thousand other vital little things that made it possible for ordinary PCs to
work outdoors — anywhere.
 View of a kiosk showing the child-friendly design

Another concern was that if a computer froze up in a remote village in

Kanyakumari, someone would have to fly from Delhi just to press the reset
button. To solve this problem, we wrote software that would enable us to
“see” our computers from anywhere using the Internet. We also wrote
software to prevent Windows from freezing and to protect our desktops from
getting accidentally deleted.

Now we were ready.

 2062: A boring past

Rita’s father is a banker, so he is home all the time. He’s a fun guy who has
BSIs, a 3-D V-screen implant and a really expensive nanotube outfit that
changes color, shape, size and texture and makes it look as though he
changes clothes frequently. When Rita told him about her history test, he said
he was a boy when the European crisis happened. He told her that, back
then, banks were housed in big buildings so everyone would know how much
money they had, and bankers drove to their offices in cars. Her father’s
stories sounded wild and unbelievable. A bank was a real, physical place
until about 10 years ago, he said, when they all dematerialized. Soon after
that, real money was discontinued. Rita had seen real money on a 3-D
museum website, and she knew that there used to be metal coins that made a
noise when they fell.
In school, Rita and her friends sometimes work in SOLEs to learn more
about the strange world in which their grandparents lived. Granny knows all
about it, of course, but she could be making up some of her stories. She is 120
years old, after all. Her hair is beginning to go gray, and she’s losing her
teeth! So, the children double-check everything Granny tells them.
When their grandparents were young, the world was full of cars. Soon it
became faster to walk than to drive. In the 2050s, the Indian government
announced tax cuts for industries that dematerialized, and people stopped
buying cars altogether. After all, you only traveled far from home for
holidays, and on those occasions you could take a Maglev, which drives
itself. But even in Rita’s time there are still some people who fly in planes
and who think having cars is cool. Crazy throwbacks.
An eMediator from the Granny Cloud once told Rita that, when she was
young, people used to have lots of clothes made from trees and plastic. They
didn’t know about nanotubes, nanofibers, nanomachines or FABs. Children
from that era also had lots of toys that they didn’t use, and all of this stuff
was kept in huge houses. Rita felt this was a bit far-fetched and quite boring.
She told her microFAB to make her a ticking clock that makes faces when
you look at it. She laughed for a while and then let the FAB turn the clock
into a reusable block again.
The eMediator told Rita that, in the past, no one knew that one day a
person would need only one set of clothing, the Internet, a microFAB and 15
square meters to have everything he or she could possibly want. But by then
Rita was fast asleep.
A new idea: Minimally Invasive Education

In 22 locations and with 100 computers installed in remote villages, our field
observers began their work. Focus groups were tested for nine months, and
the results were compared with control groups and other frequent users. An
estimated 40,000 children used these computers, and many of these children
became computer literate all on their own. The average Icon Association
Inventory scores stood at an impressive 40 percent after just nine months.
We had proof of self-regulated learning, and this time we knew that it
could happen anywhere in the world, to any child and in any climate.
I decided to call the method of instruction we had developed Minimally
Invasive Education (MIE). The rest of the world continues to call it the Hole
in the Wall.
In the meantime, interest in our project was building. The government of
India gifted five kiosks to the government of Cambodia. These would be the
first kiosks to be built outside India. The library of Alexandria in Egypt got in
touch with us and said it wanted to build 30 kiosks. We sent the kits to do so,
along with detailed instructions. The Council for Scientific and Industrial
Research (CSIR) in South Africa built two kiosks in South African villages.
The results from South Africa were identical to the results we had found in
India.
Interestingly, the French embassy in Delhi, with some coaxing from Pari,
sent Pascal Monteil, a well-known new media artist from Paris, to three of the
villages where we had installed computers on playgrounds. Pari went along
with him to ensure that he did not “teach” the children anything. The children
observed and questioned Pascal intently. Soon, they had taught themselves
digital photography using his camera and had learned how to use Adobe
Photoshop in French! Then they started to draw. Moumita, an unkempt little
girl from Bishnupur, beautifully summed up the experience of creating
artwork on the PC: “This does not exist anywhere,” she said, “except in your
mind and the computer.”
 One of the manipulated images created by children in Bishnupur,
West Bengal

Pascal took more than 100 digital creations back with him to Paris. Later, he
noted that this experiment altered his preconceptions of how art should be
taught. Art is in your head, he realized, and you can learn it all by yourself if
given the right stimuli.
After the test with Pascal, we gathered up our observations and
conclusions and asked ourselves: What did we find? What will they learn?
How do we prevent misuse of the computers? And, how does Minimally
Invasive Education work?
What did we find?

When working in groups, 6- to 13-year-old children do not need to be

“taught” how to use computers. They can teach themselves. Their ability to
do so seems to be independent of these factors:

1. educational background
2. literacy level in any language
3. social or economic status
4. ethnicity and place of origin
5. gender
6. genetic background (i.e., race, culture, etc.)
7. geographic location (i.e., city, town or village)
8. intelligence
What will they learn?

Using the Hole in the Wall setup with a single PC, children can learn to do
most or all of the following tasks in approximately three months:

1. basic computer navigation functions, such as click, drag, open, close,

resize, minimize and menu selection
2. drawing and painting pictures on the computer
3. loading and saving files
4. downloading and playing games
5. running educational software and other programs
6. playing music and videos, and viewing photos and pictures
7. surfing the Internet, if a broadband connection is available
8. setting up email accounts
9. sending and receiving email
10. using social networking programs, such as chat rooms (AIM, Google
Chat, etc.), Skype and Facebook
11. simple troubleshooting, such as fixing speakers that aren’t playing
sound
12. downloading and playing streaming media

In addition, local teachers and field observers noted that the children
demonstrated improvements in these areas:

1. enrollment, attendance and performance on school examinations,

particularly in subjects that deal with computing skills
2. English vocabulary and usage
3. concentration, attention span and problem-solving skills
4. working cooperatively and self-regulation

Our research showed that a very large number of children could benefit
from the small number of kiosks we had built. In independent studies
conducted at Madangir, New Delhi, three organizations concluded that 6,000
out of a total of 9,000 children in the area were now computer literate as a
result of the Hole in the Wall kiosks. This was achieved in three years using
only 20 computers, which suggests that up to 300 children can share one
playground computer.
Porn patrol

