Networks collection

4/1/04

4:35 PM

Page 3

Network Logic
Who governs in an
interconnected world?
Edited by
Helen McCarthy
Paul Miller
Paul Skidmore

Networks collection

4/1/04

4:35 PM

Page 8

Open access. Some rights reserved.

As the publisher of this work, Demos has an open access policy which enables anyone to access
our content electronically without charge.
We want to encourage the circulation of our work as widely as possible without affecting the
ownership of the copyright, which remains with the copyright holder.
Users are welcome to download, save, perform or distribute this work electronically or in any
other format, including in foreign language translation without written permission subject to the
conditions set out in the Demos open access licence which you can read here.
Please read and consider the full licence. The following are some of the conditions imposed by the
licence:
Demos and the author(s) are credited;
The Demos website address (www.demos.co.uk) is published together with a copy of this
policy statement in a prominent position;
The text is not altered and is used in full (the use of extracts under existing fair usage rights
is not affected by this condition);
The work is not resold;
A copy of the work or link to its use online is sent to the address below for our archive.
By downloading publications, you are confirming that you have read and accepted the terms of
the Demos open access licence.
Copyright Department
Demos
Elizabeth House
39 York Road
London SE1 7NQ
United Kingdom
copyright@demos.co.uk
You are welcome to ask for permission to use this work for purposes other than those covered by
the Demos open access licence.

Demos gratefully acknowledges the work of Lawrence Lessig and Creative Commons which
inspired our approach to copyright. The Demos circulation licence is adapted from the
attribution/no derivatives/non-commercial version of the Creative Commons licence.
To find out more about Creative Commons licences go to www.creativecommons.org

Networks collection

4/1/04

4:35 PM

Page 1

About Demos
Demos is a greenhouse for new ideas which can improve
the quality of our lives. As an independent think tank, we
aim to create an open resource of knowledge and
learning that operates beyond traditional party politics.
We connect researchers, thinkers and practitioners to an
international network of people changing politics. Our
ideas regularly influence government policy, but we also
work with companies, NGOs, colleges and professional
bodies.
Demos knowledge is organised around five themes,
which combine to create new perspectives. The themes
are democracy, learning, enterprise, quality of life and
global change.
But we also understand that thinking by itself is not
enough. Demos has helped to initiate a number of
practical projects which are delivering real social benefit
through the redesign of public services.
We bring together people from a wide range of
backgrounds to cross-fertilise ideas and experience. By
working with Demos, our partners develop a sharper
insight into the way ideas shape society. For Demos, the
process is as important as the final product.
www.demos.co.uk

Networks collection

4/1/04

4:35 PM

Page 2

First published in 2004

Demos
Some rights reserved see copyright licence for details
ISBN 1 84180 118 6
Typeset by Land & Unwin, Bugbrooke
Printed by Hendy Banks, London
For further information and
subscription details please contact:
Demos
The Mezzanine
Elizabeth House
39 York Road
London SE1 7NQ
telephone: 020 7401 5330
email: hello@demos.co.uk
web: www.demos.co.uk

Networks collection

4/1/04

4:35 PM

Page 4

Open access. Some rights reserved.

As the publisher of this work,Demos has an open access policy which enables anyone to access our
content electronically without charge.
We want to encourage the circulation of our work as widely as possible without affecting the ownership
of the copyright,which remains with the copyright holder.
Users are welcome to download,save,perform or distribute this work electronically or in any other format,
including in foreign language translation without written permission subject to the conditions set out in
the Demos open access licence which you can read at the back of this publication.
Please read and consider the full licence.The following are some of the conditions imposed by the licence:
Demos and the author(s) are credited;
The Demos website address (www.demos.co.uk) is published together with a copy of this policy
statement in a prominent position;
The text is not altered and is used in full (the use of extracts under existing fair usage rights is not
affected by this condition);
The work is not resold;
A copy of the work or link to its use online is sent to the address below for our archive.
Copyright Department
Demos
Elizabeth House
39 York Road
London
SE1 7NQ
United Kingdom
copyright@demos.co.uk
You are welcome to ask for permission to use this work for purposes other than those covered by the
Demos open access licence.

Demos gratefully acknowledges the work of Lawrence Lessig and Creative Commons which inspired our
approach to copyright.The Demos circulation licence is adapted from the attribution/no derivatives/noncommercialversion of the Creative Commons licence.
To find out more about Creative Commons licences go to www.creativecommons.org

Networks collection

4/1/04

4:35 PM

Page 5

Contents

1.
2.
3.
4.
5.
6.
7.
8.
9.

Acknowledgements

Introduction
Helen McCarthy, Paul Miller and Paul Skidmore

Living networks
Fritjof Capra

23

Towards a theory of government

Karen Stephenson

35

Connexity revisited
Geoff Mulgan

49

Untangling the threads

Ann Lieberman and Diane Wood

63

Networks, knowledge and innovation

David H Hargreaves

77

Leading between
Paul Skidmore

89

The science of inequality

Mark Buchanan

103

Old boys and new girls

Helen McCarthy

115

Networks collection

4/1/04

4:35 PM

Page 6

Network logic

10.
11.
12.
13.
14.
15.
16.
17.

Your friendship networks

Dr Perri 6

129

Developing the well-connected community

Alison Gilchrist

143

Networks and neighbourhoods

Robert J Sampson

155

Organising for success

Diane Coyle

167

The information utility

John Taylor

177

Smart mobs
Howard Rheingold

189

The rise of network campaigning

Paul Miller

205

Afterword: why networks matter

Manuel Castells

219

Networks collection

4/1/04

4:35 PM

Page 7

Acknowledgements

We are grateful to the National College for School Leadership for

supporting this publication, and to all the authors who generously
agreed to contribute essays. We would like to thank in particular
Karen Stephenson for her helpful comments and support for the
project.
Many individuals at Demos helped to make this collection happen,
not least Tom Bentley who has provided guidance and direction
throughout. Matthew Horne made important contributions at the
development and commissioning stages, and Julia Huber provided
crucial editing support. We must also thank Eddie Gibb for his input
in the final stages and Bobby Webster, who saw the collection through
to production.
As usual, all remaining errors or omissions remain our own.
Helen McCarthy
Paul Miller
Paul Skidmore
March 2004

Demos

Networks collection

4/1/04

4:35 PM

Page 8

Networks collection

4/1/04

4:35 PM

Page 9

Network Logic 1

We are some way

from being able to
structure public
and organisational
power in ways
which really
harness network
potential . . .
Introduction
Helen McCarthy, Paul Miller
and Paul Skidmore

Networks collection

4/1/04

4:35 PM

Page 10

Networks collection

4/1/04

4:35 PM

Page 11

1. Introduction
network logic
Helen McCarthy, Paul Miller and
Paul Skidmore

Networks are the language of our times. Think about Al-Qaeda. The
internet, eBay, Kazaa. The mobile phone, SMS. Think about iron
triangles and old school ties, No Logo and DeanforAmerica. Think
VISA and Amex, the teetering electricity grid, the creaking rail
network. LHR to LAX. Think about six degrees of separation. Think
small worlds, word of mouth.
Think about your networks. Your friends, your colleagues, your
social circle. How new networks take shape through introductions at
parties, over coffee breaks, via email. How your connections have
helped you, supported you and hindered you.
They are all around us. We rely on them. We are threatened by
them. We are part of them. Networks shape our world, but they can
be confusing: no obvious leader or centre, no familiar structure and
no easy diagram to describe them. Networks self-organise, morphing
and changing as they react to interference or breakdown.
Networks are the language of our times, but our institutions are
not programmed to understand them.
As individuals, we have taken advantage of the new connections: to
earn, learn, trade and travel. But collectively we dont understand
their logic. Our leaders and decision-makers have often failed to grasp
their significance or develop adequate responses. We do not know
how to avoid internet viruses or manage mass migration, structure

Demos

11

Networks collection

4/1/04

4:35 PM

Page 12

Network logic

urban communities, regulate global financial markets or combat

networked terror.
So now we live in a world held together by networks, but lacking
the language to solve its common problems. Were left with a sense of
unease a governance gap that needs to be bridged. This book brings
together some of the leading network thinkers and practitioners to
help us to do just that.
A new logic
Manuel Castells, in his afterword (see chapter 17), points out that we
are paying so much attention to networks now because of computerisation; it is electronic connections that have made the network such a
ubiquitous and public organising principle. But as both Fritjof Capra
and Karen Stephenson argue in their essays, these forms go far
beyond the digital. Networks embody a set of fundamental principles
for the ordering, distribution and coordination of different components, whether chemical, natural, social or digital. Network principles
help to explain not just the distribution of wealth in monetary
economies, but also the distribution of molecules in cellular systems.
If we can recognise and detect these patterns more accurately, we
could learn to use them for organisation and decision-making, to
make possible new forms of coordination and collective action.
There is already a huge appetite for understanding the hidden
connections in the world around us, as the combined book sales of
our authors would testify. But we are some way from being able to
structure organisational and public power in ways that really take
advantage of network potential. As Capra puts it, we need new design
principles for our future social institutions.
Whether it is harnessing the distributed processing power of
millions of networked computers as John Taylor envisages, or using
Robert Sampsons penetrating analysis of crime in the modern city to
activate problem-solving neighbourhood networks, there are
opportunities and challenges for governments, businesses and
citizens alike. Meeting them means redefining many of the principles
that currently hold our world of public institutions and assumptions
12

Demos

Networks collection

4/1/04

4:35 PM

Page 13

Introduction

together. In the rest of this essay we examine the implications for

some of these key principles:

communication
transparency
knowledge
innovation
regulation
accountability
ownership
citizenship
power.

Communication

The networks that have changed our lives most in the last two decades
have been communication networks, especially the internet and
mobile technologies. The most important shift is away from
broadcast (one to many) towards conversational (many to many)
models of communication. The music industry is not alone in finding
this new era of horizontal, peer-to-peer communication uncomfortable. No major politician, with the partial exception of Howard
Dean, has yet grasped the possibilities of the internet as an
organising, fundraising and communicating tool in the way that
Franklin Roosevelt mastered the fireside radio chat or John F
Kennedy the relaxed television address. Political communication for
the internet age has yet to truly arrive.
The media itself faces significant new challenges, not least from the
explosion in self-publishing that new media permit. The Monica
Lewinsky scandal that engulfed the Clinton presidency was initiated
by an ostensibly fly-by-night website called the Drudge Report. In the
Iraq conflict the controversy over embedding mainstream journalists
in operational military units, and the accusation that this enabled the
military to filter what was and was not witnessed, played out in
parallel with the highly personal missives of the Baghdad Blogger. An
anonymous middle-class Iraqi calling himself Salam Pax and running
Demos

13

Networks collection

4/1/04

4:35 PM

Page 14

Network logic

a weblog called DearRaed attracted global interest for the real-time,

authentic, and somehow more objective perspective he appeared to
offer.
Among organisations for whom the careful manipulation and
transmission of images and messages is a fine art, there is much to be
unlearnt. Critically, they will have to look for ways to insert themselves effectively into ongoing conversations, which they can shape
and influence but probably not control. Understanding the way that
messages are communicated and spread horizontally across networks
is becoming more and more highly prized in this regard, as the
immense popularity of Malcolm Gladwells The Tipping Point
demonstrates. In their essays, Ann Lieberman, Diane Wood and
David Hargreaves explore the possibilities of lateral networking and
collaboration between teachers to spread good practice rapidly across
school systems. It will be interesting to watch the way that individuals
and organisations who are seen to act gatekeepers or hubs in
influencing peoples behaviour are identified and enrolled by
governments keen to influence the actions of citizens or the
performance of public services, by campaigners wanting to rally
people to a cause, or by companies hoping to sell them a product.
Shoshana Zuboff and James Maxmin forecast a bright future for
those firms trusted by consumers to act as brokers, stitching together
personalised combinations of goods and services from a diverse
network of providers.1
Transparency

In a hyperconnected society, secrets get everywhere. Transparency has

gone from being something that institutions permit to something
that they find it hard to resist. A misplaced click of a mouse, a careless
conversation, and secrets are transmitted round the world in a matter
of seconds and reported globally within hours. Wave after wave of
scandals, from Iran-Contra to BSE to Enron, have illustrated the
problems that a lack of transparency can create. Each has prompted
policy-makers to extend powers of scrutiny so that such obliquity will
not be repeated.
14

Demos

Networks collection

4/1/04

4:35 PM

Page 15

Introduction

The dynamic of information in a network is one of openness. As

Stewart Brand puts it: Information wants to be free. It would be hard
to overestimate the challenge that this poses. Access to information is
an important source of power for professionals and organisations of
every stripe, and many institutional cultures have been built on
secrecy and insularity. The recent Hutton Inquiry process was
remarkable less for its final outcome than for what it revealed about
the machinations of Whitehall.
With the right kinds of transparency, it could be possible to rebuild
public trust in institutions that have taken a battering over recent
decades, but also to improve organisational performance itself. Take
government policy-making, for example. The image of policy as a
rigid, linear production line from conception to winning support to
implementation could evolve into something much more interactive
and adaptable. Policy could be developed through genuine dialogue,
tried out in small-scale ways and adapted in light of the results.
Implementation on a larger scale would not follow until evidence
about what worked had been clearly marshalled. Citizens would be
less the passive arbiters of government performance and more the
active co-creators of new policy. Such an approach would have to be
accompanied by a different set of expectations, however, among both
public and media. All institutions make mistakes, and good decisions
depend on the capacity to consider a wide range of options. Without
the scope to do so without scandal or sensation, secrecy will remain a
default position among governments.
Knowledge

Brand also argued that information equally wanted to be expensive,

because at the right time and in the right place it was so immeasurably valuable. This tension, and its specific application in debates
ranging from intellectual property and scientific publishing to
research and development and human capital banking, will manifest
itself more and more frequently in the future.
The open source movement has shown that it is possible, given
the increased capacity for coordination that new communication
Demos

15

Networks collection

4/1/04

4:35 PM

Page 16

Network logic

networks permit, to create popular, robust and user-focused goods

and services by tapping into the collaborative instincts of humans as
social animals, not merely our competitive instincts as rational
economic calculators. An approach that originated in software
programming through things like the Linux computer operating
system has now been successfully applied in the pharmaceutical
industry and other knowledge-intensive sectors. Former BBC
Director-General Greg Dyke announced plans to make the entire
BBC archive available free online, opening up the possibility of an
open source approach to broadcast content. The Creative Commons
movement has sought to anchor these developments in an evolving
statement of shared values and legal principles.
Yet despite its attractions the open source model has certain
features that may reduce its applicability to wider social and
economic endeavours. Most obviously, it depends on restricting
participation to those with high levels of technical knowledge, and
through a combination of peer review and central coordination
it has a relatively clear mechanism for validating this knowledge.
Where these conditions do not apply, then it will be crucial that we
develop much more sophisticated systems of human capital banking
for certifying what people know, and allowing us to find out
which people know what. This is not just a challenge for
governments, since the task of certifying knowledge must be
distributed across the domains in which that knowledge is acquired
and deployed. Instead, as Riel Miller has argued, it should be seen
as a long-term collective project analogous in difficulty and importance to the development of a reliable financial capital banking
system.2
Innovation

As a growing body of research makes clear, innovation is not a lonely

process undertaken by Freds in sheds. It is itself a network
endeavour.3 Good ideas are dependent on an environment that is
supportive of collaboration. Partly this is about getting the right mix
of people, skills and experience. American professor Richard Florida
16

Demos

Networks collection

4/1/04

4:35 PM

Page 17

Introduction

has argued that diversity and difference are often crucial attractors of
creative people to high-performing cities.4
But these networks of innovation must also be managed and
shaped to achieve longer-term public benefits. A key feature seems to
be a hybrid, tightloose approach: a tight, potentially even
prescriptive approach to developing the simple rules or common
standards to which all parties will work combined with a much looser
approach to specifying the particular applications that will then
emerge as a result. This reflects the fundamental point that networks
provide a platform for coordinating highly diverse activities, many of
which are not predicted in advance. WiFi, SMS, even the internet
itself all evolved quite differently from how they had been conceived.
Regulation