There were other strange changes. A reduction in petty crime! In the slum the
local adults would say, “Those computers take away the children’s free time;
petty crime and other naughty stuff will not happen.” I never thought of it
that way.
In addition, there was a frequently voiced concern that children would
have access to pornographic material through the Internet-ready kiosks.
While occasionally this has happened, it is quite rare. In five years, less than
1.5 percent of kiosk Internet usage was for pornographic content. Why? First,
our kiosks were designed for children under the age of 15. This audience has
only marginal interest in pornography. Additionally, our kiosks are placed in
highly visible public places. It is rather difficult for children to browse
pornographic content when they are in heterogeneous groups and when the
screen is visible to passing adults. Finally, our kiosks are monitored remotely
and the screens are visible over the Internet. The children know this, as every
kiosk has a sign announcing the surveillance.
In the past four years, the few instances of pornographic access recorded
by the remote monitoring system were all committed by adults. Such access
by adults is not more common because we incorporated some specific design
features into the kiosks in order to avoid this. Since the kiosks are intended
for children, they are placed lower than the average adult’s height. Thus,
adults must stoop to see the screens. A lid placed on top of the screen acts as
a sunshade and makes it difficult for taller people to stoop under it and access
the screen. Also, the protective cowl on top of the keyboard and mouse has a
gap that is just large enough for small hands. Last but not least, the seating
rod is placed close enough to the kiosk to not allow sufficient legroom for an
adult to sit on it.
All of these features ensure that an adult needs to be in a rather peculiar
position to use the kiosk. It is interesting that some adults have, nevertheless,
not been thwarted in their attempts to misuse the kiosk. Fortunately, their
numbers are small.
How Minimally Invasive Education works

Certain common observations from our experiments emerged, suggesting the

following learning process occurs when children self-instruct in computer
usage:

1. Discoveries tend to happen in one of two ways: When one child in a

group already knows something about computers, he or she shows off
those skills to the others. Or, while the others watch, one child explores
randomly in the GUI (Graphical User Interface) environment until an
accidental discovery is made. For example, the child may discover that
the cursor changes to a hand shape at certain places on the screen.
2. Several children repeat the discovery for themselves by asking the first
child to let them try it.
3. While in Step 2, one or more children make more accidental or
incidental discoveries.
4. All the children repeat all the discoveries made and, in the process,
make more discoveries. They soon start to create a vocabulary to
describe their experiences.
5. The vocabulary encourages them to perceive generalizations, such as,
“When you click on a hand-shaped cursor, it changes to the hourglass
shape for a while and a new page comes up.”
6. They memorize entire procedures for doing something, such as how to
open a painting program and retrieve a saved picture. Whenever a child
finds a shorter procedure, he or she teaches it to the others. They
discuss, hold small conferences, make their own timetables and research
plans. It is important not to underestimate them.
7. The group divides itself into the “knows” and the “know-nots,” much as
they might divide themselves into “haves” and “have-nots” with regard
to their possessions. However, a child that knows will share that
knowledge in return for friendship and reciprocity of information, unlike
with the ownership of physical things, where they can use force to get
what they do not have. When you “take” information, the donor doesn’t
 “lose” it!
8. A stage is reached when no further discoveries are being made and the
children occupy themselves with practicing what they have already
learned. At this point, intervention is required to plant a new seed for
discovery, such as, “Did you know that computers could play music?
Here, let me play a song for you.” In the Hole in the Wall computers,
such minimal intervention happens accidentally from passing adults or
just by accidental discoveries. Usually, a spiral of discoveries follows
and another self-instructional cycle begins.

In order for the above instructional objectives to be met, it is important that

the following actions take place:

1. The computer should be in an outdoor, public, and safe location.

Children and their parents are apprehensive of enclosed spaces such as
closed rooms or “clubs.” Locating computers indoors, even inside a
school, may trigger fear of regimentation, control, examination,
pedophilia and other negative behaviors, as perceived by children.
Locating a computer on a school playground, on the other hand, gives
connotations of fun, accessibility, exploration and freedom. This is
because the computer does not belong to anyone nor has an institutional
context, again as perceived by children.
2. Children should use the computer in heterogeneous groups. Since the
MIE process depends on exploration and discovery, working in groups
is essential. Collaborative constructivism is the main paradigm of MIE.
Children teach one another very effectively, and they also are highly
capable of self-regulating the process. This is the main reason 100
children can use one computer.
3. There should be no adult intervention or supervision in the children’s
use of the computer, except with regard to the children’s safety and
well-being. Adults should not use the kiosk. All activity should be
monitored remotely to ensure that the kiosk is being used for the right
purpose.
4. PC functioning and Internet connectivity should be reliable and checked
frequently.
 Based on the experiences and data we have gathered over the past 12
years, we feel that such playground access points should be a part of every
primary school, where such schools exist. In locations where primary schools
are not available, these kiosks could provide even more vital educational
inputs.
The Hole in the Wall study showed conclusively that groups of children
can teach themselves to use a computer and navigate the Internet, irrespective
of who or where they are, what language they speak and whether they’ve
attended school or not.
Minimally Invasive Education for children through kid-friendly public
Internet kiosks should form an integral part of primary education in the 21st
century. It has the potential not only to rapidly close the digital divide but
also to unlock the sort of creative potential for children’s self-development
that eminent educationists (those who study the theory of education) have
sought to inspire for more than a century.
 2062: The cube in the tank

When Rita was 8, her father bought her a fish tank. It was a large, cylindrical
one, and it held four lovely goldfish. To Rita, their tails looked like sails of
silk. Rita looked after the fish, programming her microFAB to feed them the
exact amount of food they needed each day.
But everything went wrong, and it was all because of Lak. It was a Sunday
in March when Lak came to spend the day with her. They decided to build a
virus using their microFAB — just a single strand of DNA coiled up inside a
protein sheath. Lak found the genome for goldfish, and they modified the
gene to make the tail four times larger than it would be naturally. Rita
wanted to make her goldfish’s sail-tails huge! When the microFAB had done
the job, they took a dozen of the new cells and put them into her fish’s food.
The next morning, the fish were dead, Lak was gone and Rita was really
unhappy. She put the dead fish in the input chute of the microFAB, emptied
and cleaned out the tank, and erased the virus program from the microFAB.
Then she cried.
The tank has remained empty. Her father put a light over it so that the
empty tank would look like a sculpture. Rita still felt bad when she looked at
the tank, but as the months passed her sadness diminished.
Rita thought of the cube on a cloudy Saturday afternoon when her father
was a bit groggy from his bloody mary and had put his V-screen on dream
mode. She grouped with Dev, Lee and Gina on a Presence site so they could
talk and see and all be together, to design what the cube would do.
The cube would be 5 millimeters long on each side. Two sides would have
a microphone and a light that could produce colors by combining red, green
and blue. Two other sides would have a light sensor and a speaker. The
remaining two sides would have solar cells to power the electronics. When
the cube sensed a sound, it would produce a flicker of light in response to the
frequency of the sound. When the cube sensed light, it would produce a burst
of sound in response to the color and intensity of the light.
Lee made the first two cubes on his microFAB. They worked well. When he
shone a light on one, it turned the other one off, and then the two cubes made
all sorts of lights and sounds for a while. It was really cool, and Rita decided
to borrow the fabrication program from Lee.
The microFAB uses air to make things. It takes carbon, hydrogen, nitrogen
and oxygen out of the air and combines them to make physical materials.
There isn’t all that much hydrogen in the air, so you do need to put some
water into the microFAB’s input chute in order to speed up the process. The
toilet flush is connected to the microFAB, and it gets water and all other
trace elements from the toilet waste. Rita’s microFAB is quite small — only 1
meter wide, 2 meters long and 2 meters high. Her father’s FAB is much
bigger, but then it does so many other things. He uses it to make all their
food, and he even made her bicycle with it.
Rita used her microFAB to make 10 cubes, and she put them all in the
empty goldfish tank. She laughed as the cubes squealed and flashed colors
every time she put the lights on. She put the fabrication program into an
endless loop, and her microFAB calculated it could produce 144,000 cubes
every 24 hours. Rita pressed Go even though she knew her father would
complain about the energy bill. To help things along, she put most of her
dinner into the input chute as well. It just fit into the chute and would provide
lots of hydrocarbons and polymers from proteins to make the work easier,
she figured.
Beyond the Hole in the Wall

A decade has passed since the first children encountered a Hole in the Wall
computer, and we continue to hear success stories from those we’ve reached
with the kiosks. “The Konkan Harvest — An ICT Story,” a recent
documentary film by Swati Desai, reports on a girl in rural Maharashtra who
has gone on to study engineering as a result of her encounter with the
computer in the wall. Another Hole in the Wall child researches
biotechnology at the Tata Institute of Fundamental Research, one of India’s
premier institutions. In several schools, we find evidence of an improvement
in English, math and science test scores.
Once I knew that children could learn to use computers without adult
intervention or formal instruction, I decided to find out what else children
could learn on their own. Among the many experiments we did, there are
three that stand out as examples of new ways of learning that children are
able, and indeed eager, to use.
In Hyderabad, India, groups of children showed significant improvements
in English pronunciation with only a few hours of independent practice. They
used a computer and a speech-to-text program that used a native English
accent. The results are published
(mitpressjournals.org/doi/pdf/10.1162/itid.2003.1.1.75?cookieSet=1), and
they show that speech-to-text engines can be used for self-regulated
improvements in pronunciation.
In 2006, professor James Tooley of Newcastle University received a large
research grant to improve private education for the poor in Hyderabad. He
invited me to join the university and work with him on this project. In
November 2006, I moved to Newcastle. It was the first time in my life I had
not lived in India.
The Newcastle faculty loved the idea of the Hole in the Wall, but they
thought it was a bit ambitious to say that children might teach themselves
complicated things on their own. Pointing and clicking at a computer screen
is one thing, but learning a subject is another. Or is it?
In an attempt to find a limit to self-organized learning, we studied 10- to
14-year-old Tamil-speaking children in a remote Indian village in order to
determine their capacity for learning basic molecular biology. Initially, the
children worked on their own using a Hole in the Wall public computer
facility, and later they worked with a mediator who had no knowledge of the
subject. The research question would be simple but ambitious: Using a street-
side computer, can Tamil-speaking children in a village in southern India
teach themselves biotechnology in English? We would compare these
learning outcomes with those of children at a nearby average to below-
average state government school who were about the same age as the Hole in
the Wall children and who were not fluent in English but who were taught
biotechnology in school. We would also compare the results with those of
another group of children at a high-performing private school in New Delhi
who were fluent in English and had been taught this subject by qualified
teachers.
We began our experiment at the village of Kalikuppam in southern India,
which was still struggling to regain its footing after a devastating tsunami.
Using access to a Hole in the Wall computer, children taught themselves
basic biotechnology and were able to attain a test score of 30 percent in just
two months. They had started with a score of zero.
If Tamil-speaking children could teach themselves biotechnology in
English on their own, how far could they go with someone there to help
them? A score of 30 percent correct may seem impressive for children who
had started at zero, but it was still a failing grade. We decided to ask a local
young woman, Rekha Peter, who was working for a nongovernmental
organization, to help us go further.
Rekha was a good friend of the children in the town; sometimes they even
played football together. Rekha had no background whatsoever in
biotechnology, but she took on the role of an untrained, unknowledgeable but
friendly mediator. She admired the children’s learning and encouraged them
to go further. She used a method that grandmothers have used instinctively
for ages – tapping into the children’s desire to impress one another and their
adult friend.
In two more months the biotechnology test scores in Kalikuppam rose to
more than 50 percent. We found that the village children who had access to
computers and Internet-based resources only through the Hole in the Wall
learning stations achieved test scores comparable to those at the local state
school and, with the support of the mediator, equal to their peers in the
privileged private urban school of Delhi.
 I brought these results back to Britain, and a fortuitous accident happened:
“Slumdog Millionaire,” the film that won eight Oscars in 2009, was released.
The movie tells the story of a Mumbai teenager competing in a popular TV
game show called “Who Wants to Be a Millionaire?” On the show, the boy is
very successful at answering the game’s questions because he has picked up a
great deal of useful knowledge throughout his impoverished life. “Slumdog”
was based on a novel called “Q&A” by Vikas Swarup called. To my delight
and surprise, Swarup told an Indian newspaper, “I was inspired by the Hole
in the Wall project, where a computer with an Internet connection was put in
a Delhi slum. When the slum was revisited after a month, the children of that
slum had learned how to use the World Wide Web. That got me fascinated,
and I realized that there’s an innate ability in everyone to do something
extraordinary, provided they are given an opportunity.”
Newspaper journalists interviewed me, and I took advantage of my
newfound celebrity. In an article that appeared in the Guardian, I invited
British grandmothers to volunteer an hour of their time, over Skype, to talk
with children in the slums and villages of India. Within days, I had over 200
volunteers — men and women of all ages, many of them retired teachers, all
willing to give their time for free. It was a wonderfully energetic and very
British response.
In the months that followed, 40 of these eMediators went on to conduct
more than 200 hours of contact with children in India. They read the children
stories, played games with them, and chatted about their two countries. We
are currently measuring the effects of this on the children’s English
communication skills.
Suneeta Kulkarni, who has a Ph.D. in child development, stepped in to
manage the Granny Cloud, as everyone fondly called it. Using a wikisite,
solesandsomes.wikispaces.com, she put together a network of free teachers
on the Internet. Many were highly experienced retired teachers, and others
were practicing teachers. They were from all over the world. Suneeta went on
to create a Facebook page, and the Granny Cloud became a vibrant
community.
Schools from Colombia and other countries around the world joined in. In
Gateshead, England, 10-year-olds working in groups could already answer, in
less than 20 minutes, GCSE questions that they would encounter six years
later. I asked them if they could have done this in less time if they had not
shared a computer in groups but had worked individually on their own
laptops. They said they could not have done it that way at all.
Rise of the SOLE