Successful innovation also depends on effective regulation. Given the

benefits to citizens and consumers in terms of the goods and services
that networks provide, and the potential incentives for actors to
exploit their position within the network to behave in ways that
reduce these benefits, special attention must be paid to the peculiar
regulatory challenges that networks create. In particular, it is often
necessary to separate the regulation of the platform that networks
provide from the actual services which are delivered via that platform.
The difficulties of finding an organisational model for promoting the
maintenance and improvement of the post-privatisation rail network
is one recent example of this. Creating a competitive market in
broadband internet access given the dominance of an incumbent
market player is another.5
Looking to the future, similar problems are looming in relation to
the provision of digital television services,6 and even the hitherto very
lightly regulated internet itself. The recent stock market flotation of
the webs most popular search engine Google, and the possibility that
it could be acquired by a company with a vested interest in influencing its search results, was a timely reminder that in certain respects
searchability of networks through transparent and reliable search
engines is a public good that cannot be left simply to the market. How
Demos

17

Networks collection

4/1/04

4:35 PM

Page 18

Network logic

we approach such regulatory conundrums remains unclear, but it

seems unlikely that the models we have relied on up to now will be up
to the job. The role of regulatory agencies may need to shift from that
of enforcement to brokerage, identifying the organisations and
strategies that could be enrolled in tackling a given problem.7
Accountability

In an interconnected world simple chains of cause and effect are

difficult to establish. Most organisations are nodes in a whole series of
complex networks, some of which they may not even realise they
belong to. The impact of those networks and the outcomes they
produce is rarely the result of any one organisation but rather the
interactions between them. At the same time, membership of such
networks does not diminish the need to give account for ones
actions; on the contrary, it amplifies it. A typical FTSE 100 today is
now required to give account to: shareholders, and particularly the
large institutional investors whose judgements are so crucial in
influencing perception of company performance and management
competence; the media, through which its reputation and trustworthiness will often be mediated; an array of regulatory agencies,
perhaps sector-specific, others concerned with generic issues like
employment practices or health and safety; supranational bodies,
including the European Union; and its wider stakeholders, including
customers, suppliers, employees, NGOs and politicians.
Yet despite this complex picture, we still treat accountability as a
linear process one party being held to account by another. This is
expressed in the way that the Public Accounts Committee of the
House of Commons works, for example, and in the way that the
scrutiny functions of regulatory agencies continue to be understood.
In the future, our conception of accountability seems likely to evolve
away from simple lines of answerability towards something more
complex and messy, with lines of accountability that are:

18

multiple, so that any one actor was accountable to a

number of other actors in a number of different ways

Demos

Networks collection

4/1/04

4:35 PM

Page 19

Introduction

overlapping, so that at different times in different

circumstances one source of accountability might take
priority, but at no point could there be no accountability
at all
based on deliberative as well as procedural processes
generating opportunities for genuine discussion and
learning, rather than fostering defensive mindsets or
going through the motions.

Ownership

The blurring of responsibility and the growth of organisational

interdependence also force us to rethink our binary assumptions
about public versus private ownership. New ways of organising and
providing public goods and services have emerged, which downgrade
the role of the state from that of provider to that of regulator or
coordinator, and seek to draw other actors into complex webs of
provision cutting across traditional institutional boundaries. This
process has been theorised in the rise of the governance paradigm
across the social sciences. The common theme is an interest in
patterns of governing that do not rest on the traditional authority of
the state to coerce and command, and which involve institutions
drawn from within but also beyond government.8
Citizenship

Network citizens understand their connections to the wider world, as

well as to those in their neighbourhood. They are capable of
participating in networked forms of social action, as Howard
Rheingold elaborates, but they are also able to respect the informal
norms of collaboration in everyday life that enhance quality of life in
the terms set out by Robert Sampson. Many of our public goods rely
on the network effects of individuals being prepared to act as coproducers of the service they receive from the state. Networked
citizens will participate in the creation of new decision-making
capabilities as well as understanding their informal power and
responsibilities.
Demos

19

Networks collection

4/1/04

4:35 PM

Page 20

Network logic

Power

The changing nature of power in a network society may be the

hardest nettle to grasp. As Manuel Castells explains in his afterword,
power is as crucial as ever in structuring the contours of the network
society. But power no longer resides in individual institutions (even
states) but in what he calls the switchers through which networks
regulate terms of entry and privilege or exclude particular interests or
positions. These structural conditions help to explain the persistence
of particular kinds of systematic disadvantage even where the wider
environment appears to be in flux. Mark Buchanans cogent analysis
of the network causes of income inequality illuminates this point.
It is interesting that governments have been markedly more willing
to accept the logic of network power in some policy areas than in
others. Faced with the current outcry over the outsourcing of service
employment to call centres in India and elsewhere, for example, a
coherent policy response has yet to emerge. British politicians have
simply reiterated the neoliberal consensus that in a global networked
economy, national policy and employment priorities are inevitably
subservient to investment flows that they have little or no capacity to
control.9 By contrast, on the equally thorny question of how to deal
with substantially higher levels of migration from developing
countries, European governments have desperately sought to shore
up creaking asylum and immigration systems in order to convince
hostile electorates that control is still feasible.
In the face of interdependence, neither defeatism nor control
freakery is acceptable as a strategy for governance. Public intervention
to pool risk, counteract economic insecurity and counteract social
dislocation is becoming more, not less, necessary. The question is
what kinds of public intervention are going to be both legitimate and
effective. The answer, as far as we can see, is to develop institutions
able to channel this interdependence in positive ways rather than be
swamped by it, and which work with the grain of our desire for selfdetermination rather than suppress it.
In other words, if we can only connect the pursuit of freedom to
systems of organisation that will not be undone by its exercise, a
20

Demos

Networks collection

4/1/04

4:35 PM

Page 21

Introduction

networked world can become a more sustainable and a more

enriching place. Making it so requires us to change not just our tools
of intervention, but also our ways of seeing the world. Fortunately,
the connections we have forged make it possible to pursue this task
together.
Notes
1
2
3
4
5
6
7
8
9

S Zuboff and J Maxmin, The Support Economy: why corporations are failing
individuals and the next stage of capitalism (London: Allen Lane, 2003).
R Miller, Towards the learning society? In T Bentley and J Wilsdon (eds) The
Adaptive State: strategies for personalising the public realm (London: Demos,
2003).
I Tuomi, Networks of Innovation: change and meaning in the age of the internet
(Oxford: Oxford University Press, 2002).
R Florida, The Rise of the Creative Class: and how its transforming work, leisure,
community and everyday life (New York: Perseus, 2002).
See J Wilsdon and D Stedman Jones, The Politics of Bandwidth (London:
Demos, 2002).
B Cox, Free for All? Public service television in the digital age (London: Demos,
2004).
J Black, Enrolling actors in regulatory systems; examples from UK financial
services regulation, Public Law, Spring 2003.
G Stoker, Governance as theory: five propositions, International Social Science
Journal, no 155, March 1998.
As Tony Blair put it in December 2003: I feel desperately sorry for anyone
whose job is at risk as a result of this change, but that is the way the world is
today... Blair warns jobs exodus part of new order, Financial Times, 3 Dec.
2003.

Demos

21

Networks collection

4/1/04

4:35 PM

Page 22

Networks collection

4/1/04

4:35 PM

Page 23

Network Logic 2

One of the most

important insights
of the new
understanding of
life is that the
network is a
pattern common
to all life . . .
Living Networks
Fritjof Capra

Networks collection

4/1/04

4:35 PM

Page 24

Networks collection

4/1/04

4:35 PM

Page 25

2. Living networks
Fritjof Capra

In recent years networks have become a major focus of attention in

science, business, and also in society at large and throughout a newly
emerging global culture. Within a few years the internet became a
powerful global network of communications, and many of the new
internet companies act as interfaces between networks of customers
and suppliers. Most large corporations today are organised as
decentralised networks of smaller units, connected to networks of
small- and medium-sized businesses that serve as their subcontractors and suppliers, and similar networks exist among
non-profit and non-governmental organisations. Indeed,
networking has been one of the main activities of political grassroots
organisations for many years. The environmental movement, the
human rights movement, the feminist movement, the peace movement, and many other political and cultural grassroots movements
have organised themselves as networks that transcend national
boundaries.
With the new information and communication technologies,
networks have become one of the most prominent social phenomena
of our time. Sociologist Manuel Castells argues that the information
technology revolution has given rise to a new economy, structured
around flows of information, power and wealth in global financial
networks. Castells also observes that, throughout society, networking
has emerged as a new form of organisation of human activity, and he
Demos

25

Networks collection

4/1/04

4:35 PM

Page 26

Network logic

has coined the term network society to describe and analyse this new
social structure.1
In science, the focus on networks began in the 1920s, when
ecologists viewed ecosystems as communities of organisms, linked
together in network fashion through feeding relations, and used the
concept of food webs to describe these ecological communities. As
the network concept became more and more prominent in ecology,
systemic thinkers began to use network models at all systems levels,
viewing organisms as networks of cells, and cells as networks of
molecules, just as ecosystems are understood as networks of
individual organisms. Correspondingly, the flows of matter and
energy through ecosystems were perceived as the continuation of the
metabolic pathways through organisms.2
The Web of Life is, of course, an ancient idea, which has been used
by poets, philosophers, and mystics throughout the ages to convey
their sense of the interwovenness and interdependence of all
phenomena. In this essay I shall discuss the fundamental role of
networks in the organisation of all living systems, according to
complexity theory and other recent developments in the natural and
social sciences, and I shall analyse the similarities and differences
between biological and social networks in some detail.3
The nature of life
Let us begin with biology and ask: what is the essential nature of life
in the realm of plants, animals, and micro-organisms? To understand
the nature of life, it is not enough to understand DNA, genes,
proteins, and the other molecular structures that are the building
blocks of living organisms, because these structures also exist in dead
organisms, for instance in a dead piece of wood or bone.
The difference between a living organism and a dead organism lies
in the basic process of life in what sages and poets throughout the
ages have called the breath of life. In modern scientific language, this
process of life is called metabolism. It is the ceaseless flow of energy
and matter through a network of chemical reactions, which enables a
living organism to continually generate, repair and perpetuate itself.
26

Demos

Networks collection

4/1/04

4:35 PM

Page 27

Living networks

There are two basic aspects to the understanding of metabolism.

One is the continuous flow of energy and matter. All living systems
need energy and food to sustain themselves; and all living systems
produce waste. But life has evolved in such a way that organisms form
ecological communities, or ecosystems, in which the waste of one
species is food for the next, so that matter cycles continually through
the ecosystems food webs.
The second aspect of metabolism is the network of chemical
reactions that process the food and form the biochemical basis of all
biological structures, functions and behaviour. The emphasis here is
on network. One of the most important insights of the new
understanding of life that is now emerging at the forefront of science
is the recognition that the network is a pattern that is common to all
life. Wherever we see life, we see networks.
Self-generation
It is important to realise that these living networks are not material
structures, like a fishing net or a spiders web. They are functional
networks, networks of relationships between various processes. In a
cell, these processes are chemical reactions between the cells
molecules. In a food web, the processes are processes of feeding, of
organisms eating one another. In both cases the network is a nonmaterial pattern of relationships.
Closer examination of these living networks has shown that their
key characteristic is that they are self-generating. In a cell, for
example, all the biological structures the proteins, enzymes, the
DNA, the cell membrane, and so on are continually produced,
repaired and regenerated by the cellular network.4 Similarly, at the
level of a multicellular organism, the bodily cells are continually
regenerated and recycled by the organisms metabolic network. Living
networks are self-generating. They continually create or recreate
themselves by transforming or replacing their components. In this
way they undergo continual structural changes while preserving their
web-like patterns of organisation.

Demos

27

Networks collection

4/1/04

4:35 PM

Page 28

Network logic

Boundaries of identity
All living organisms have a physical boundary that discriminates
between the system the self , as it were, and its environment. Cells,
for example, are enclosed by membranes and vertebrate animals by
skins. Many cells also have other boundaries besides membranes,
such as rigid cell walls or capsules, but only membranes are a
universal feature of cellular life. Since its beginning, life on Earth has
been associated with water. Bacteria move in water, and the
metabolism inside their membranes takes place in a watery
environment. In such fluid surroundings, a cell could never persist as
a distinct entity without a physical barrier against free diffusion. The
existence of membranes is therefore an essential condition for cellular
life.5
A cell membrane is always active, opening and closing continually,
keeping certain substances out and letting others in. In particular, the
cells metabolic reactions involve a variety of ions, and the membrane,
by being semi-permeable, controls their proportions and keeps them
in balance. Another critical activity of the membrane is to continually
pump out excessive calcium waste, so that the calcium remaining
within the cell is kept at the precise, very low level required for its
metabolic functions. All these activities help to maintain the cellular
network as a distinct entity and protect it from harmful
environmental influences. The boundaries of living networks, then,
are not boundaries of separation but boundaries of identity.
Social networks
The main goal of my research over the past ten years has been to
extend the systemic conception of life to the social domain, and in my
last book, The Hidden Connections, I discuss this extension in terms of
a new conceptual framework that integrates lifes biological, cognitive
and social dimensions.6 My framework rests on the assumption that
there is a fundamental unity to life, that different living systems
exhibit similar patterns of organisation. This assumption is supported
by the observation that evolution has proceeded for billions of years
by using the same patterns again and again. As life evolves, these
28

Demos

Networks collection

4/1/04

4:35 PM

Page 29

Living networks

patterns tend to become more and more elaborate, but they are
always variations on the same basic themes.
The network pattern, in particular, is one of the very basic patterns
of organisation in all living systems. At all levels of life the
components and processes of living systems are interlinked in
network fashion. Extending the systemic conception of life to the
social domain, therefore, means applying our knowledge of lifes basic
patterns and principles of organisation, and specifically our
understanding of living networks, to social reality.
However, while insights into the organisation of biological
networks may help us understand social networks, we should not
expect to transfer our understanding of the networks material
structures from the biological to the social domain. Social networks
are first and foremost networks of communications involving
symbolic language, cultural constraints, relationships of power, and
so on. To understand the structures of such networks we need to use
insights from social theory, philosophy, cognitive science,
anthropology and other disciplines. A unified systemic framework for
the understanding of biological and social phenomena will emerge
only when network theories are combined with insights from these
other fields of study.
Social networks, then, are not networks of chemical reactions, but
networks of communications. Like biological networks, they are selfgenerating, but what they generate is mostly non-material. Each
communication creates thoughts and meaning, which give rise to
further communications, and thus the entire network generates
itself.7
The dimension of meaning is crucial to understand social
networks. Even when they generate material structures such as
material goods, artifacts or works of art these material structures
are very different from the ones produced by biological networks.
They are usually produced for a purpose, according to some design,
and they embody some meaning.
As communications continue in a social network, they form
multiple feedback loops, which eventually produce a shared system of
Demos

29

Networks collection

4/1/04

4:35 PM

Page 30

Network logic

beliefs, explanations and values a common context of meaning, also

known as culture, which is continually sustained by further
communications. Through this culture individuals acquire identities
as members of the social network, and in this way the network
generates its own boundary. It is not a physical boundary but a
boundary of expectations, of confidentiality and loyalty, which is
continually maintained and renegotiated by the network of
communications.
Culture, then, arises from a network of communications among
individuals; and as it emerges, it produces constraints on their
actions. In other words, the rules of behaviour that constrain the
actions of individuals are produced and continually reinforced by
their own network of communications. The social network also
produces a shared body of knowledge including information, ideas
and skills that shapes the cultures distinctive way of life in addition
to its values and beliefs. Moreover, the cultures values and beliefs also
affect its body of knowledge. They are part of the lens through which
we see the world.
Living networks in human organisations
In recent years it has become very popular in management circles to
use metaphors like the living company, trying to understand a
business organisation as a living, self-organising system.8 It is
therefore instructive to apply our network approach to the analysis of
human organisations.
Living social systems, as we have seen, are self-generating networks
of communications. This means that a human organisation will be a
living system only if it is organised as a network or contains smaller
networks within its boundaries, and only if these networks are selfgenerating. Organisational theorists today speak of communities of
practice when they refer to these self-generating social networks.9 In
our daily activities, most of us belong to several communities of
practice at work, in schools, in sports and hobbies, or in civic life.
Some of them may have explicit names and formal structures, others
may be so informal that they are not even identified as communities.
30