Throughout our discoveries, a nagging question still remained: Is this

learning?
Emma Crawley is a primary school teacher working at St. Aidan’s Church
of England Primary School in the underprivileged town of Gateshead,
England. She took me to her school in June 2009, and we began a set of
experiments that would lead us toward a strange understanding of children’s
learning in a self-organized, connected world.
Emma and I created a Self-Organized Learning Environment (SOLE) in a
regular classroom. The children worked in groups of three, four or five, and
each group had a computer connected to the Internet. They were allowed to
talk with one another and to move around and see what other groups were
doing. (Remember, we saw in the Hole in the Wall experiment that clustering
around a computer is an effective learning strategy for children.) The children
were given selected GCSE questions to work on, and they answered most of
them correctly. We then waited two months and tested them again, this time
without a computer and with each student taking the test alone — just like a
regular examination.
The children’s scores never dropped. Strangely, some children scored
higher on certain questions after the two-month break than they had on the
day of the SOLE. We suspect this was because the children continued
researching on their own. They had near perfect recall of the answers.
But again, was this learning?
The problem is that there appears to be no test of “deep” learning, that I
know of, in which the learner analyzes new information and links it to what
he or she already knows with the goal of long-term retention and
understanding. Everyone seems to know what deep learning means, but I
can’t find a test for it. Governments define learning as the ability to answer
questions on tests such as the GCSE (General Certificate of Secondary
Education). By that criterion, children seem to be able to “learn” using
SOLEs alone.
I had a new hypothesis: Given the appropriate digital infrastructure, a safe
and free environment, and a friendly but unknowledgeable mediator, groups
of children can pass their school-leaving tests (such as the GCSE in the U.K.
and the Central Board of Secondary Education, or CBSE, exam in India) on
their own.
By 2010, schools all over the world were beginning to show interest in
SOLEs. Two school principals from Melbourne, Australia — Brett Millott
and Paul McKenna — have picked up the method and have taken the idea of
the SOLE far beyond what I was doing. Instead of making a SOLE an
alternative to schools, they have found ways to integrate the new
methodology into regular schooling, something I could never have done.
It’s not all good news everywhere, though. A SOLE will work in a school
if a principal wants it to. Ravi Bisht and Vikram Kumar built beautiful
SOLEs in slum schools of Hyderabad. Many are now used for storing old
files or are used by teachers and other adults. There is pornography on the
system logs. When asked if they use their SOLEs, the children look afraid,
forlorn. That expression sits like a stone on my soul. It makes Suneeta
physically ill. I question my abilities as an administrator and implementer. I
must find out why.
As I moved from school to school all over the world, I started to
experiment with test questions that were harder and deeper than the dry-as-
dust questions of the GCSE. In Hong Kong, a wonderful teacher constructed
a brilliant question: “How does an iPad know where it is?” The children got
to work and came up with the answer: GPS. The teacher followed up: “What
is GPS, and how does it work?”
Thirty minutes later, the children told us that the iPad uses three satellites
to find its position. Why three? They said one satellite will locate you in a big
circle, and two satellites will locate you in a long line, but three satellites will
use trigonometry to find the exact point you are on. So, three satellites is the
minimum number of satellites you can use for finding an exact position. They
wanted to learn more about how it worked.
Again and again, we saw SOLEs open doors to subjects that typically
make no sense to children. In Turin, Italy, and Montevideo, Uruguay, 10-
year-old children confronted the question “Who was Pythagoras and what did
he do?” Their computer screens were filled with right-angle triangles in about
45 minutes. The mood was solemn. They described the Pythagorean theorem
to me, but they wanted more time to investigate further.
In Turin, the children told me how they went from Pythagoras to Lorentz
transformations and the special theory of relativity. In Uruguay, the children
said space doesn’t necessarily have to be three-dimensional, but if it is, the
Pythagorean theorem lets us study it. Mabel Quiroga, an enterprising
educator from Argentina, was with me. Mabel asked the children what they
liked about this way of learning. “Four brains are better than one,” they told
her.
In Gateshead, 9-year-olds told me electricity is a flow of electrons and that
electrons are so small you can’t really tell where they are because, when you
shine light on them, the light blows them away. Electrons are really clouds of
probability, they said.
What happens to us after we die?
Can trees think?
Why do we dream?
Were Vikings smelly?
Each question triggers much discussion and many animated arguments and
bitter disputes. Finally, invariably, the discussions and the arguments and the
disputes open the door to learning.
 2062: Learning and babble