Demos

Networks collection

4/1/04

4:35 PM

Page 31

Living networks

Whatever their status, communities of practice are an integral part of

our lives.
As far as human organisations are concerned, we can now see that
they have a dual nature. On the one hand, they are social institutions
designed for specific purposes, such as making money for their
shareholders, or managing the distribution of political power. On the
other hand, organisations are communities of people who interact
with one another to build relationships, help each other, and make
their daily activities meaningful at a personal level.
This dual nature as legal and economic entities as well as
communities of people derives from the fact that various
communities of practice invariably arise and develop within the
organisations formal structures. These are informal networks
alliances and friendships, informal channels of communication, and
other webs of relationships which continually grow, change and
adapt to new situations.
Within every organisation there is a cluster of interconnected
communities of practice. The more people are engaged in these
informal networks, and the more developed and sophisticated the
networks are, the better will the organisation be able to learn, respond
creatively to new circumstances, change and evolve. In other words,
the organisations aliveness resides in its communities of practice.
In order to maximise a companys creative potential and learning
capabilities it is crucial for managers and business leaders to
understand the interplay between the organisations formal, designed
structures and its informal, self-generating networks.10 The formal
structures are sets of rules and regulations that define relationships
between people and tasks, and determine the distribution of power.
Boundaries are established by contractual agreements that delineate
well-defined subsystems (departments) and functions. The formal
structures are depicted in the organisations official documents the
organisational charts, bylaws, manuals and budgets that describe the
organisations formal policies, strategies and procedures.
The informal structures, by contrast, are fluid and fluctuating
networks of communications. These communications include nonDemos

31

Networks collection

4/1/04

4:35 PM

Page 32

Network logic

verbal forms of mutual engagement in joint enterprises, informal

exchanges of skills, and the sharing of tacit knowledge. These practices
create flexible boundaries of meaning that are often unspoken.
In every organisation there is a continuous interplay between its
informal networks and its formal structures. The formal policies and
procedures are always filtered and modified by the informal
networks, which allows them to use their creativity when faced with
unexpected and novel situations. Ideally, the formal organisation will
recognise and support its informal networks of relationships and will
incorporate their innovations into the organisations formal
structures.
Biological and social networks
Let us now juxtapose biological and social networks and highlight
some of their similarities and differences. Biological systems exchange
molecules in networks of chemical reactions; social systems exchange
information and ideas in networks of communications. Thus,
biological networks operate in the realm of matter, whereas social
networks operate in the realm of meaning.
Both types of networks produce material structures. The metabolic
network of a cell, for example, produces the cells structural
components, and it also generates molecules that are exchanged
between the networks nodes as carriers of energy or information, or
as catalysts of metabolic processes. Social networks, too, generate
material structures buildings, roads, technologies and so on that
become structural components of the network; and they also produce
material goods and artefacts that are exchanged between the
networks nodes.
In addition, social systems produce non-material structures. Their
processes of communication generate shared rules of behaviour, as
well as a shared body of knowledge. The rules of behaviour, whether
formal or informal, are known as social structures and are the main
focus of social science. The ideas, values, beliefs and other forms of
knowledge generated by social systems constitute structures of
meaning, which we may call semantic structures.
32

Demos

Networks collection

4/1/04

4:35 PM

Page 33

Living networks

In modern societies, the cultures semantic structures are

documented that is, materially embodied in written and digital
texts. They are also embodied in artefacts, works of art and other
material structures, as they are in traditional non-literate cultures.
Indeed, the activities of individuals in social networks specifically
include the organised production of material goods. All these
material structures texts, works of art, technologies and material
goods are created for a purpose and according to some design. They
are embodiments of the shared meaning generated by the societys
networks of communications.
Finally, biological and social systems both generate their own
boundaries. A cell, for example, produces and sustains a membrane,
which imposes constraints on the chemistry that takes place inside it.
A social network, or community, produces and sustains a nonmaterial, cultural boundary, which imposes constraints on the
behaviour of its members.
Conclusion
The extension of the systemic conception of life to the social domain
discussed in this essay explicitly includes the material world. For
social scientists, this may be unusual, because traditionally the social
sciences have not been very interested in the world of matter. Our
academic disciplines have been organised in such a way that the
natural sciences deal with material structures while the social sciences
deal with social structures, which are understood to be, essentially,
rules of behaviour.
In the future, this strict division will no longer be possible, because
the key challenge of our new century for social scientists, natural
scientists and everyone else will be to build ecologically sustainable
communities.11 A sustainable community is designed in such a way
that its technologies and social institutions its material and social
structures do not interfere with natures inherent ability to sustain
life. In other words, the design principles of our future social
institutions must be consistent with the principles of organisation
that nature has evolved to sustain the web of life. A unified conceptual
Demos

33

Networks collection

4/1/04

4:35 PM

Page 34

Network logic

framework for the understanding of material and social structures,

such as the one offered in this essay, will be essential for this task.
Fritjof Capra, PhD, physicist and systems theorist, is a founding director
of the Center for Ecoliteracy in Berkeley. He is the author of several
international bestsellers, including The Tao of Physics and The Web of
Life. This essay is based on his most recent book The Hidden
Connections. www.fritjofcapra.net
Notes
1
2
3

10
11

M Castells, The Rise of the Network Society (Oxford: Blackwell, 1996).

F Capra, The Web of Life (London: HarperCollins, 1996).
For a more extensive analysis, see F Capra, The Hidden Connections (London:
HarperCollins, 2002).
See H Maturana and F Varela, The Tree of Knowledge (Boston: Shambhala,
1987) and Capra, Web of Life.
Capra, Hidden Connections.
Ibid.
See N Luhmann, The autopoiesis of social systems in N Luhmann (ed.) Essays
on Self-Reference (New York: Columbia University Press, 1990) and Capra,
Hidden Connections.
See, for example, A de Geus, The Living Company (Boston: Harvard Business
School Press, 1997).
See E Wenger, Communities of Practice (Cambridge: Cambridge University
Press, 1998).
Capra, Hidden Connections.
Ibid.

34

Demos

4
5
6
7
8
9

Networks collection

4/1/04

4:35 PM

Page 35

Network Logic 3

Reciprocity is key
to the power of
networks, the
alchemy of mutual
give and take over
time turning to a
golden trust . . .
Towards a theory of
government
Karen Stephenson

Networks collection

4/1/04

4:35 PM

Page 36

Networks collection

4/1/04

4:35 PM

Page 37

3. Towards a theory of
government
Karen Stephenson

Governments dont move; they morph. Built on the skeletal remains

of past policies they grow incrementally like a coral reef, changing the
ecosystem around them. And, like coral reefs, they are vast structures,
difficult to chart thoroughly. But governments are not bureaucracies,
no matter how much they may look and feel like a barrier reef at
times. So what are they?
Commentators on both sides of the Atlantic have long lamented
the existence of big government. But size isnt everything. Sure, scale
still matters: government sprawl going from 1,000 to 50,000 to
200,000 employees can and does make a difference. But what if
there was an iterative pattern to the structuring of increasing scale?
Thats exactly the reason the French anthropologist Levi-Strauss once
claimed to have found an atom of kinship, an elementary structure
common to every biological or fictive family.1 As it turned out, his
idea was more provocative than practical: there was no such reliably
repeating structure. But what if there was an atom of organisation, a
recurring structure of how people organise? And, if so, what
implication does it have for leadership and governance?
Markets, hierarchies and networks
Lets try to answer these questions by starting with what we know.
Assume a hierarchy. It was long thought that hierarchy was an island
of planned coordination in a sea of market relationships, a pristine
Demos

37

Networks collection

4/1/04

4:35 PM

Page 38

Network logic

paradise inhabited by vertically integrated tribes of employees.2 But

corporate anthropologists debunked the myth of the CEO, the savage
noble burned at the stake in a bonfire of vanities, just as British
anthropologists overturned the myth of the noble savage a century
before. Why? Because whether the jungles are green and leafy or
concrete, they are brimming with intricate webs of relationships,
which when viewed from afar reveal elementary structures.
Initially, these elementary structures were sorted into a triptych of
organisational forms: markets, hierarchies and networks. Standing at
one end, a chorus of economists droned market as their mantra the
genesis of organisational life. On the other end perched hierarchy, the
logical evolutionary endpoint, the height of civilised achievement.
Networks were largely ignored or dismissed as a mixed breed, a
doomed hybrid nesting somewhere between the two.
But the resulting theoretical continuum of market network
hierarchy was misguided, as was made profoundly apparent on
September 11, 2001. Networks would prove to be core to the
continuum, and hierarchy the hybridised half-breed. Networks were
and are deeply linked to cultural genesis and genocide, and comprise
the core DNA within any governance structure.
Transaction costs and organisational form
To understand how the old continuum, though mistaken, became
accepted, we start with Ronald Coases classic paper on the nature of
the firm.3 Coase suggested that firms and markets, while different
organisational structures, nevertheless share common transactional
practices. The distinction, later amplified by Williamson,4 was based
on the amount of knowledge about a transaction (or asset
specificity) that was required. Disinterested, non-repetitive
exchanges occurred as market transactions (simple contracts). But
exchanges that entailed greater uncertainty, and therefore a
proportional amount of asset specificity, were best sheltered within
the firm as a way to mitigate the greater risk.
Once these transactions were harboured in the firm, inefficiencies
devolved. Williamson argued that these inefficiencies were tolerated,
38

Demos

Networks collection

4/1/04

4:35 PM

Page 39

Towards a theory of government

in fact preferred, because of the firms own bounded rationality

better the devil you know than the devil you dont and theories of
firm and hierarchy were conceived.5 In refining these propositions
and terms a continuum of transactions emerged, beginning with the
disinterested and discrete market transactions on one end and the
interested, asset-specific transactions of a firms hierarchy on the other.
Powell suggested that networks deserved a place on the
continuum, for they permitted exchanges that required asset
specificity but at the same time could move through semi-permeable
organisational boundaries typical of family businesses, guilds and
cartels.6 Networks collide and collude and are able to elude
organisational boundary checkpoints nimbly because of the trust
and mutual reciprocity between the actors in essence invisibly
storing the asset-specific knowledge of the transaction in the trust
relationship itself.7 Networks produce nuanced asynchronous and
asymmetric exchanges, deftly avoiding both the visible hand of
hierarchy8 and the invisible hand of the market.9
Sounds right, feels right but its wrong. Lets rethink the
continuum in light of the logic of exchange and put the argument
right. At one extreme are disinterested, non-repetitive exchanges
typically found in markets. The logical inversion would be repetitive
exchanges of mutual interest evidenced in networks. Hierarchy,
comprised of routine exchanges (repetitive like a network) with a
governing authority (more in keeping with the contractual
characteristics of market exchanges), is now squarely in the middle,
having qualities of both network and market. Table 1 summarises
these organisational forms and features in their logical order.
Ramifications for organisations
What does this mean for organisations? In markets, its a free-for-all
a knowledge diaspora, information spreading virally to where it is
most needed regardless of legal (or moral) boundaries. But in
organisations, the interests are best if not self served when
knowledge is more closely held because of the routine and repetitive
exchanges that result from established procedures and the knowledge
Demos

39

Networks collection

4/1/04

4:35 PM

Page 40

Network logic

Table 1 Organisational forms and their associated

exchange rates
Organisational form

ABCs of exchange

Market
Hierarchy
Network

Disinterested, non-repetitive
Routinised by a governing authority
Mutually interested, repetitive

shared among those in authority. Therefore, organisations and their

governance are directly shaped by only hierarchies and networks. In
broad strokes, these organisational structures can be distinguished by
a very simple rule of thumb. First, in a hierarchy one person can
perform the work of one: work is directly related to the job
assignment one for one and one for all; second, in a network shared
collective intelligence is exponential: two people can perform the
work of four. This multiplier effect is a result of the leveraging of
individual efforts through bonds of mutual trust and reciprocity, as
evidenced by sturdy hunter-gatherers who daily survived
overwhelming ecological odds through cooperation. Small groups
everywhere share this ancient and larger-than-life capability with
their Kalahari counterparts.
Reciprocity is key to the power of networks, exerting a governing
logic over them the alchemy of mutual give and take over time
turning to a golden trust. Primordially, trust was determined through
face-to-face interactions, and to a large extent is still today. Therefore
one needs to appreciate the profound truth that the face of trust is
still a human face, a face that can mask a fundamental fear of
differences. The stark truth about trust is that if you dont look like
me, or dress like me, walk or talk like I do, then I am not likely to
know or understand you. This fetish for the familiar is fundamentally
tribal and resistant to diversification.10 A network of trust is the real
invisible hand behind every act of deceit, fraud and betrayal. Trust is
the caress of forgiveness and friendship, the unshakeable grip of
family or tribe.
40

Demos

Networks collection

4/1/04

4:35 PM

Page 41

Towards a theory of government

What happens when networks and hierarchy are misaligned? A

bureaucracy is the logical but unfortunate result. Bureaucracies write
over their past wrongs, exacting obedience by methodically numbing
collective consciousness through the destruction or stoppage of
careers. In fully mature bureaucracies, two people perform the work
of one-half. This is the real truth behind budgetary shortfalls. So
whether you are a hunter-gatherer or hunting and pecking your way
through a bureaucracy, you are operating as you would in a network,
not a market or hierarchy. Transactors and transactions change all the
time in markets and hierarchies in markets because the transactions
are disinterested and non-repetitive, and in hierarchies because they
are routinised regardless of who performs them. Although hierarchies
are lauded, weve all heard the laughable refrain, Ive worked for five
heads of department! Staying power is found not in whos at the top
but in whos in the network.
Ramifications for government
Human networks have evolved from bands, tribes, segmentary
lineages and chiefdoms right up to the modern state. Government
operations betray the signs of this evolutionary heritage, strikingly
similar to a segmentary lineage system.11 When policies change or
new needs arise, teams are created not from the ground up, but as
sub-units of existing segments. As layers of hierarchy proliferate, units
at each layer compete against one another, combining to work as a
larger unit only when these too are drawn into competition. So
within a government department one team jockeys for position with
another, one directorate attacks another to protect its budget, and
the department as a whole fights other departments to defend its
turf. In these systems there is no internal structure or infrastructure
to join the system as a whole; it is simply a network of hierarchies
(vertically integrated silos). As such, they are never more than (and
are often less than) the sum of their parts. Segmentary systems
calculate power by comparing and contrasting their stock or status
with that of other segments. Competition, not collaboration, is the
watchword.
Demos

41

Networks collection

4/1/04

4:35 PM

Page 42

Network logic

So it comes as no surprise when people in the public sector

complain of being stuck in organisational silos, obstacles to be
overcome by joined-up working. What we dont realise is that we are
hard-wired to create these silos because of the constraints of
segmentary systems. Figure 1 shows what a segmentary system looks
like depicted graphically. It is the image of an organisational analysis
that was conducted of the US government, three hierarchical layers
down from the president. Within each organisational box or segment
you can see smaller subgroups denoted by densely connected circles.
The perimeters of the circles are made up of microscopic dots
denoting real people and the criss-crossing lines that fill each circle
represent the reciprocal communications between the individuals in
each box. It is obvious that the relative density within the boxes
eclipses the connections among the boxes, indicating that people
Figure1 Segmentary lineage within the US government

(Courtesy of NetForm International, Inc.)

42

Demos

Networks collection

4/1/04

4:35 PM

Page 43

Toward the theory of government

spend their limited resources investing in segmentary or silo-ed

relationships within each box or segment. The unhappy result was
that segmentary politics consisted largely of errors and their
associated rework, achieved at the peril of overall organisational
effectiveness.
It is easy to see from this example how segmentary politics can
easily and perversely thwart overall objectives and goals. Real
problems do not come neatly packaged. Whether in fighting
terrorism or tackling corporate fraud, cross-cutting problems require
cross-cutting solutions. Yet in segmentary systems of government
there is no process in place to allow the seamless exchange of vital
information laterally among the different units with an interest in
tackling the problem with the knowledge or resources to do so. The
result is that information disappears, deadlines are missed, fingers
point and wrongs are papered over. The knowledge of how to put it
right to put a process in place is in the networks of the people
doing the work.
That is why it is important to understand the differences between
how people actually work in their networks and how the process says
they should work. When it was first elected, the Blair administration
made much of the need for joined-up working. This was exemplified
by the agglomeration of several smaller departments and budgets into
the giant Department for the Environment, Transport and the
Regions (DETR). The failure and subsequent break-up of the DETR
showed that putting people in the same building or changing the
name on the signs does not automatically lead to integrated working.
But this misalignment between the network of work and the formal
processes is a continuous feature of every organisation, not least
because market conditions are always changing and organisations are
always adapting to keep up.
This phenomenon is exacerbated by the growing recognition
within government of its limited capacity to tackle policy problems
autonomously, and a growing desire instead to build partnerships of
public, private and voluntary sector actors better equipped to address
them. Are there any examples where governments are transforming
Demos

43

Networks collection

4/1/04

4:35 PM

Page 44

Network logic

the way they partner with their stakeholders? Is there an established

way to benchmark these kinds of publicprivate collaborations?
A step in the right direction
One exciting project involves a collaboration between HM Customs
and Excise, the Government Office for the East Midlands and the East
Midlands Development Agency. The project is using a new technique
known as virtual network analysis (VNA) to try to improve
partnership working.12 The approach combines process mapping and
modelling with what is known as social network analysis, or people
mapping. In human networks the assets are tacit, intangible and
intellectual. VNA maps and measures these intangible assets in order
to make them tangible and the effective management of them more
possible.
Three trials were commissioned by the Director of Crime
Reduction in the East Midlands, addressing:

domestic violence in Northampton

burglary in Nottingham
sport partnerships in Nottinghamshire.

The VNA trials are currently in various stages of completion. Already

the findings are positive. Blockages have been identified and efficient
integrated working solutions are being implemented. Some
improvements have been immediate; others take more time. In the
domestic violence trial in Northampton, for instance, the study
revealed that four different agencies all carrying out multi-agency
training to different standards were unaware of each others work in
this area. When these shared practices were identified, knowledge was
much more effectively leveraged through group learning. In the
burglary trial in Nottingham it was discovered that one crime-solving
division was holding crucial knowledge close to its chest to the
detriment of other parts of the organisation. Systems for effective
knowledge transfer and reward mechanisms for sharing information
were explored.
44

Demos

Networks collection

4/1/04

4:35 PM

Page 45

Towards a theory of government

Looking to the future

We know from the science of networks that there is hard-wiring
embedded in the apparent softness of social capital. Indeed, social
capital has been the missing link in a performance equation that only
valued human capital as measured hierarchically. The very nature of
traditional performance measures is silo-esque people are measured
by those to whom they report. Now, by combining social and human
capital measures, the intellectual assets of the organisation can be
asserted, assessed and quantified. People can be measured by whether
they run to, not away from, problems. This recognition carries with it
the potential, at the very least, to neutralise the myopic exclusivity of
the silo mentality. At best, it could hold the key to neutralising the
destructive networks of international crime, fraud and terrorism.
VNA is an important approach in facilitating more effective
cooperation between segments, but more work lies ahead. In
particular, while processes may be mapped and aligned, who is
looking after the leaders? Where does the leadership reside and what
does it look like? Traditional leadership theory focuses on the typical
one leader approach and falls short of the mark. Within a
segmentary system we must rethink leadership as a network of
segment leaders, cooperating and coordinating at improved
efficiencies for the greater good of the total enterprise. This is true
infrastructure. Figure 2, developed for the VNA trials, tries to capture
this concept. In this VNA world, leaders at some time or to some
degree are followers.13 At any one time leaders may be leading:

laterally for external partnerships

laterally for internal partnerships
downwardly for vertically integrated hierarchies and
intermittently according to the exigencies of a decision
process.14

The real lesson here is that the recombinant properties of a leadership

network are just as important to model and understand as the
processes they govern.
Demos

45

Networks collection

4/1/04

4:35 PM

Page 46

Network logic

Figure 2 Model for government

(Courtesy of Enterprise Modeller Solutions Ltd)

The key point is this: just because you have an organisational chart
at your fingertips does not mean you have a charter for your world.
And having the networks mapped does not tell you about the cultural
terrain you have to cross in order to lead effectively; the map is most
certainly not the territory. Rather it is the lack of a coordinated
leadership network within a network of hierarchies that produces the
lurches, lunging and sputtering we frequently experience in
government. What I have tried to do here is explain the science
underlying the practical vagaries, which is essential for planning and
predicting effective change.
With e-government and virtual operations increasing, it becomes
even more salient to have government segments coordinated for rapid
learning and operational adaptation. Its difficult to resign ourselves
to living in a world where even the best-intentioned can undermine
or devastate operations by a simple flip of a switch. But we do live in
such a world and it is filled with this kind of risk. Its up to us to
46

Demos

Networks collection

4/5/04

2:35 PM

Page 47

Towards a theory of government

engender the kind of trust and build the kind of government that will
make it a better and safer place.
Karen Stephenson is professor of management at the Harvard Graduate
School of Design and President of Netform, Inc. Her book The Quantum
Theory of Trust will be published by Financial Times Prentice Hall later
this year.
Notes
1
2
3
4

6
7
8
9
10
11

12

C Levi-Strauss, The mathematics of man, International Social Science Bulletin,

no 6, 1955 and The Elementary Structures of Kinship, tr. J Harle Bell, J R von
Sturmer and R Needham (Boston: Beacon Press, 1971)
Adapted from G B Richardson, The organization of industry, Economic
Journal, no 82, 1972.
R Coase, The nature of the firm, Economica, no 4, 1937.
O Williamson, Markets and Hierarchies (New York: Free Press, 1975), The
Economic Institutions of Capitalism (New York: Free Press, 1984) and
Transaction cost economics and organization theory, Industrial and Corporate
Change, no 2, 1993.
Management theorists put their imprimatur on the debate with derivative
theories based on hybrid organisational forms and managerial approaches,
such as the U and M organisation and Theories X and Y (D McGregor, The
human side of enterprise, Management Review, Nov 1957) and Z (W Ouchi,
Theory Z (New York: Addison-Wesley Publishing, 1981)) respectively.
W Powell, Neither market nor hierarchy: network forms of organization,
Organizational Behavior, no 12, 1990.
M Granovetter, Economic action and social structure: a theory of
embeddedness, American Journal of Sociology, no 91, 1985.
A Chandler, The Visible Hand (Cambridge, MA: Harvard University Press,
1977).
A Smith, The Wealth of Nations (Oxford: Clarendon Press, 1976).
K Stephenson, Networks, in CRC The Technology Management Handbook, ed.
R Dorf (Boca Raton: CRC Press, 1999).
M Sahlins,The segmentary lineage: an organization for predatory expansion,
American Anthropologist, no 63, 1961. In the early stages of human
organisation leaders were typically charismatic and built a following by
creating loyalties using generosity or fearful acquiescence through magic,
demonstrated wisdom, oratorical skill, and so on. A change occurred between
the segmentary lineage and chiefdom, where leadership was no longer carried
out by a charismatic leader but by an instituted office authorised by God, coup
or chad.
For more information see www.benchmarking.co.uk or contact Jeremy
Hawkins at the East Midlands Crime Reduction Unit.

Demos

47

Networks collection

4/1/04

4:35 PM

Page 48

Network logic

13
14

48

M Sahlins, Poor man, rich man, big man, chief: political types in Melanesia and
Polynesia, Comparative Studies in Society and History, no 5, 1963.
Detailed in M Kriger and L Barnes, Executive leadership networks: top
management group dynamics in a high-performance organization, in The
SYMLOG Practitioner: Applications of small group research, eds R Polley, A Hare
and P Stone (New York: Praeger, 1988).

Demos

Networks collection

4/1/04

4:35 PM

Page 49

Network Logic 4

Seeing the
connectedness of
things is the
starting point for
understanding a
world that
otherwise appears
baffling . . .
Connexity revisited
Geoff Mulgan

Networks collection

4/1/04

4:35 PM

Page 50

Networks collection

4/1/04

4:35 PM

Page 51

4. Connexity revisited
Geoff Mulgan

Connexity was published in 1997 in the UK and has subsequently

been published (and translated) in various parts of the world, from
the US and Greece to China.1 The book made three arguments.
First, that the growing connectedness of the world is the most
important social and economic fact of our times. It is manifest in the
growth of physical links like telecom networks; in rising flows of
goods, money, ideas and people; in the interconnectedness of culture
and the environment; and in new forms of social organisation. Hence
the reuse of an old English word, connexity.
Second, that this connectedness renders redundant many of the
dominant concepts of political, social and economic thought that
assumed a world of sovereign and separate entities: nation states,
companies and individual citizens. The key intellectual methods
needed to understand the contemporary world, by contrast, focus on
the character of systems, connections and feedback loops, and on
subjects of action who are not seen as complete and bounded in the
manner of nineteenth-century liberalism or twentieth-century
economics.
Third, that this connectedness poses major moral challenges, above
all concerning our responsibilities to others. For example, how should
we rethink our consumption patterns in the light of climate change?
How should we and our governments respond to genocide and failing
states in distant countries? I argued that these issues were set to rise
Demos

51

Networks collection

4/1/04

4:35 PM

Page 52

Network logic

up the political agenda and would require radical changes to our

institutions.
Since I wrote the book all of these trends have intensified. The
world has continued becoming more connected; the boundaries
between international and domestic policy have blurred irrevocably;
and the gap between the scale and nature of the problems and the
capacities of institutions to deal with them has grown. Within
business, governments and civil society the importance of networks
in helping with coordination, learning and the creation of value has
become more commonplace.
Networks, hierarchies and control
Connexity built on the more theoretical account set out in an earlier
book, Communication and Control.2 This argued that the information
revolution was best understood as an increase in capacities to control,
and that these would be exploited both by new horizontal networks
and by traditional hierarchies.
The empowerment of networks is becoming widely understood in
everything from knowledge management to protest. Oddly, it is the
empowerment of hierarchies that is now less well recognised.
For many the promise of the information revolution was that
power would be distributed evenly. The network would displace the
hierarchy, as in Tom Paines marvellous description of democracy
shattering the crown of royal dominion and giving each citizen a
jewelled piece. A networked world would empower citizens against
states, consumers against companies, the weak against the strong.
Networks can, indeed, be genuinely empowering. But some of the
characteristics of knowledge its zero marginal cost, replicability and
fungibility and some of the characteristics of networks their
reach, and exponentially rising value have led to greater not lesser
concentrations of power and have reinforced some hierarchies. The
key characteristic of hierarchies is concentration: concentration of
resources at the points where it can make most impact, and
concentration of control over resources that others need: money,
knowledge, votes, even processing power.
52

Demos

Networks collection

4/1/04

4:35 PM

Page 53

Connexity revisited

We still live in a world of powerful hierarchies. Governments take a

larger not smaller share of GDP than they did 10 or 20 years ago. The
military depend on tighter command systems than ever before to
avoid mistakes, since the response of a warship to air attack now has
global ramifications. Power in the global media, and power over the
memes that shape minds, is more concentrated than ever despite
the proliferation of magazines and websites. For all the talk of the
network economy most businesses are organised as fairly tight
hierarchies, albeit with fewer layers, and some that used not to be, like
partnerships, are taking more traditional corporate forms. Again, one
of the drivers of this is globalisation, since what a subsidiary does in a
distant country using child labour, say may have an impact on
consumers here. The same is true in NGOs. Look closely at
Greenpeace, for example, and you see a fairly tightly controlled
hierarchy, not loose democracies. Within and around all of these are
networks: networks for managing relationships, knowledge and
information. But at their heart lie hierarchical organisations of power
and authority able to act decisively and quickly, with concentrations
of resources and with some of the properties of Benthams
panopticon, able to see everything from the centre in real time.
So although networks have become much more important to the
way we live, we do not live in a world dominated by networks.
Networks are extraordinary ways of organising knowledge,
cooperation and exchange. They are far more effective means of
sharing learning than hierarchies and generally better at adapting to
change. But they remain poor at mobilising resources, sustaining
themselves through hard times, generating surpluses, organising
commitments, or playing games of power. This is why, for example,
the interesting feature of the anti-globalisation movement is its
weakness not its strength, and why Al-Qaeda can inflict huge damage
but cannot create.
Risk and the state
The continuing importance of hierarchy in a networked world partly
reflects the nature of risk. States have their origins as protections
Demos

53

Networks collection

4/1/04

4:35 PM

Page 54

Network logic

against risk: protection from invasion and attack, then in the

nineteenth century from disease, and later still protection from
poverty and unemployment. Much of what we count as progress in
urbanised societies has been the increasing success of many states in
protecting their citizens from war, disease and poverty.
Some expected a networked world based on information to reverse
this, or at least to offer a different way of handling risk. It is certainly
harder for states to build certain kinds of walls or control what
information and knowledge reaches their citizens (though the Taliban
tried). The other widespread prediction was that a more networked
world would inevitably encourage greater individualisation of risk,
which might leave each person, or at least everyone with the means to
do so, with their own insurance, purchased on global markets,
calibrated using sophisticated algorithms, particularly with new
genetic knowledge. People would buy their own security, education,
or healthcare, as the state, and pooled risk, withered away.
The reality is very different. We remain in a world where many
risks are collective and where the public rationally looks to public
institutions to protect them. These include the risks of instability in
the global market, the risks of attacks on systems on which we depend
(from food distribution to electricity), risks to personal privacy or
global warming (in each of which states can be as much the problem
as the solution). Paradoxically the very speed with which networks
have advanced has reinforced the need for stronger legitimate
authorities to protect people from the risks they have brought in their
wake.
The nature of risk also underscores the importance of moral
obligations to others. September 11 was a reminder that even the
richest and most powerful remain vulnerable. Thomas Hobbess
words from over 300 years ago have special resonance now.
In Leviathan he wrote that the weakest has strength enough to kill
the strongest, either by secret machination or by confederacy with
others , which is why the strongest need to care about the lives and
needs of the weakest.

54

Demos

Networks collection

4/1/04

4:35 PM

Page 55

Connexity revisited

The implications of connexity

What follows for governments, whether of cities or states? In
Connexity I set out some of the emerging principles for government
from the relationship with citizens and the role of trust to the use of
more horizontal structures in bureaucracies. Five aspects stand out.
Transparency

The first is the potential for radically greater transparency because of

the ease and cheapness of sharing information. On the Web I can look
up my street on a site called upmystreet.com. Through it I can
instantly find out information not just about local contacts or
discussion groups but also about the relative performance of my local
schools or local police force. I can do this because government has
been through a revolution in how information is organised. Much
that used to be internal, the prerogative of management, is now
external. Information has been turned inside out, rather like a
Richard Rogers building.
This shift was bitterly resisted by the professions and many of the
experts in each field, who feared that the information was crude and
that the public wouldnt be able to make sense of it. Yet having
happened it is irreversible, and has turned out to be a powerful force
for changing cultures of provision, driving up performance and
encouraging weak performers to learn from their more successful
counterparts.
Transparency has also changed the working practices of government. A good example is food, where public trust in government
collapsed in the UK in the 1990s because of BSE. Now the Food
Standards Agency makes all its decisions in public, publishes all the
scientific and other data underpinning those decisions, and has
quickly shown that a more open, adult approach increases public
trust. My guess is that future governments will simply take that sort of
greater transparency for granted, as an unavoidable aspect of the
environment in which they have to work.