In school, the children were studying history. Rita’s group decided to

investigate Mongolia because Scott thought it was a cool name. They made
SOLEs and started to research the rather scary Ghengis Khan character.
Rita was surprised to learn that he had managed to spread his genes right
across the world, even though they had no FABs when he was alive! They
made a 3-D film and reconstructed the moment when Khan was poised to
attack Vienna. Those horses were cool!
On another note, Rita’s fish tank now had more than 150,000 cubes in it.
Rita put some water in the tank in the hopes of making the cubes float, but
they all sank. So she asked her BSIs about sinking, and she learned about
density. Rita made a lot of salt with the microFAB and added it a little bit at
a time to the water in the tank until the water was dense enough for the
sunken cubes to become floaters. Now the cubes were babbling and making
light patterns, and she watched them for a long time before she went off to
sleep.
In the morning, Rita churned the salt water in the tank, and the cubes
reconfigured to make different patterns. She kept the microFAB on, and it
continued making more cubes.
In a week, her father started to complain about the energy bill. But Rita
was lucky; she had the support of one of his partners. They had six living
spaces occupying an entire 90 square meters. Her bio-mom, Bidyutparna,
didn’t really understand the cubes, but Kimberley did. She told Rita’s father
that the girl was building a self-organizing network, even if Rita wasn’t
aware she was doing so. Arshia and Joan, her father’s other two partners,
were busy with projects and were hardly ever in their living spaces, so they
didn’t count. Well, thought Rita, I have it better than children in a lot of other
spaces. Look at Scott, for instance, with his bio-mom and her three partners,
none of them his biological father. No one bothered about what Scott wanted,
did they?
Her tank was quite big, thought Rita. It was more than a meter in diameter
and a meter high. She figured out the volume from a website and queried how
many 5-by-5-millimeter cubes would fit inside of it. Her BSIs said more than
10 million! It was 10 days now, and the tank was not even 10 centimeters full.
She put in a handful of cubes from the output chute and the whole mass in the
tank started blinking and babbling for hours. Her father was getting quite
interested now, and sometimes he would drop a handful of cubes in the tank
and watch them for a while. The tank made a babbling noise almost all the
time; it was very soft and a bit scary. “That’s really clever,” her father told
her, and Rita was very, very happy.
Learning as emergent phenomena

What is deep learning? This question continues to follow us throughout our

experiments. Surely, we need to understand the conscious brain and how it
works in order to understand the nature of learning. Yet, that seems to be a
tall order.
I think the nature of learning is hidden in the new science of self-
organization and emergence. To understand learning, we must understand
how self-organization happens and what leads to this mysterious process
called “emergence.”
When an audience claps continuously for a long time, the claps start to
come together in a boring, rhythmic kind of way, as if there were a conductor
waving his baton and saying, “Clap, clap, clap, clap.” Only there is no
conductor. The sounds of the claps self-organize, and the rhythm is emergent.
No one organized it. It just happened.
When gusts of wind move about, they sometimes come together into a
swirling vortex of air. A storm is born. No one coordinated the wind, it just
happened.
A flock of birds will take flight in no particular order. In moments, they
form a perfect V as they fly away. No single bird was chosen leader, and no
single bird said to make a V. It just happened.
A beehive of perfect hexagons is made by thousands of bees working
together. Bees don’t know anything about geometry or hexagons, and they
don’t have a master architect guiding them. They just deposit their wax. The
hexagons happen.
Each of these is an example of emergence — the appearance of a property
not apparent in a system.
The only condition for self-organization and emergence seems to be that
every part of the system must be connected in some way to every other part.
For example, neurons in the brain are simple switches, but connect them all
together and the whole mass begins to think. Could education be a process of
self-organization, with learning being the emergent outcome?
Connectedness

The fact that carbon atoms can form bonds, which in turn form long chains, is
central to the emergence of life on Earth. The primordial soup of organic
material in the oceans of ancient Earth self-organized to form DNA and
related genetic material. Such materials are self-referential; that is, they carry
information about themselves. This enables reproduction, evolution and life.
Self-reproducing and mutating organisms adapt themselves to their
environment. The fittest survive. On Earth, self-organizing systems have,
invariably, developed sense organs with which to perceive their
environments. Such perception of the environment is crucial to the
organisms’ survival.
Moreover, living organisms are also capable of sensing their own internal
condition and of modifying their behavior on the basis of their understanding
of themselves and the external environment. This “understanding” is referred
to as cognition. Finally, cognition seems to lead to consciousness, a state that
we understand instinctively but have difficulty in defining. This chain of
events can be generalized as follows:

Connected systems can, under suitable conditions, lead to self-

organization.
Self-organizing systems can, under suitable conditions, develop
cognition.
Cognitive systems can, under suitable conditions, become conscious.

Let’s look at each type of system.

Connected system. A connected system is a collection of interacting parts
functioning as a whole. It is distinguishable from its surroundings and has
recognizable boundaries. It has emergent properties not contained within any
of the individual parts. In popular language, a system is defined as being
“greater than the sum of its parts.”
Self-organizing system. Connected systems tend to self-organize. There
are many examples of such self-organization: flocks of birds, lines of
marching ants, rhythmic applause. Self-organizing systems operate with a
few simple rules to produce patterns. For example, an ant may be
programmed to simply follow another ant, and the larger pattern that resulted
would be a line of “marching” ants. A bird may fly at a certain distance and
angle from another, and the larger pattern that resulted would be a “flock.”
Fractals are perhaps the most dramatic example of a self-organizing system in
the realm of computing. Self-organization is being extensively studied and
could form one of the fundamental branches of the natural sciences in the
near future.
Self-organization in human networks has been less studied, but
collaborative environments, relay chats and digital communities on the
Internet seem to indicate that simple rules of connectivity can lead to larger
patterns of human behavior.
Cognitive system. A cognitive system is one that is aware both of its own
state and of the state of systems other than itself. In other words, we could
define a cognitive system as one that has knowledge about itself and about its
external environment. The problem with such definitions lies in the meaning
of the words such as “aware” and “knowledge.” For our present purpose, we
will equate awareness with knowledge and also propose that both words be
defined with respect to connectedness. Hence, we propose that a system is
aware of a parameter (i.e., has knowledge of it), either internal or external to
itself, only when a change in that parameter causes a change in its own state.
Being connected would then become a prerequisite for awareness and
knowledge.
It is interesting to note that human beings generally consider knowledge to
be irrelevant unless it “does something” for them. Much of what’s wrong
with formal education lies in the fact that students don’t perceive how the
knowledge they are getting will, in any way, change their state.
Conscious systems