Demos

55

Networks collection

4/1/04

4:35 PM

Page 56

Network logic

Holism

The second principle is holism: the potential for governments to see

how things connect. Systems thinking, and the possibilities of
networks, are together pushing towards what we call holistic, joinedup government reversing the logic of nineteenth-century
functionalism, and the logic of the new public management which
divided every task into component parts, separate functions or
markets. Instead, smart networks make it possible to organise the
forms of government in very different ways, starting with needs,
outcomes, or client groups and then working backwards to fit
functions to them.
In welfare and justice, for example, the case management approach
has been widely adopted, so that if a 16 year old gets into trouble the
different agencies police, social services and voluntary organisations
are brought together by a single person to fit the young persons
individual needs. On the Web, government services have been
clustered by life experience such as having a baby or retiring. At a
national level some budgets have been moved away from departments
to cross-cutting structures like Surestart for under-3s or the fund
for conflict prevention in sub-Saharan Africa, so that the money
follows the problem rather than being the preserve of departments,
agencies and professions. Old-fashioned vertical hierarchies still have
their place. But over time the likelihood is that horizontal, networked
structures will increasingly displace them.
Directness

The third principle is directness. In the past governments had to

operate through many layers of mediation, tiers and hierarchical
bureaucracies. Now more direct relationships are feasible. Take
health. A big advance in the UK has been NHS Direct, a 24/7 phone
and online service of advice and diagnosis, which is in part about
empowering individuals to take more command of their own health.
If my daughter is sick at 3am I can speak to a nurse or look up
guidance. The idea is not all that radical in itself. But it brings
together old elements (phones and nurses) in a new way, just as
56

Demos

Networks collection

4/1/04

4:35 PM

Page 57

Connexity revisited

printing, for example, was not a new invention but rather a

combination of raised letters from coins, the wine press and etching.
Directness applies in other fields too. Learndirect and the National
Grid for Learning now provide a full range of courses accessible
online, with back-up from tutors both phone-based and face to face.
Such approaches have the potential over time to transform education
into a much more user-driven system. In the field of democracy there
has been constant experimentation to increase direct engagement. In
the UK one-fifth of local councils have used citizens juries. There
have been more referendums in the last seven years than in the last
70. Power has been devolved to Scotland, Wales and Northern
Ireland, which now have their own very self-contained political lives.
Postal voting has doubled turnouts in some areas and very local polls
for neighbourhood committees with control over large budgets have
achieved high turnouts.
These changes in method do not solve the underlying problems of
democracy. As ever, the scarce resource is time and with it
understanding. But they are steps towards more direct engagement by
the public in the decisions made on their behalf.
Multiple levels

The fourth principle is the potential to see every issue, every task
through many frames from the neighbourhood to the global.
Governance now takes place at multiple levels local, regional,
national, European, global between which there are few clear
boundaries. Local phenomena, like asylum or drugs, or the pressures
on the education system, cannot be understood or addressed in
isolation from global events. Policy increasingly straddles old divides.
The ugly word intermestic describes how issues like energy security,
cybercrime or migration cut across older definitions of foreign
affairs.
At the same time one of the great achievements of the information
society has been the rapid emergence of something akin to a global
demos. Global public opinion has made itself felt around Kyoto,
Afghanistan, debt cancellation and Africa. It makes its strength felt in
Demos

57

Networks collection

4/1/04

4:35 PM

Page 58

Network logic

the actions to tackle conflict, in humanitarian responses to disaster, in

the outlawing of genocide, and in the way that some of the memes of
democracy, social justice and human rights have spread faster than
trade or foreign direct investment in the last decade.
Yet the global demos poses acute problems. No one has designed
even in theory a way for people to sense sovereignty in exercising
power along with 6 billion others. Some bodies operate on the
principle of one nation one vote, some one dollar one vote, and some
are effectively weighted by military power while oddly no
international institution works on the principle of one citizen one
vote. In a world of imbalanced power this may be inevitable, but it is
in tension with the promise of distributed control.
I doubt that we will see the emergence of nation-state type forms
at the global level: parliaments, armies, currencies. More likely is a
much more complex system of parallel agencies, variable geometry
alliances and difficult models of accountability, in all of which the
informal aspects are as important as the formal ones. A good example
of this is the online network of legislatures being promoted by Earth
Action, bringing together 25,000 legislators representing 60 per cent
of the worlds population with the potential to organise discussion
groups over everything from stem cell regulation to terrorism. This
will have no formal power but potentially could have a great deal of
influence.
Leanness

The fifth principle is leanness. The first wave of productivity gains

from information technology were exploited in manufacturing. Only
belatedly did private services experience major gains. Now public
sectors stand to gain even more than the private sector because so
much of what they do involves the collection, processing and
dissemination of information and knowledge; these lie at the very
heart of government. Yet the realisation of these gains depends on
radical reform of processes, structures and rules.

58

Demos

Networks collection

4/5/04

2:29 PM

Page 59

Connexity revisited

Conclusion
Seeing the connectedness of things is the starting point for
understanding a world that otherwise appears baffling. Economics,
environment and security do not exist in neatly demarcated boxes.
Nor do nations, companies or even families. Yet it is far easier to
assume a world without connections, a world of fewer dimensions
where simpler heuristics carry us through. This is perhaps the hardest
aspect of a connected world and the reason why our concepts and
institutions may be doomed to lag behind the reality they seek to
make sense of.
Geoff Mulgan is Head of Policy in the Office of Prime Minister Tony
Blair, but writes here in a personal capacity.
Notes
1
2

G Mulgan, Connexity: responsibility, freedom, business and power in the new

century (London: Vintage, 1998).
G Mulgan, Communication and Control: networks and the new economies of
communication (Cambridge: Polity Press, 1991).

Demos

59

Networks collection

4/1/04

4:35 PM

Page 60

Networks collection

4/1/04

4:35 PM

Page 61

Editors note
A significant feature of the changing landscape of educational reform in
the UK and beyond is the increasing emphasis on collaboration and
networking. This shift challenges the dominance of the standards agenda
and the competitive ethic that has defined educational policy since the
1980s. Networking between schools is increasingly recognised as a key
driver of school improvement in so far as it encourages professional
collaboration, innovation, the spread of good practice, and the
strengthening of mutual accountability and transparency across groups
of schools and communities of practitioners.
To help us understand the nature of this shift and what it means for
the future of educational reform, we asked leading thinkers on both sides
of the Atlantic to give us their perspective. In chapter 5 distinguished US
academics Ann Lieberman and Diane Wood provide some important
insights into networks and teacher learning gleaned from the National
Writing Project. Demos Associate and leading UK education thinker
David Hargreaves then responds to their essay and offers some broader
reflections on disciplined innovation within educational networks in
chapter 6.

Demos

61

Networks collection

4/1/04

4:35 PM

Page 62

Networks collection

4/1/04

4:35 PM

Page 63

Network Logic 5

Networks these
loose and flexible
organisational
forms are
becoming an
important way to
organise teachers
and schools . . .
Untangling the threads
Ann Lieberman
Diane Wood

Networks collection

4/1/04

4:35 PM

Page 64

Networks collection

4/1/04

4:35 PM

Page 65

5. Untangling the threads

networks, community and teacher
learning in the National Writing
Project
Ann Lieberman and Diane Wood
In the press for higher standards on the one hand and test results on
the other, much of what we are learning about the necessity and
support for teacher learning in professional communities is getting
lost. Questions about how teachers learn, what they do with what
they learn, and whether it ever shows up in what students accomplish
are all being buried. But these are precisely the kinds of questions that
might lead us to understand better how to build an essential bridge
between teachers professional development and student learning.
It was these questions that excited us, serving as an impetus for a
19982000 study of two sites of the National Writing Project (NWP),1
arguably the single most successful professional development
network in the United States. We wanted to find out what teachers
learned in the NWP; what, if anything, they took back to their
classrooms; and how it showed up in the work they did with
students.2 Although we already understood that the sites were
organised locally and yet were linked nationally, we wanted to know
more about how these organisational arrangements were developed
and sustained and what effects such arrangements had on teacher
learning.
In the beginning
Long before we began our study we had heard from teachers that the
NWP transformed their lives. We wanted to know exactly what they
Demos

65

Networks collection

4/1/04

4:35 PM

Page 66

Network logic

meant by that. How, from the perspective of teachers, could a

professional development project be called transformative? We were
particularly interested in this question because, when most teachers
talked to us about professional development, they described it as a
waste of time and an interruption to their work rather than a support
for it. Typical in-services, they often complained, rarely provide
follow-up and, even when a topic is interesting, it rarely has anything
to do with helping them in their classrooms. What was it, we
wondered, about the NWP that seemed to earn such passionate
accolades from teachers?
We began our study aware that networks these seemingly loose
and flexible organisational forms are becoming an important way to
organise teachers and schools in productive efforts toward school
improvement.3 We also recognised there are growing and deepening
understandings about the power of professional communities for
teacher learning, though such communities are complicated, rare and
fragile.4 A crucial question left hanging, however, is whether or not
network learning ever actually finds its way into classroom teaching
and student accomplishment.
We learned that the NWP is a national network made up of many
local networks throughout the country. Each local site comprises a
schooluniversity partnership offering summer and year-long
professional development for teachers. The NWP bills itself as a
professional development initiative, providing multiple avenues for
teachers to learn about literacy and writing. Because the NWP has
been around since 1974 we had met many NWP teachers over the
years and knew how ardently they affiliated themselves with their
professional community and its mission. Research literature exists, of
course, about networks, about professional communities, and about
teacher learning. But the NWP had integrated all of these so
completely and placed them in such synergistic relationship with one
another that to study the dynamics of this interplay offers a
particularly intriguing, yet daunting challenge. The more we became
immersed in our study, the more we struggled to untangle these
tightly woven threads in order to understand what was going on.
66

Demos

Networks collection

4/1/04

4:35 PM

Page 67

Untangling the threads

Given the complexities, we decided to begin by studying two sites

that are, according to the national directors, particularly well
functioning. Both we and the directors wanted the study to focus on
sites that negotiated different geographical regions and socioeconomic conditions. We settled on the site centred at University of
California at Los Angeles (UCLA), a decidedly urban setting, and one
centred at Oklahoma State University (OSU) in Stillwater, Oklahoma,
a site encompassing a mid-size city, small towns and rural areas. By
closely observing activities in both sites and interviewing directors
and participating teachers, we gleaned some important understanding about the longevity and success of this professional
development initiative.
The five-week invitational institute
We were told by the NWP leadership, both national and site directors,
that we needed to attend and make sense of the five-week summer
invitational, the initiating experience for new NWP teachers, because
there lay the heart of the writing project. After we had seen the
institutes first-hand at both UCLA and OSU, we recognised that a
seamless set of activities unfolding over the five weeks succeeded in
galvanising a professional community. The model, as everyone called
it, has three interrelated features:

creating forms for teachers to teach one another their best

practices
providing opportunities for teachers to write and share in
response groups
engaging teachers in reading and discussing relevant
literature and research.

These seemingly straightforward ideas provide the frame for the fiveweek institutes. As teachers teach and learn from one anothers
practice, write and share their writing, and read and discuss research,
they rotate through a series of roles. For examples, they become their
colleagues teachers when giving teaching demonstrations or
Demos

67

Networks collection

4/1/04

4:35 PM

Page 68

Network logic

providing helpful feedback on colleagues teaching and writing. By

turns, they also become learners as they play audience to
presentations, get feedback after sitting in the authors chair, and
engage in research and professional discussions. By rotating such jobs
as logging the days activities and bringing in food for snacks, they
play the role of nurturing and responsible colleagues, attentive to the
needs of a developing professional community.
Subtly underpinning these activities are the habits of mind
necessary for the process of good writing. Teachers learn to reflect on
their own intentions as human beings and professionals and choose
compelling topics to pursue. They have the space and time for prewriting reflection, for drafting and getting feedback, for editing and
revising. They learn how their writing changes depending on their
purposes and intended audiences.
These activities, done in community, help teachers make
important connections. Teaching, like writing, requires clear and
compelling purpose, trials and errors, and revisions. And, like writing,
it flourishes best when taken public for collegial edification and
critique. Moreover, writing and teaching are both bolstered by
reading and research. Teachers who teach writing effectively and
comprehensively have more to do than simply teach the process. They
also draw on composition theory and cover such topics as genre, style,
grammar, structure, voice, audience and rhetoric. Over the five weeks,
an institute curriculum takes shape, incorporating both the active
process of writing and salient topics in literacy education. Moreover,
the curriculum makes clear connections to local educational
challenges. But it was initially honouring teachers knowledge that
engages teachers and encourages them to hold up their teaching for
feedback and critique.
Many sites (ours included) ask teachers to write three pieces: an
autobiographical re-creation of an experience; a personal essay
inspired by an experience; and an experimental piece a poem, story,
or one-act play. For a concluding assignment, teachers are asked to
address an issue on the teaching of writing or to write a journal article
for submission or a policy statement for their school. In this way
68

Demos

Networks collection

4/1/04

4:35 PM

Page 69

Untangling the threads

teachers experience what it means to be a student and they feel

viscerally what it means to engage in writing.
By the second day of the institute, teachers are reading their
unfinished pieces, working in their writing groups, teaching lessons to
a large group, reading educational literature and participating in
discussion groups. As the weeks progress, the institute runs itself as
the teachers take over primary responsibility for its rich, varied and
intense curriculum. Out of the give and take of these learning
activities, institute fellows develop characteristic ways of interacting
with one another. They develop, in effect, social practices that weave
together the process and content of their learning and embed the
cultural expectations and norms of the NWP, creating a strong
professional community. By the end of the institute, fellows have
internalised these practices, which include:

approaching every colleague as a potentially valuable

contributor
encouraging teachers to teach other teachers
creating public forums for sharing, dialogue, and critique
turning ownership of learning over to learners
situating learning in practice and relationships as well as
relevant knowledge
providing multiple entry points into learning
communities
thinking about teaching through reflection on learning
sharing leadership
adopting an enquiry stance
and rethinking professional identity and linking it to the
professional community.