Using this thinking, let’s define consciousness in terms of connectedness and

anticipation. In other words, a conscious system is affected by the states of all
the systems that are, have been or will be connected to it (including its own
state). It is specifically this awareness of future states that differentiates
conscious systems from other connected systems and enables conscious
systems to anticipate.
A dog or a cat is affected by its past for a very long period of time and by
its future for only a few minutes. This is instinctive behavior. Animals, as far
as we know, can imagine only minutes into the future (something we could
call a “past priority consciousness”).
A human being is affected by a long past but also can imagine years or
decades into the future. Some humans rely more on the past (“traditional”),
and some rely more on the future (“speculative”) when forming their
awareness of the world. It is possible to imagine consciousness that relies
heavily on the far future and near past. We could call this “future priority
consciousness,” though I don’t think any examples of this exist.
Finally, a system whose state is determined by all its states in the past, to
the infinitely distant past, as well as by all its states in the future, to the
infinitely distant future, would be the highest form of consciousness. The
actions of such a maximal consciousness would be as unknowable to us as
human actions are to the animal world.
We all know that cause comes before effect. The cause is in the past, the
effect in the present. Causality, as we know it, is one way. We could call this
“time asymmetric causality.”
What if a cause could be in the future, and the effect in the present. This
would be the opposite of the kind of causality we know — also asymmetric
but in the other direction.
Nature does not like asymmetry; there is symmetry in almost everything.
Could it be that we’ve got causality all wrong? Could an effect in the present
be affected by causes in the past as well as in the future? We could then call
this “time symmetric causality.”
Consciousness, then, would be a manifestation of time symmetric
causality. I don’t know if this is right, but it’s somehow beautiful to think
about.
With the help of my friend Sujai Kumar, I tested what would happen to a
self-organizing system on a computer if it had a view of its own future. We
got a strange result: The system went into chaos and then, repeatedly,
reproduced the image of its future as though it could remember it, and then
made that future come true. We published this study: (complex-
systems.com/pdf/16-3-1.pdf).
Therefore, we can surmise the following:

Connected systems can self-organize.

Self-organizing systems show emergent behavior.
Emergence can produce cognition.
Cognition and an ability to sense a past and a future can produce
consciousness.

Could it be, then, that education is a process of self-organization? Given

enough stimuli, do our brains start to self-organize? If so, then surely
emergent phenomena will take place. Is learning an emergent phenomenon?
If learning is an emergent phenomenon, then the teacher needs to provide
stimulus — lots of it – in the form of “big” questions. These must include
questions to which the teacher, or perhaps anyone, does not have the answer.
These should be the sorts of questions that will occupy children’s minds
perpetually. The teacher needs to help each child cultivate a vision of the
future.
Thus, a new primary curriculum needs to teach only three skills:

1. Reading comprehension: This is perhaps the most crucial skill a child

needs to acquire while growing up.
2. Information search and analysis: First articulated at the National
Institute of Technology in India by professor J.R. Isaac in the early
1990s — decades ahead of its time — this skill set is vital for children
searching for answers in an infinite cyberspace.
3. A rational system of belief: If children know how to search, and if they
know how to read, then they must learn how to believe. Each one of us
 has a belief system. How soon can a child acquire one? A rational belief
system will be our children’s protection against doctrine.

Children who have these skills scarcely need schools as we define them
today. They need a learning environment and a source of rich, big questions.
Computers can give out answers, but they cannot, as of yet, make questions.
Hence, the teacher’s role becomes bigger and stranger than ever before:
She must ask her “learners” about things she does not know herself. Then she
can stand back and watch as learning emerges.
 2062: Emergence

Three months later, the tank got filled. It was a particularly bad day for Rita.
Her father was away on a rare trip to the other side of the world, and Arshia
came to pick her up from school. Arshia kept complaining about not having
time for herself and criticized Rita about everything — how she walked, how
she was too loud, how it would have been better if her bio-mom had never
had her. Rita kept quiet and walked as fast as she could. When they reached
their space, she didn’t say goodbye to Arshia.
It was December and quite dark when Rita entered her space. She was
annoyed, but she forgot her mood when she saw the tank. It was glowing with
a pretty blue-violet color. “Wow”, said Rita, as loudly as she could, just to
annoy Arshia.
The tank reacted. Spots of black and white began to blink all over it. Rita
moved closer, and she could not believe her eyes. Something that looked
vaguely like a face was flickering on and off in the tank. In fact, it looked a
bit like her own face. “Wow”, she said again, very softly this time. The
babbling from the tank stopped.
“Wow, Wreeta,” said the tank.
A step-by-step guide to Method ELSE

For more than a decade, I have conducted experiments that were designed to
draw out the best technological and pedagogical methods for implementing
self-organized learning. These are the results of research on the development
of SOLEs and of investigations into remote mediation over the Internet
within Self-Organized Mediation Environments (SOMEs) — two facets of
what I imagine education in the connected environments of the 21st century
will look like.
If you are designing a school or are in charge of improving the quality of
education in a school, I would like to offer a blueprint for Method ELSE
(short for Methods for Emergent Learning Systems in Education), a
collection of techniques and facilities that I believe can be constructed in any
school and which I have found to result in significant improvement in
children’s learning and examination performance. Before we dive into step-
by-step instructions for implementing Method ELSE, let’s go over some
definitions of terms.

Self-organizing system: A self-organizing system is one in which the

system structure appears without explicit intervention from outside the
system. Within such a system, critically interacting components self-organize
to form potentially evolving structures that exhibit hierarchy of emergent
systems properties.

Emergence: The appearance of a property not previously observed as a

functional characteristic of the system.

Now, here is my advice on how to construct a learning environment using

Method ELSE.

Step 1: Build
 Example of a SOLE for 12 to 15 children

Identify a location in the school, typically a room that is highly visible to

passing adults — including the principal, teachers and parents coming to
pick up children — as well as to other children.
Create glass walls for the room so that the entire area is visible. Put in
bright compact fluorescent lighting and paint the walls in light, cheerful
colors. The flooring should be easy to clean and should be kept dust-
free.
Include furniture with no sharp edges. All furniture should be easy to
clean. About six items of furniture will enable a class of 24 to 30
children (at four to six per group) to use the facility without
overcrowding.
Purchase enough desktop computers so that each group of four to six
children has access to one. Each computer should have a fast processor,
a large (at least 19-inch) LCD monitor, speakers, a wireless keyboard
and a wireless mouse. Secure the speakers to the desk or wall.
Place the CPU of each computer in a safe place under the tables. The
power switches on the computer and the connecting cables for the power
 supply and monitor input should not be accessible to the children. Place
monitors on a stand so that they are raised at least 12 inches above the
surface of the table. This will ensure that the monitors are clearly visible
from the outside when children are using them and that they are not
blocked from view by the children’s heads or bodies.
Ensure that all the computers have broadband Internet access at speeds
of 2 Mbps or more, if possible. Do not use firewalls, unless unavoidable.
Wireless broadband is recommended.
All electrical wiring and outlets should be concealed but easily
accessible when needed. It should not be necessary to crawl under tables
to access connections.
Install any freeware, such as Open Office, for the children to work with.
Image software such as Microsoft Paint, Adobe Photoshop or equivalent
freeware is a must.
One of the computers in the SOLE should have a web camera and
microphone installed. The camera should be permanently mounted so
that it enables a full view of the SOLE. A camera with pan and tilt
facilities and a built-in microphone is recommended (such as the
Logitech QuickCam Sphere). A person contacting the SOLE over Skype
should be able to see most of the children as they gather around the
computer with the camera.
Install adequate and appropriate power conditioning and backup. An
uninterruptible power supply (UPS) is recommended in areas where the
electricity supply is not reliable. A generator or solar panels and
batteries should be used in areas that have no electricity.
Keep a small table and chair in the SOLE for an attendant.
Check to see that all monitors are clearly visible from outside the SOLE.
Ideally, the SOLE should be set up by a vendor who will also provide a
technical person for attending to any problems. This person should be
capable of troubleshooting electrical, hardware, software and
connectivity problems.

Step 2: Manage
 A SOLE in use

At least one teacher should be trained in applying self-organizing

systems concepts in children’s education. They should understand the
principles and be trained in their use. You can contact me at
sugata.mitra@newcastle.ac.uk or sugata.mitra@gmail.com for guidance
on how to train the teachers.
There should be one attendant at the SOLE at all times. The attendant is
best described as a friendly but not knowledgeable mediator. An ideal
attendant would be an educated grandmother in her 50s or early 60s. The
attendant would ensure that the children are safe, resolve disagreements,
call for help when required and, in general, keep an eye on things,
particularly when no teacher is present. The attendant will not teach,
suggest or direct the children’s activities in any way. However, the
attendant should intervene in the unlikely event of inappropriate material
being accessed by the children. The key role of the attendant is to
admire children’s learning and encourage them to go further.
The SOLE should be open for use approximately one hour before school
and up to two or more hours after school. Preferably, it should be
 available on weekends and holidays as well. No adult should be allowed
to use the SOLE unless there are at least four children present. Any adult
using the SOLE will need to ask permission from the children and the
attendant before using the facility. The adult’s identity and the amount
of time that the adult uses the SOLE should be recorded. The SOLE is
predominantly for use by children between 5 and 17 years old who
attend the school. The chief users should be between 6 and 12 years old.
The keys to the SOLE should be with the attendant, and she or he should
know how to switch everything on and off.

Step 3: Use

Collaborative discoveries in SOLEs

Timetabled usage: Each class should have at least one session of about
90 minutes in the SOLE every week. During this time, a teacher will
engage the children with a question for them to answer using the SOLE.
Examples of questions could be, “Who built the pyramids and why?”
“What are fractals?” “What are they looking for with the Large Hadron
Collider at CERN in Geneva?” “Who is Gandhi and what did he do?”
“Where is Botswana and what is it famous for?”
For each session, the children form their own groups of four to six.
Children are allowed to change groups, talk to one another, talk to other
groups and walk around looking at other students’ work. There are very
few rules. The teacher’s role is minimal, and she or he should remain on
the sidelines. Since an attendant is present, the teacher can leave the
SOLE if desired.
About 30 minutes before the end of the session, the groups should
produce a one-page report describing what they have discovered. The
teacher can then expand on this in a later class. Use this method for 8- to
12-year-olds.
Curricular usage: This is similar to the previous example except that the
driving question is pulled from the school-leaving examination (for
instance, the CBSE in India or GCSE in the U.K.). If this is done
regularly with children between 12 and 15 years old, there will be
improvements in their examination results. Children seem to enjoy
attempting these questions on their own, preferably in the absence of the
teacher, and they seem to retain the answers very well.
Aspirational usage: In these sessions, children listen to a short lecture
from an interesting website — for example, TEDTalks. They then
research the talk in groups and present their findings. If children listen to
interesting people on a regular basis, there is evidence that it has a
positive effect on their aspirations and general knowledge.
Free usage: The SOLE should be open for use by any child in the school
for a period of time before and after regular school hours. It should be
made clear to the children that they can use this time to play games, chat
or do whatever they wish. As usual, working in groups is to be strongly
encouraged.
Remotely mediated sessions: At certain times, the SOLE can be used for
connecting to eMediators. If used correctly, this type of session can have
a strong and positive impact on cultural development and English
fluency. This is described in more detail in the next section, “Remote
links.”
 Step 4: Remote links

Children interacting with a remote mediator

Experimental results show that children who are left to their own devices to
work in groups and use the Internet can quickly and effectively attain
educational objectives. However, there are limits to their understanding and
performance. When children are encouraged by a friendly but not
knowledgeable mediator, they may be better able to reach the learning levels
of “taught” children in good schools. Such mediation is possible at a distance
using the following method.