The writing project approach to developing good writers and good

teachers underscores two principles. First, deep understanding arises
from practice in both of its common meanings. That is, both writers
and teachers learn from doing. Writers get better at writing by
actually writing and teachers get better by teaching. Moreover, both
Demos

69

Networks collection

4/1/04

4:35 PM

Page 70

Network logic

improve by practising in public. Valuing knowledge grounded in

practice gives rise to the second principle: teachers are the best teachers
of other teachers. Those actually involved in practice gain
contextualised understandings that make their ideas especially
compelling and believable. By taking the position that those who
practise writing or teaching are most likely to be good at teaching it,
the NWP foregrounds expertise rooted in practice and elevates the
peer-to-peer approach to learning rather than the expert to novice
approach.
Developing a local network
As we observed both of these institutes, we saw the seeds of a network
sown. The institutes induct teachers into a learning community
where they can concentrate on their own individual learning and yet
grow from the learning agendas of others. While providing
opportunities for teachers to discover and nurture a potential talent
teaching other teachers what they have learned from practice the
institutes lay the fundamental basis for the network. Having seen the
worth of teachers learning from one another, many fellows leave the
institutes committed to the professional development of their
teaching colleagues. At UCLA, for instance, there are over 40 teacher
consultants (TCs) who now run professional development
programmes throughout LA county.
Many of the active TCs with whom we talked told us that, when
they taught other teachers, they continued to learn themselves. For
instance, a number of teachers found that in sharing their teaching
practices with their fellow teachers, they became quite articulate
about the hows and whys. Despite having built expertise over the years
as they tried, honed and shaped these practices, teaching them to
other teachers always provided fresh insights and deeper
understandings.
Although many of those teachers who successfully complete the
institutes become TCs, there are some who choose not to participate
in teaching other teachers. Those who do, however, are frequently
kept quite busy providing professional development for schools and
70

Demos

Networks collection

4/1/04

4:35 PM

Page 71

Untangling the threads

districts, and they get paid for their work. Some get involved in
special interest groups on topics of particular concern, such as writing
in bilingual classrooms or improving advanced placement teaching.
Others lead or participate in teacher research groups that meet
throughout the year. Figures indicate that a substantial leadership
cadre develops in each site and that the experience is powerful.5 As
one teacher consultant explained:
On one level TCs work a lot on their own teaching. The
continued discussions with teachers about teaching, whether
giving a presentation or at meetings regarding new NWP
projectsOn another level, you learn a lot about teacher
learning through experience. It is not explicityou have to
stretch your thinking as a presenter as to what texts and
structures you can use to give your audience a chance to
experience the presentation rather than to watch it.
Moreover, the NWP itself provides opportunities for leadership in the
organisation. At each local site an advisory board of TCs is created
ensuring that the local site stays rooted in teachers definitions of their
classroom needs. The UCLA site leadership consists of a director, codirector, three associate directors (all teachers) and others in the
university who works with schools. Similarly at OSU, several teacher
consultants help create the policies for a year-long programme as they
gain experience and provide an important voice in the development
of the local network.
Summer and year-long programmes
Although all sites hold a summer invitational institute in common,
the types of formats they develop and the content of their
professional development differ in interesting ways. The particulars of
the context place demands on the networks as they respond to state
and local policies as well as to differences in urban, rural and
suburban environments. The social practices learned in the summer
invitational enact and flesh out core values in the sites, allowing
Demos

71

Networks collection

4/1/04

4:35 PM

Page 72

Network logic

managers to make the necessary adaptations to their contexts without

losing the core of what it means to be a NWP site. For example, at
UCLA, a 24-year-old site, two state-wide initiatives have greatly
affected teaching and learning in the Los Angeles district. Proposition
227 limits a bilingual student to one year of English immersion, and
high stakes standardised tests are used as the criteria for passing and
failing courses and graduating from high school. Summer offerings of
the UCLA site respond to such changing policy contexts for teachers.
The UCLA site, in response to Proposition 227 and high stakes
testing, organised summer workshops, such as Writing from Day
One, a week-long offering to help teachers with strategies that enable
bilingual students to improve their writing while they are learning
English. Research to Reading, a state-wide programme, became
another offering, this one designed to help teachers avoid narrowing
the literacy curriculum by demonstrating that serious incorporation
of language arts standards could ultimately lead to higher test scores
and to better writing.
At OSU, because the site encompasses such large distances, housing
arrangements are made for the five-week invitational. Whenever
possible, workshops are arranged to piggy-back on other occasions
and minimise the driving time for teachers. To provide a measure of
consistency the OSU site has created its own model of writing project
work, designed to reach out to teachers in remote towns and rural
areas and to bring new and experienced teachers the benefits of
teaching demonstrations, opportunities for teacher writing, writing
groups and teacher research.
These examples speak to ways that teachers can contribute to the
learning of colleagues while continuing their own learning. They also
speak to the flexibility and adaptability that a network-like way of
organising affords. The NWPs network structure keeps it sensitive to
a dynamic and changing environment, while its social practices keep
its cultural identity intact and sustain its broader mission.
Besides responding to local needs, the NWP has further ways to
answer the professional development needs of teachers. The national
office, for instance, helps facilitate sub-networks that cut across sites.
72

Demos

Networks collection

4/1/04

4:35 PM

Page 73

Untangling the threads

These networks work on common problems identified by the TCs

and brought to the attention of the national office. To date, the NWP
has four such sub-networks: Project Outreach (formed when the
NWP realised that it was not serving poor and minority populations
equitably); the Urban Sites (which works through particular
problems of language and low-achieving schools); Rural Sites (a
network focusing on rural needs with a particular focus on
technology); and the English as a Second Language network (focusing
on second language learners). The national office coordinates the
evaluation and documentation of all 167 sites, its quarterly and
bimonthly publications and, increasingly, books written by and about
the writing project.
Teacher learning in the National Writing Project
Also enmeshed in each summer institute (and the forthcoming
workshops facilitated by TCs) is a view of learning that turns
conventional professional development on its head. Instead of
teaching teachers someone elses ideas of how they should improve
their teaching, the institute begins with what teachers know and then
moves to what others have learned. This respect for teacher
knowledge is so powerful that teachers eventually open themselves up
to going public with their writing and teaching in an environment
that invites feedback and critique. It is not hard to see over time how
teachers learn from one another, constantly thinking about how
particular strategies can be adapted to suit their particular students.
Besides learning from one another, teachers are introduced to a
wide variety of reading materials including books to expand their
own classroom library. They learn by reading and discussing research
on literacy and by being taught by experienced TCs who have become
expert in particular areas (such as assessment and multicultural
literature). In turn, teachers, through the process of being in a new
group, learn how to become good community members (for instance,
sharing the responsibility for being a presenter on one day and an
audience the next, or being the logger of the days events). Teacher
learning here becomes multifaceted. Teachers are learning by
Demos

73

Networks collection

4/1/04

4:35 PM

Page 74

Network logic

presenting their own practice, by listening to others, by reading and

discussing research and literature together, by being in a group, by
taking responsibility for the groups needs, and by taking risks
together. In short, they are learning how to be members of a
democratic community that values them, their knowledge and their
continued growth.
In this article we have discussed the site as a local network
embedded in a national network bringing its group of teachers into a
professional community that respects the variety of ways that
teachers learn. But the magic of the writing project is that all these
things are tightly integrated into an experience that socialises teachers
into a new way of thinking about themselves as teachers. While the
social practices provide the core, the network-like way of organising
helps develop a local constituency. Untangling these threads has
helped us to see a powerful model for professional development and
yet another way to develop a professional community situated in
teachers practice.
Ann Lieberman is an emeritus professor at Teachers College, Columbia
University, and a senior scholar at the Carnegie Foundation for the
Advancement of Teaching. Diane R Wood is an assistant professor in the
Educational Leadership Program, College of Education and Human
Development, University of Southern Maine. This essay is an edited
version of an article that first appeared as Untangling the threads:
networks, community and teacher learning in the National Writing
Project, Teachers and Teaching 8, no 34 (2002), pp 295302.
Notes
1

A site in the NWP grows out of a universityschool partnership. The university

is the owner of the site. Beginning sites receive $20,000 and must document
their plans and their work and keep accurate information about who
participates and the nature of their formats for work during the year. The
programme is sponsored by the federal government.
This paper quotes liberally from A Lieberman and D Wood, Inside the National
Writing Project: network learning and classroom teaching a new synthesis (New
York: Teacher College Press, 2003).
J Adams, Taking Charge of Curriculum: teacher networks and curriculum

74

Demos

Networks collection

4/1/04

4:35 PM

Page 75

Untangling the threads

implementation (New York: Teachers College Press, 2000); A Lieberman and

MW McLaughlin, Networks for educational change: powerful and
problematic, in Phi Delta Kappan 73, no 9 (1992); A Lieberman and M
Grolnick, Networks and reform in American education, Teachers College
Record 98, no 1 (1996).
P Grossman and S Wineburg (forthcoming), In pursuit of community,
Teachers College Record; MW McLaughlin and J Talbert, Professional
Communities and the Work of High School Teaching (Chicago: University of
Chicago Press, 2001); E Wenger, Communities of Practice: learning, meaning,
and identity (Cambridge: Cambridge University Press, 2000); J Westheimer,
Among Schoolteachers: community, autonomy and ideology in teachers work
(New York: Teachers College Press, 1998).
M St John et al with assistance from A Murphy, The National Writing Project:
client satisfaction and program impact (Inverness, Ca: Inverness Research
Associates, 2001).

Demos

75

Networks collection

4/1/04

4:35 PM

Page 76

Networks collection

4/1/04

4:35 PM

Page 77

Network Logic 6

Innovation is a
social, interactive
process rather than
one of individual
creativity, and
networks play a
vital role . . .
Networks, knowledge
and innovation
David H Hargreaves

Networks collection

4/1/04

4:35 PM

Page 78

Networks collection

4/1/04

4:35 PM

Page 79

6. Networks, knowledge
and innovation
reflections on teacher learning
David H Hargreaves

A response to the previous chapter, Untangling the threads, by Ann

Lieberman and Diane Wood.
Academic educationists, like teachers in schools, have strongly held
values. Naturally they choose to study phenomena that relate to those
values and preferably support them. Ann Lieberman has a
distinguished reputation for her work on teachers. I count myself
among those influenced by her edited collection of essays, Building a
Professional Culture in Schools and her book Teachers Their World
and their Work.1 I imagine that she was thrilled to discover what was
happening in the National Writing Project, for this is what appears to
be a resounding success story about her passion for professional
learning among teachers. Her essay, with Diane Wood, glows with the
excitement and enthusiasm experienced by educationists who
discover some evidence that supports their beliefs and that is
welcome good news to a profession longing to have its own beliefs
and commitments validated, especially when these collide with
government-led initiatives driven by the standards agenda.
In this short response to their essay, which in many ways inspires
me as much as it does them, I want to set their ideas and findings in a
wider conceptual framework and a rather more sceptical perspective
on the achievements of the National Writing Project (NWP). In both
Demos

79

Networks collection

4/1/04

4:35 PM

Page 80

Network logic

ways I seek to contribute to the advancement of academic theory and

research, as well as to teacher practices, in a field where rapid change
and development can surely be expected.
Professional development and student learning
Lieberman and Wood seek to untangle the threads of the dynamics
of the NWP that, in their view, potentially builds the essential bridge
between teachers professional learning (their own passion) and
student learning (and its outcomes in measured achievements, the
passion of politicians pressing for higher standards). If a causal link
can be demonstrated between the kinds of teacher practices
developed in the NWP and improved student learning, then this not
merely adds to our knowledge of what works in schooling; it also
validates the approaches to professional learning that Lieberman has
long advocated rather than the top-down interventions that have
been attractive to political reformers. Lieberman and Wood do not
adduce evidence for this causal link: they assume it. If it does exist
and we must look elsewhere for the hard evidence then the
relationship between teachers professional development and student
test scores will, as they claim, be a complex one, with many threads
that need to be untangled. Or, to put the matter in more conventional
research terms, there will be many variables involved and these are
likely to interact with one another in complex ways.
In essence, the essay offers a list of possible variables, but little
insight into which of them might be of particular importance or into
the ways in which they interact (or, in their terms, become tightly
woven). The evidence is based on interviews with the participants
and observation of some (unspecified) activities. It appears that the
participants were enthusiasts for the project and their claims of its
success are taken at face value. It is not clear how much of the analysis
is based on the accounts yielded by the interviews and how much on
what is inferred from researchers observations. It is, in short, a piece
of preliminary or exploratory soft research, but the findings are not
then set within a harder framework, such as a set of hypotheses that
might be tested. This is a pity, for if there is firm evidence of a causal
80

Demos

Networks collection

4/1/04

4:35 PM

Page 81

Networks, knowledge and innovation

link between certain approaches to teachers professional development and student achievement, then, as is implicitly suggested, this
may extend beyond student writing and apply to a far wider range of
teacher practices in classrooms. It could potentially entail the
discovery of what has hitherto been elusive: teacher-friendly ways
to engage with professional development resulting in improved
student outcomes, to the delight of politicians demanding higher
standards.
The reluctance to formulate any hypotheses is thus a surprise as
well as a pity, for 15 years ago Lieberman presciently understood the
need for some hard analysis:
We often think of solving problems in our own institutions. We
rarely think of forming coalitions or networks outside existing
formal channels. And it is even rarer that we think of loose,
informal collections of people (networks) as catalysts for change.
We may very well be in a period where we grossly underestimate
both the attack on teachers and the amount of support needed to
make improvements in practice Our concern is understanding
[networks] from the inside, getting a sense of the subtleties, and
using examples as a way of conceptualising what we know about
networks.2
At that time, the key ingredients of a successful network, derived from
their analysis of educational networks then in existence, were:

organising for participation

developmental substance or some agreed content
developmental mechanism for network interaction and
sharing
planning new rewards, including experimenting with new
ideas
problem-solving orientation to help participants find
their own solutions to problems
diagnosis of what participants know and need
Demos

81

Networks collection

4/1/04

4:35 PM

Page 82

Network logic

strategy-building through action plans

organising for linkage among participants with different
perspectives and potential contributions.

There appears to be some overlap between these eight processes that

we consider essential for the creation of a network and the key
ingredients of the NWP, but there are also real differences. The NWP
study emphasises, among other factors:

teachers teaching professional colleagues about their best

practices
teachers honouring and respecting the knowledge of
colleagues
teachers writing as well as meeting and talking
teachers getting feedback on their learning and actions
teachers rotating through a variety of roles.

How are these differences to be explained? This takes us back to the

need to determine the strength of the evidence for NWP, since in their
turn the 1990 networks were largely assumed to be successful. Is
NWP demonstrably more successful than its predecessors and thus a
more reliable source of sound evidence of what makes for success?
Does the current evidence on NWP replace the key ingredients
suggested in 1992? If our knowledge about networks is accumulating,
can we now with confidence proclaim what works and what does not?
Understanding networks
Over the intervening 10 to 15 years, interest in networking, as well as
its associated literature, has grown dramatically, but this is not well
reflected in the Lieberman and Wood report. Today we have a better
grasp of the issues of both the structure and culture of networks.
Networks have many different kinds of structure. Does the NWP have
a distinctive structure that differs from those studied in 1990? Indeed,
exactly what kind of structure does NWP have? We are told that the
networks are organised locally but linked nationally. What are the
82

Demos

Networks collection

4/1/04

4:35 PM

Page 83

Networks, knowledge and innovation

local networks like? Do they vary much in structure, for example

those in the urban rather than rural sites? What are the units of the
networks? Are they networks of individual teachers? Or are they
sometimes networks of schools? Or could they be both? Does it
matter that NWP is in some ways a network of networks? Do
networks of networks function differently from networks of
institutions (schools) or individuals (teachers)? Are some network
structures more successful than others in their impact on student
learning? While NWP may be rated an overall success, it seems likely
that some sub-networks in some areas will be more successful than
others. What are the characteristics of the most and least successful
networks? Does this help us to understand the nature of the most
important variables in systems of professional networking? Does the
teacher consultant act as a kind of hub that helps the network to
flourish? Are networks without hubs of some kind more liable to
failure or disintegration? What is the role of weak ties in a national
network as opposed to what are probably the stronger ties of a local
network? In short, is the structure of a network critical to the way it
works and influences teacher development and student learning?
While it is difficult to glimpse the structure of the networks from
the Lieberman and Wood description, their culture is more visible.
Back in 1990 there was relatively little work on social capital, and
virtually none in education. Today it is a highly developed even
overdeveloped concept, not least in the field of networks. For
networks have a structural side, the nature of the links between the
nodes, but also a cultural aspect, which is usually encapsulated in the
term trust. The glue behind the strong ties of local networks is trust:
without trust, networks rarely prosper. It is possible to argue from the
Lieberman and Wood account of the way the NWP was organised
and functioned that high levels of social capital were built up among
the participants and crucial to the projects success was this
generation of high levels of trust between teachers. For example,
getting teachers to show their writing to fellow teachers and obtaining
feedback on it depends on there being a level of initial trust; if the
experience is then felt to be professionally rewarding, it will generate
Demos

83

Networks collection

4/1/04

4:35 PM

Page 84

Network logic

more trust, which in turn strengthens the network itself. It is arguable

that NWP succeeded because it tapped and refreshed professional
trust among the participating teachers.
To say this is not simply to play with words for, if this hypothesis is
correct, there may well be many different ways of generating high
levels of social capital in addition to those specified in the analysis of
NWP. The explanation may not lie in particular practices, such as
teachers engaging in writing, but in the fact that here is one activity
that in this particular context helped to generate high social capital.
In a different context another curriculum subject or another aspect
of schooling very different practices may be successful in generating
the high social capital that allows teachers to learn with and from
their colleagues in a way that benefits their classroom practices.
Indeed, one of the things we really do need to untangle is whether
peer-to-peer approaches are always better than expert-to-novice ones
in education, or whether the latter have often been less successful
because they have lacked a basis of social capital. Indeed, expert-tonovice systems do seem to work in apprenticeship-type relationships
where trust is well established, as documented by Jean Lave and
Etienne Wenger among others.3 The problem may be not that expertto-novice approaches are ineluctably destined to failure in the case of
teachers professional learning, but that we have missed out of our
analysis the key underlying feature, social capital, and mistakenly
assumed that the surface features, the participants identities as
experts and novices, are the critical variables that explain the low
success rate.
Networks, knowledge management and innovation
This brings us to another aspect of the wider conceptual framework
for networks since 1990, namely the dramatic growth of interest in
knowledge management and innovation.
From a knowledge management perspective, what we have
traditionally called professional learning is very often a form of
knowledge creation and knowledge transfer, alternatively conceived as
innovation and the dissemination of such innovation. We now
84