Connect to a “cloud” of eMediators. These are usually experienced

teachers willing to provide an hour or more of their time each week
conducting sessions in SOLEs via Skype videoconferencing. They form
a Self-Organized Mediation Environment (SOME). To understand how
this works, see solesandsomes.wikispaces.com.
To organize SOME sessions for your SOLE using this method,contact
Suneeta Kulkarni at suneeta.kulkarni@gmail.com. She will inform you
of the current eMediator status and help you schedule a SOME session.
Once a SOME session is scheduled, connect to Skype using a computer
with a web camera and dial your eMediator, who should be online at
that time. It is a good idea to teach the children and the attendant to do
this.
Once a video connection is set up, the remote mediator will take over
and conduct the session.
SOME sessions are targeted for very young children, between 5 and 7
years old. The mediators will read stories to the children, and the
ensuing conversation will improve the children’s English and social
skills. Basic English literacy is absolutely essential for Method ELSE to
work.
SOME sessions can also be used for older children who wish to interact
with experienced teachers for subject-related queries or difficulties.
SOME sessions work better if a life-size image of the mediator is
projected on a wall of the SOLE. However, this can be expensive
because it requires a projector. If no projector is available, displaying the
mediator’s image on a clearly visible monitor will do, particularly if the
full-screen mode is used.
Clear audio and a noise-free environment are absolutely essential so that
the mediator can hear the children clearly, and vice versa.
It would help considerably if the children wear badges with their names
printed in large letters so that the mediator can read their names.
It is necessary to connect exactly on time and to let a mediator know
well in advance if you cannot connect at a scheduled time. Otherwise
they will lose interest in your school.
Tips from the field

I hope you find this new method of education as exciting as I do. In

conclusion, I’d like to offer the field observations of Emma Crawley of St.
Aidan’s Church of England School in Gateshead, England. She is an
instructor who has found great use in Minimally Invasive Education and is an
experienced practitioner of Method ELSE.

Practitioner field observations

Set aside time to observe the class dynamics.

Groups chosen and the roles taken by individual children within each
group may surprise you.
Children fall into a natural role easily without being given specific jobs.
There is no pressure for the children to perform as individuals, which
gives children confidence to ask questions and follow their own
curiosity.
Children discuss ideas, make connections and extend their knowledge to
new areas.
Children can give reasons for group choices and movements between
groups and can suggest improvements that can be made to the way they
work.

Developments with frequent use

 Children begin to question their attitudes toward others and their peers’
perception of themselves, which improves group performance.
Children start to find keywords in a question or topic rather than type in
a whole question.
Children begin to skim texts to find specific information, refining
searches and choosing keywords.
Children automatically check multiple sites to make sure information is
correct.
Children begin to share notes they have gathered more freely,
recognizing a common goal.
Children begin to follow leads that are of interest to them and become
increasingly curious to find out more.
Discussing the children’s notes and extending their understanding
through questioning at the end of each session deepens the children’s
understanding and allows the teacher to stretch the learning further.
The children begin to freely offer ideas as to where they would like their
learning to lead them.

Impact on Other Areas of Learning

The children begin to apply research skills independently as they work

at home, participate in their other lessons and access the Internet.
As a practitioner, you may choose to apply the group method to other
areas of the curriculum, and you should question your reasons for
choosing partnerships or groups within the class yourself.
The children become fast and efficient at producing detailed reports on
topics and become familiar with text organization and the inclusion of
pictures, photographs and diagrams.
You will see an improvement in the quality of class cohesion as
 teamwork develops.
The relationship between the teacher and children becomes stronger as
the children have the confidence to ask questions, discuss current affairs
and suggest areas of learning that they know will be welcomed,
discussed and shared.

What to Expect

Be patient. You will have to hand control over to the children for their
own learning. Trust them.
At first they will get excited and move around a lot because it’s a
novelty; this will become less chaotic the more the method is used.
Noise levels will rise and fall, but this is OK. The children need to talk
and discuss what they find.
There is generally a calm, relaxed atmosphere in the classroom.
Take the opportunity to observe the group dynamics; make notes but try
not to get involved.
If the children begin to argue or complain about others, hand the control
straight back to them by reminding them of the group rules.
Never stop a session because the children seem distracted or removed
from the learning; this is not the case. The children will become more
relaxed and quicker at research, but this is due to experience of the
method and not complacency.
 About the author

Sugata Mitra is Professor of Educational Technology at the School of

Education, Communication and Language Sciences, Newcastle University,
England, and a visiting professor at the MIT Media Lab in Cambridge,
Massachusetts. Professor Mitra works in the areas of cognitive science,
information science and educational technology, and he has a keen interest in
engineering and software development. His current research interests include
technologies for remote and rural education, distance education, instructional
robotics, self-organizing systems and collaborative systems on the Internet.
 About TED

TED is a small nonprofit devoted to "Ideas Worth Spreading." It started out,

in 1984, as a conference bringing together people from three worlds:
Technology, Entertainment, Design. Since then its scope has become ever
broader. Along with two annual conferences — the TED Conference in Long
Beach and Palm Springs, California, each winter, and the TEDGlobal
conference in Edinburgh, Scotland, each summer — TED provides the
award-winning TEDTalks video site, the Open Translation Project and Open
TV Project, the inspiring TED Fellows and TEDx programs, and the annual
TED Prize.

The annual TED Conferences in Long Beach/Palm Springs and Edinburgh

bring together the world's most fascinating thinkers and doers, who are
challenged to give the talk of their lives (in 18 minutes).

On TED.com, we make the best talks and performances from TED and its
partners available to the world for free. More than 1,000 TEDTalks are now
available online, with more being added each week. All of the talks are
subtitled in English, and many are subtitled in various other languages. These
videos are released under a Creative Commons license, so they can be freely
shared and reposted.
Beyond the Hole in the Wall: Discover the Power of Self-Organized Learning. Copyright 2012
by Sugata Mitra. All rights reserved. Published in the United States by TED Conferences, LLC.
No part of this publication may be used or reproduced, distributed, or transmitted in any form or
by any means whatsoever without the prior written permission of the publisher, except in the
case of brief quotations embodied in critical articles and review and certain noncommercial uses
permitted by copyright law. For permission requests, contact the publisher at:

TED Conferences, LLC

250 Hudson Street
New York, NY 10013
TED.com

Published simultaneously in the United States and wherever access to Amazon, the iBookstore,
and Barnes & Noble is available. First edition. First published January 2012. ISBN: 978-1-
937382-08-7

TED is a registered trademark, and the TED colophon is a trademark of TED Conferences, LLC.