Demos

Networks collection

4/1/04

4:35 PM

Page 85

Networks, knowledge and innovation

understand better than ever that innovation is very often a social,

interactive process rather than one of individual creativity, and that
networks play a vital role in the creation and the transfer of new
knowledge and innovation. In NWP teachers share best practices
and do so in a way that seeks to improve on them as well as spread
them. The question thus arises: to what degree would it be
advantageous to set the Lieberman and Wood analysis not only in its
traditional location of professional learning but also in new locations
of knowledge management and innovation? There are at least two
advantages of so doing. First, it offers the analysis a new and
potentially richer set of concepts in which to set NWP and the
analytic frame adopted. Second, it would bring into view a new and
different research base.
Best practice

Three examples might help here. Lieberman and Wood commend the
ways teachers share their best practices, but they treat the concept of
best practice as entirely unproblematic. Exactly what are the defining
characteristics of a good practice in NWP? It might, of course, mean
nothing more than an ideologically approved or politically correct
pedagogical practice within the values of NWP. It might mean a
practice that an individual teacher has found to be effective in her
private experience and judgement. It might mean one that is
demonstrably more effective in ensuring student learning. If NWP is
effective in knowledge transfer through its networks, it is rather
important that the good practices disseminated are ones that are
indeed demonstrably effective by some objective evidence. What
action was taken to ascertain the basis of the good practices involved
in NWP? How much of this learning could be used in other forms of
professional development?
Good practice and best practice are often treated as synonyms,
even though best is not the same as good. A practice could be
considered best if it has been shown to be better more effective or
efficient than other practices. Did NWP uncover or elaborate on a
process of moving from good practice, however defined, to a
Demos

85

Networks collection

4/1/04

4:35 PM

Page 86

Network logic

demonstrable best practice, rigorously defined? If so, this process

would have relevance to a wide range of professional development
and reform activities in education.
The process of transfering knowledge

We know that transferring knowledge from one person or one

context to another is often difficult, and especially so, as in the case of
many of the classroom practices of teachers, when the knowledge
involved is heavily tacit rather than explicit, and not easily put into
words. It is possible that NWP owed much of its success to the fact
that unusual amounts of knowledge were successfully transferred
between teachers, which would be rewarding to both donors and
recipients of such knowledge. But what exactly were the processes by
which the knowledge was transferred and what role did particular
forms of networking play in this? The knowledge management
literature would entail treating knowledge transfer between teachers
as a topic worthy of detailed investigation.
Networks, ICTs and best practice

A third example is the new information and communication

technologies (ICTs), which again have grown dramatically over the
last decade. We do not know what role the ICTs might play in creating
and sustaining professional networks: they will have limits and will
have yet to be discovered strengths, even though it is virtually certain
that they will complement face-to-face interactions rather than
displacing them. ICTs may be of particular importance for
networking by teachers who are isolated in their teaching specialism
or in their physical location.
At the same time, we should not underestimate the capacity of
innovation and best practice networks to devise solutions to
problems that arise and to borrow ideas from the internet. Take the
way Amazon.com works, for instance. You look up a topic, and are
provided with a list of books. You look up a book and in addition to
details of its content, price and so on, two further resources are put at
your disposal. First, you are offered reviews of the books, by the
86

Demos

Networks collection

4/1/04

4:35 PM

Page 87

Networks, knowledge and innovation

author or professional reviewers, as well as other Amazon customers.

You are also told whether customers found these reviews helpful.
Second, Amazon tells you which other books a purchaser of the target
book has also bought. Displayed before you is an elaborate set of
factual information and evaluations to help you make a more
informed decision about book-buying.
Epinions.com offers a similar service. It will search millions of
products and services such as books, movies, cars, restaurants,
computers, sports and travel and tell you where to obtain the lowest
price for them and which stores are most trusted by customers.
Products and services are reviewed and rated by customers, and these
are available to all other customers. Customers rate reviewers for the
quality of their reviews, and reviewers whose judgements are trusted
by their peers are designated top reviewers. You are also told which
other reviewers the top reviewers most trust. You become a top
reviewer only if you have earned such a reputation for your advice to
other customers.
An innovation and best practice network for teachers might have a
similar infrastructure. The quality of an innovation and validity of a
claim to good or best practice could be rated by those who had tried
to transfer it, as well as by experts such as researchers or Ofsted.
Indeed, the trustworthiness of the judges would also be rated by
practitioners, for this would be particularly important in relation to
judgements or claims about high leverage and ease of transferability.
For academic researchers to have to earn their reputation for
trustworthiness would be a gain for both them and for teachers. The
system would also need to give information on the location of
the nearest helper-practitioner or consultant, since accessibility
and opportunities for a face-to-face meeting as well as coaching
and mentoring are vital. The NWP might be the kind of project to
gain from such a development borrowed from ICT in the business
world.
Conclusion
Ann Lieberman has been at the forefront of developments in teachers
Demos

87

Networks collection

4/1/04

4:35 PM

Page 88

Network logic

professional learning and its associated communities and networks.

Today understanding the dynamics involved is more important than
ever. If we are challenged by more demanding questions, this is
because there are more opportunities both for imaginative
innovation and for better methods of investigating and analysing
these creative developments.
David H Hargreaves is a fellow of Wolfson College, Cambridge, and
Chairman of the British Educational Communications and Technology
Agency (BECTA). He is a senior associate of Demos and Associate
Director (Development and Research) of the Specialist Schools Trust,
and was formerly Chief Inspector of the Inner London Education
Authority and Chief Executive of the Qualifications and Curriculum
Authority. His latest publication is Education Epidemic: transforming
secondary schools through innovation networks, published by Demos
in 2003.
Notes
1

2
3

88

A Lieberman (ed.), Building a Professional Culture in Schools (New York:

Teachers College Press, 1988); A Lieberman and L Miller, Teachers their World
and their Work: implications for school improvement (New York: Teachers
College Press, 1992).
Lieberman and Miller, Teachers their World and their Work.
For example J Lave and E Wenger, Situated Learning: legitimate peripheral
participation (Cambridge: Cambridge University Press, 1991).

Demos

Networks collection

4/1/04

4:35 PM

Page 89

Network Logic 7

Asked about
leadership, most
people reach for
the organogram.
But when it comes
to networks there
are no such easy
answers . . .
Leading between
Paul Skidmore

Networks collection

4/1/04

4:35 PM

Page 90

Networks collection

4/1/04

4:35 PM

Page 91

7. Leading between
leadership and trust in a network
society
Paul Skidmore

The institutional landscape of modern society is being ripped up. Be

they companies or public agencies, individual organisations are
finding that the only way to satisfy the changing demands and
expectations of customers and citizens is to be embedded in networks
of organisations able to stitch together different products, services,
resources and skills in flexible combinations and deliver them when
and where they are most needed. Splendid isolation is out.
Collaboration is in.
But this radical disruption also spells trouble for many of the
assumptions we have about what leadership means, what it is for and
where we might look for it. Networks challenge our conceptions of
leadership, which too often are still rooted in an outmoded great
man theory that mistakes the formal authority of status, rank or
station with the exercise of leadership. When you ask people about
the leadership of an organisation, most people reach for the
organogram and point to the top. When it comes to leading across
networks there are no such easy answers.
New network-based ways of organising social and economic
activity will only thrive if we can evolve new models of leadership that
embrace the distinctive organising logic of networks, and do not seek
to apply an old set of principles in an environment that has been
dramatically altered. We must learn what it means to lead effectively
not just within individual organisations, but across the networks of
Demos

91

Networks collection

4/1/04

4:35 PM

Page 92

Network logic

which they are part. Leading between will be the new leadership
imperative of the coming decades.
The challenge of leadership in a network society
Our increasing personal and institutional interconnectedness, the
long-term trends driving it, and the challenges that arise from it, are
all familiar terrain.1 The organisational responses to these developments have also been chronicled. Companies have been reorganised
internally as networks of sub-units, and externally as specialised hubs
in distributed production networks involving other suppliers and
subcontractors, often crossing national boundaries.2 This model is
exemplified by Cisco Systems, a company that mediates between
customers and a diverse array of manufacturers of components used
in information technology networks.
According to Shoshana Zuboff and James Maxmin, these trends
are set to accelerate and intensify in the coming decades with the
emergence of the support economy. Their thesis is that the drive to
specialisation has left individual corporations unable to provide the
deep support that consumers need to help them navigate through
ever-more complex arrays of choice and offering, or to engage with
the personals needs and aspirations of individual customers. As a
result, most will therefore find themselves drawn into federated
support networks: fluid configurations of firms brought together to
provide unique aggregations of products and services.3
The same drive to integrate has also been felt across the public
sector. Public policy problems are now understood to cut across
traditional institutional boundaries. As Prime Minister Tony Blair put
it, Even the basic policies, targeted at unemployment, poor skills, low
incomes, poor housing, high crime, bad health and family
breakdown, will not deliver their full effect unless they are properly
linked together. Joined-up problems need joined-up solutions. Public
services are under growing pressure to offer genuinely personalised
solutions if they are to meet the individual needs of an increasingly
demanding citizenry.4 Yet the agencies charged with meeting these
challenges have spent the last century retreating into ever-more
92

Demos

Networks collection

4/1/04

4:35 PM

Page 93

Leading between

specialised functional silos, supported by powerful institutional,

professional and disciplinary cultures or tribes that make effective
coordination very difficult.
The trouble is that this increasing interconnectedness does not
reduce our requirement for leadership. By creating new and tough
problems, and undermining the legitimacy and effectiveness of some
of the traditional institutional responses to them, it actively increases
it. But the question is what kind of leadership do we need?
Old theories die hard
Too often in the face of these pressures we look to locate leadership at
the top of institutions. In an uncertain world, we expect individual
leaders to somehow provide certainty where previous leaders could
not: We call for someone with answers, decision, strength, and a map
of the future, someone who knows where we ought to be going
someone in short who can make hard problems simple.5
In education, for example, we have seen the rise of superheads
headteachers and principals brought into failing schools and given
more resources and higher remuneration on the assumption that they
will personally be able to reverse the decline, often with very mixed
results.
In local government we have seen the introduction of US-style
elected mayors, in the hope that concentrating power in a single office
will create more visible and effective leadership. But so far these have
failed to capture the imagination or energise citizens, and in the few
municipalities that have opted for local mayors electoral turnouts
have not markedly improved.
In business we have seen the cult of the CEO, with senior
executives paid vast salaries because corporate survival is seen to
depend on attracting and retaining talent. But as the controversy over
rewards for failure indicates, many such remuneration packages are
only tenuously linked to actual business performance, and in a
number of notorious cases (for example, at GE Marconi) executives
have been given multi-million pound severance packages even after
leading their companies to the verge of ruin.
Demos

93

Networks collection

4/1/04

4:35 PM

Page 94

Network logic

Finally, in the reform agenda of the European Union, particularly

as embodied in the recent Constitutional Convention, we are seeing a
push towards a conventionally hierarchical model of political
leadership, with greater decision-making power concentrated at the
centre. As Mark Leonard has argued, this search for some neat
institutional arrangement ignores the fact that the EU is more like a
network than a traditional organisation. It misses the chance to
breathe new life into a debate hamstrung by the false choice between
federal superstate and a free trade area, and could undermine the very
flexibility on which EU integration has depended.6
What is striking about all these examples is that the response to a
crisis of authority is to reinforce the traditional model of leadership.
We seek saviours, and then berate them when they fail. Wherever we
look, our instinctive response to the complexity of organisational life
is to strengthen the very forms of institution, and institutional
authority, that it has exhausted. The command-and-control form of
authority on which most large organisations were built does not tally
with the underlying social reality. It is at odds both with the
complexity of the context in which they are asked to operate and the
prevailing social expectations about how they should behave.
Control, as Veenkamp et al argue, seems more important but less
feasible than ever before.7
We need to take a different starting point.
Leading with questions not answers
In Leadership Without Easy Answers, Ronald Heifetz argues that
conventional models of leadership confuse it with authority. In so
doing, they perpetuate the seductive but dangerous myth that
leadership is about influence and persuading people to follow a
particular vision. So followers look to a leader to solve their
problems for them, ignoring their own capacity (and responsibility)
to solve it for themselves. People in authority believe that their vision
of change is legitimate simply because they are leaders. And when
things go wrong, it is the leaders who are blamed and replaced, with
little or no reflection on the underlying causes of the problem.
94

Demos

Networks collection

4/1/04

4:35 PM

Page 95

Leading between

To put it another way, leadership is not something you are but

something you do; it is an activity, not a position. For Heifetz,
leadership is about mobilising people to do what he calls adaptive
work. It is about forcing them to confront the gap between the
rhetoric of what they are trying to achieve and the reality of their
current capacity to achieve it. Leaders do not try to impose change.
Instead they make the case for why change is necessary, and
then make the space for it occur. Leaders create a holding environment for those they lead, managing the tension and stress that
change inevitably generates but never allowing them to run away
from it.
This simple insight is profound in relation to leadership within an
organisation, but it is revolutionary in helping us to see the challenge
of leadership across networks. Divorced from formal positions of
authority, leadership mobilising people to do adaptive work is as
feasible between organisations as it is within them, even if the
resources that are deployed and constraints experienced may vary
depending on the context.
The six characteristics of network leadership
So what is it that network leaders do?
Network leaders lead from the outside in

As the Global Business Network notes, many firms think about their
strategy from the inside-out, beginning with the organisations
purpose and core strengths, then working out to explore its
marketplaces and only then looking externally for broader,
underlying shifts that might matter.8 The problem is that by the time
they get there they have imposed so many filters that theyre not
seeing the real world at all. They are looking through the lens of their
own perspectives and assumptions about what matters, not those of
the customers, users or citizens they are there to serve.
Network leaders start from the outside-in. They start with the
deepest needs of their users, and work back to establish the
configuration of organisations, resources and capacities needed to

Demos

95

Networks collection

4/1/04

4:35 PM

Page 96

Network logic

meet them. The task then is to find ways of persuading other

organisations of the need to work together.
Network leaders mobilise disparate supplies of energy

Leadership is often seen to be synonymous with decisive action:

defining a vision and pursuing it. Network leaders understand that
decisive action may be of little use in an unpredictable world, particularly when the knowledge about how best to improve performance
often resides in the tacit and explicit knowledge of front-line staff.
As Nonaka and Takeuchi argue, finding ways to unlock and
harness this knowledge by developing procedures for the creation and
sharing of knowledge among staff is therefore a crucial leadership
task.9 In this context, leadership is less about decision than
deliberation.
Douglas Rushkoff argues that the real power and attraction of the
internet is not the knowledge or facts or ideas it supplies but the
opportunity to interact with others: Content is not king. Contact is
king.10 The same goes for leadership. Network leaders know that they
cannot provide some definitive vision statement but they can
structure the right kind of conversation. They can create a language
that enables people to cross boundaries within or beyond their
organisation that they otherwise would not.
Network leaders foster trust and empower others to act

But deliberation does not mean inaction. Networked leadership is not

leadership by committee, where the sole criterion for action is the
lowest common denominator. As Danny Chesterman argues in his
study of leadership in local multi-agency partnerships, The first
assumption is that consensus is necessary by all before any
partnership can act collaborativelyWe talk as if agreement is a
precondition for action. It isnt. But sufficient trust is.11 Network
leaders understand that different actors will not always agree on the
appropriate course of action, not least because in a complex world the
correct path will rarely be clear, and stumbling upon it may require
processes of trial and error, and learning by doing.
96

Demos

Networks collection

4/1/04

4:35 PM

Page 97

Leading between

By sharing perspectives and building understanding, however, it is

possible to foster the trust and the set of common values on which all
networks depend, and which are robust enough to withstand
considerable variety in the actions undertaken by others. Geoff
Mulgan describes the medals once awarded to the general who
disobeyed orders, but in so doing changed the course of battle.12 It is
a spirit that lives on today. The day before the Eden Centre in
Cornwall opened to the public, managing director Tim Smit called
the staff together and said, Tomorrow, people will ask you for things,
or to do things, we havent thought of. If you respond in a way which
goes wrong, no one will blame you. If you do nothing, Ill sack you.13
True authority, as Capra puts it, consists in empowering others to
act.14
Network leaders help people grow out of their comfort zones

Network leadership would not be necessary if the organisational silos

in which many of us find ourselves were not so attractive. As senior
managers responsible for multi-professional learning in Britains
NHS explained, these tribes provide us with stability, a sense of
identity and a shared language that allows information to travel fast.
Above all, they allow us to maintain our existing routines: Its very
seductive to fall back into old behaviours because thats the known
world, as one put it.15 In the public sector, multi-agency working is
now de rigueur, with local service delivery of everything from
education (such as Excellence in Cities clusters) to economic
development (like local strategic partnerships) structured around
networks of agencies. Unfortunately, partnership is often treated as a
structure rather than an activity, and formal mechanisms for
decision-making are put in place before the different actors have had
a chance to move out of their particular silos.16
Network leaders understand the attraction of these comfort zones
but look for ways to help people grow out of them. Traditional
performance management systems typically reward people for staying
within particular silos and running away from the problems that fall
between the gaps but, as Karen Stephenson shows, developments in
Demos

97

Networks collection

4/1/04

4:35 PM

Page 98

Network logic

social network analysis have allowed a number of organisations to

begin to reward people for running towards them. In the public
sector, extrinsic rewards have often proved effective in the short run,
with organisations happy to work in partnership so long as there is a
clear and immediate return on their investment. But longer-term
commitment seems to reside in more intrinsic rewards: tapping into
peoples sense of professionalism, and reconnecting them with the
higher moral purpose that first motivated them to enter that
particular field.
Network leaders are lead learners not all-knowers

Certainty of vision is wrapped up in many of our mental models of

leadership. But in the modern world this can be a dangerous myth,
leading us down seductive avenues that turn out to be blind alleys
think of Cable and Wireless doomed foray into business services
under Chief Executive Graham Wallace, a strategy that may have
seemed sensible at the height of the internet bubble but a few years
and 35 billion in lost shareholder value later increasingly looked
like a great example of what Michael Fullan calls the visions that
blind.17
The original meaning of authority, Fritjof Capra has noted, is not
power to command but a firm basis for knowing and acting.18
Given the complexity of modern organisational life, it seems the only
firm basis for acting is to be a permanent learner. Network leaders do
not see themselves as all-knowers but as lead learners. They
understand that a large part of leadership is about shutting up and
listening. Network leaders make a point of not having all the answers.
Network leaders nurture other leaders

At Lipson Community College, a large secondary school in

Plymouth, the pool of potential leadership talent is drawn very
widely. In fact, it extends to students themselves. Older students have
received coaching as mediators to help younger pupils settle disputes
or other problems getting in the way of their learning without
involving staff. Students of any age are encouraged to become lead
98

Demos

Networks collection

4/1/04

4:35 PM

Page 99

Leading between

learners, trained to take real responsibility for their learning and to

mentor others.
Candidates for new teaching posts are asked to teach a lesson, and
students in the class give feedback on their performance to the
schools management team. Smaller groups of students who have
been given special training then comprise one of the interview panels
and often have the main say in who is appointed. Principal Steve
Baker talks of playing a long game through a sustained programme of
activities that brings in, reaches out to and raises expectations of the
whole community.
Network leaders like these understand that leadership is not about
a simple transaction between leaders and led. Instead they reach back
to the ancient ideal of self-government as the ultimate goal of
leadership. They understand that most systems from organisations
to cities to biological ecosystems are too complex and unpredictable
to be controlled from the top-down. Yet they display an underlying
tendency towards self-organisation and order, leading to what Briggs
calls meaningful patterns of uncertainty.19 This self-organisation can
be shaped in purposeful ways, provided we can develop leadership
models that distribute leadership across organisations rather than
imposing it from the top. To align leadership with the built-in
instinctive adaptive responses of organisations, in other words,
network leaders understand the need to nurture other leaders
wherever they may be found. As Sun Tzu put it long ago: The good
leader is the one the people adore; the wicked leader is the one the
people despise; the great leader is the one the people say we did it
ourselves.
Trust, betrayal and network leadership
Network leadership is increasingly necessary if organisations are to
satisfy the needs of those they serve. But the mental leap involved in
accepting network leadership is not easy. Perhaps the most important
commodity for this new conception of leadership to take hold is trust.
Leaders in hierarchies rely on chains of command and clear lines of
accountability to ensure that the right decisions are made, and the
Demos

99

Networks collection

4/1/04

4:35 PM

Page 100

Network logic

right people censured if they fail. Network leadership rejects that

model of authority, and the blame games it promotes.
But network leaders nonetheless carry responsibility, in particular
to preserve the trust on which their networks depend. In an
unpredictable world in which some failures are almost bound to
happen, that is a tough challenge. Acknowledging the depth of our
interdependence with others, and the limited capacity of our leaders
to manage it, will be a frightening experience. It is much more
convenient to think that leaders will be saviours and that we have
someone to blame when things do not go our way. But if it wakes us
up to the potential within each of us to solve our own problems, then
so much the better.
Paul Skidmore is a senior researcher at Demos.
Notes
1
2
3
4
5
6
7
8
9
10
11
12
13
14

100

See for example G Mulgan, Connexity: responsibility, freedom, business and

power in the new century (London: Vintage, 1998).
M Castells, Rise of the Network Society (Oxford: Blackwell, 1996); A Giddens,
The Third Way: the renewal of social democracy (Cambridge: Polity Press, 1998).
S Zuboff and J Maxmin, The Support Economy: why corporations are failing
individuals and the next stage of capitalism (London: Allen Lane, 2003).
C Leadbeater, T Bentley and J Wilsdon, The Adaptive State: strategies for
personalising the public realm (London: Demos, 2003).
R Heifetz, Leadership Without Easy Answers (Cambridge, MA: Belknapp Press,
1994).
M Leonard, Network Europe (London: Foreign Policy Centre, 1999).
T Veenkamp, T Bentley and A Buenfino, People Flow: managing migration in a
New European Commonwealth (London: Demos, 2003).
Global Business Network, What Next? Exploring the new terrain for business
(Cambridge, MA: Perseus, 2002).
I Nonaka and H Takeuchi, The Knowledge Creating Company (Oxford: Oxford
University Press, 1995).
D Rushkoff, Open Source Democracy (London: Demos, 2003).
D Chesterman with M Horne, Local Authority? (London: Demos, 2003).
G Mulgan, Communication and Control: networks and the new economies of
communication (Cambridge: Polity Press, 1991).
T Smit, Eden (Bantam Books, 2001), cited in J Chapman, System Failure
(London: Demos, 2002).
F Capra, The Hidden Connections (London: HarperCollins, 2002).

Demos

Networks collection

4/1/04

4:35 PM

Page 101

Leading between

15
16
17
18
19

M Horne, P Skidmore and J Holden, Learning Communities and NHSU

(London: Demos/NHSU, 2004).
Chesterman with Horne, Local Authority?
M Fullan cited in D Wilkinson and E Appelbee, Implementing Holistic
Government (Bristol: Demos/Policy Press, 1999).
Capra, Hidden Connections.
Cited in Zuboff and Maxmin, The Support Economy.

Demos

101

Networks collection

4/1/04

4:35 PM

Page 102

Networks collection

4/1/04

4:35 PM

Page 103

Network Logic 8

Disparities of
wealth appear as a
law of economic
life that emerges
naturally as an
organisational
feature of a
network . . .
The science of
inequality
Mark Buchanan

Networks collection

4/1/04

4:35 PM

Page 104

Networks collection

4/1/04

4:35 PM

Page 105

8. The science of
inequality
Mark Buchanan

Historically, economists have been obsessed with arguments over how

equitable or inequitable the distribution of wealth is among
individuals. This is perhaps the most controversial and inflammatory
of all economic topics. As economist John Kenneth Galbraith noted
in his history of the field: The explanation and rationalization of the
resulting inequality has commanded some of the greatest, or in any
case some of the most ingenious, talent in the economics profession.1
What is this inequality that Galbraith refers to? We all know, of
course, that a few people are very wealthy and that most of us have far
less. But the inequality of the distribution of wealth has a surprisingly
universal character. You might well expect the distribution to vary
widely from country to country, as each nation has its own distinct
political organisation and resources some nations relying on
agriculture, others on heavy industry while their peoples have
unique cultural expectations, habits and skills. However, towards the
end of the nineteenth century, an Italian engineer-turned-economist
named Vilfredo Pareto discovered a pattern in the distribution of
wealth that appears to be every bit as universal as the laws of
thermodynamics or chemistry.
Suppose that in Britain, China or the United States, or any other
country for that matter, you count the number of people worth say,
10,000. Suppose you then count the number worth 20,000, 30,000
and so on, covering many levels of wealth both large and small, and
Demos

105

Networks collection

4/1/04

4:35 PM

Page 106

Network logic

finally plot the results on a graph. You would find, as Pareto did, many
individuals at the poorer end of the scale and progressively fewer at
the wealthy end. This is hardly surprising. But Pareto discovered that
the numbers dwindle in a very special way: towards the wealthy end,
each time you double the amount of wealth, the number of people
falls by a constant factor.
Big deal? It is. Mathematically, a Pareto distribution of this form
has a notable characteristic, as it implies that a small fraction of the
wealthiest people always possesses a lions share of a countrys riches. It
could be the case that the bulk of humanity in the middle of
distribution was in possession of most of the wealth. But it isnt so. In
the United States something like 80 per cent of the wealth is held by
only 20 per cent of the people, and the numbers are similar in Chile,
Bolivia, Japan, South Africa or the nations of Western Europe. It
might be 10 per cent owning 90 per cent, 5 per cent owning 85 per
cent, or 3 per cent owning 96 per cent, but in all cases, wealth seems
to migrate naturally into the hands of the few. Indeed, although good
data is sadly lacking, studies in the mid-1970s based on interviews
with Soviet emigrants even suggested that wealth inequality in the
communist Soviet Union was then comparable to that of the UK.2
An underlying order?
What causes this striking regularity across nations? Does it simply
reflect the natural distribution of human talent? Or, is there some
devilish conspiracy among the rich? Not surprisingly, given the strong
emotions stirred by matters of wealth and its disparity, economists in
the past have, as Galbraith noted, flocked to such questions. Today,
these questions again seem quite timely, as, if anything, the degree of
inequity seems to be growing.
In the United States, according to economist Paul Krugman:
The standard of living of the poorest 10 percent of American
families is significantly lower today than it was a generation ago.
Families in the middle are, at best, slightly better off. Only the
wealthiest 20 percent of Americans have achieved income
106

Demos

Networks collection

4/1/04

4:35 PM

Page 107

The science of inequality

growth anything like the rates nearly everyone experienced

between the 1940s and early 1970s. Meanwhile, the income of
families high in the distribution has risen dramatically with
something like a doubling of real incomes of the top 1 percent.3
A similar story could be told for the United Kingdom and many other
countries, especially in Eastern Europe, over the past two decades.
Something similar is taking place on the global stage. Globalisation
is frequently touted especially by those with vested economic
interests such as multinational corporations and investment banks
as a process that will inevitably help the poor of the world. To be sure,
greater technological and economic integration on a global scale
certainly ought to have the potential to do so. Yet as Nobel Prizewinning economist and former chief economist of the World Bank
Joseph Stiglitz notes in his recent Globalization and its Discontents:
A growing divide between the haves and the have-nots has left
increasing numbers in the Third World in dire poverty, living on
less than a dollar a day. Despite repeated promises of poverty
reduction made over the last decade of the twentieth century, the
actual number of people living in poverty has actually increased
by almost 100 million. This occurred at the same time that total
world income actually increased by an average of 2.5 per cent
annually.4
What is the origin of these distinct but seemingly related trends?
Economists can, of course, offer a great many relevant observations.
Stiglitz condemns the international economic policies of the Western
nations and the International Monetary Fund as reflecting the needs
of special financial and commercial interests regardless of the damage
inflicted on the developing nations. One can point also to tax cuts for
the very wealthy (a general theme of the 1980s and 1990s, especially
in the US and UK), to changes in international markets, the influence
of new communication technologies and so on. These are all
legitimate and insightful points.
Demos

107

Networks collection

4/1/04

4:35 PM

Page 108

Network logic

But might there be a more general perspective a general science

that would illuminate the basic forces that lead to wealth inequity?
Conventional economic theory has never managed to explain the
origin of Paretos universal pattern. Ironically, however, a pair of
physicists, venturing across interdisciplinary lines, has recently done
so. To understand their thinking, forget for the moment about
personal ingenuity, intelligence, entrepreneurial skills and the other
factors that might clearly influence an individuals economic destiny.
Instead, take one step into the abstract and think of an economy as a
network of interacting people and focus on how wealth flows about
in this network.
Thats the way the money goes
Each of us has a certain amount of wealth and, over the days and
weeks, this amount changes in one of two fundamental ways. Your
employer pays you for your work; you buy groceries; you build a
fence to keep in the dog; you take a holiday in Tuscany. Transactions
of this sort form the bread-and-butter of our daily economic lives,
and serve to transfer wealth from one person to another along the
links in the network. This is one mechanism by which our wealth
goes up or down.
When wealth flows from one person to another, however, the total
amount doesnt change. And yet wealth can also be created or
destroyed. Say you purchased a house in the UK in 1995. Today, you
are probably happy to see that its value has skyrocketed in the recent
real-estate boom. Your total wealth has gone up. On the other hand,
in 1998, you may have invested some spare cash into the stock market
perhaps even buying shares in Enron or Worldcom and gone on
to read the daily newspapers with a sense of unfolding doom.
Investments lead both to the creation and the destruction of
wealth.
From this extremely simple perspective, then, two basic factors
control the dynamics of wealth. This is hardly contentious, and it may
also seem unworthy of consideration. Yet, as physicists Jean-Philippe
Bouchaud and Marc Mzard of the University of Paris have recently
108

Demos

Networks collection

4/1/04

4:35 PM

Page 109

The science of inequality

shown, the interplay of these two basic forces goes a long way to
determining how wealth is distributed.5
For a network of interacting individuals, Bouchaud and Mzard
formulated a set of equations that could follow wealth as it shifts from
person to person, and as each person receives random gains or losses
from his investments. They also included one further feature to reflect
the fact that the value of wealth is relative. A single parent trying to
work and raise her son might face near ruin over the loss of a 20
note; in contrast, a very rich person wouldnt flinch after losing a few
thousand. In other words, the value of a little more or less wealth
depends on how much one already has. This implies that when it
comes to investing, wealthy people will tend to invest proportionally
more than the less wealthy.
The equations that capture these basic economic processes are
quite simple. However, there is a catch. For a network of many people
say, 1,000 or more the number of equations is similarly large. For
this reason, a model of this sort lies well beyond anyones
mathematical abilities to solve (and this explains why it has not
appeared in conventional economics). But the philosopher Daniel
Dennett has for good reason called digital computers the most
important epistemological advance in scientific method since the
invention of accurate timekeeping devices and Bouchauds and
Mzards work falls into a rapidly growing area known as
computational economics which exploits the computer to discover
principles of economics that one might never identify otherwise.
Bouchaud and Mzard explored their model in an exhaustive series
of simulations. And in every run they found the same result after
wealth flows around the network for some time, it falls into a steady
pattern in which the basic shape of wealth distribution follows the
form discovered by Pareto. Indeed, this happens even when every
person starts with exactly the same amount of money and moneymaking skills. This pattern appears to emerge as a balance between
two competing tendencies.
On the one hand, transactions between people tend to spread
wealth around. If one person becomes terrifically wealthy, he or she
Demos

109

Networks collection

4/1/04

4:35 PM

Page 110

Network logic

may start businesses, build houses and consume more products, and
in each case wealth will tend to flow out to others in the network.
Likewise, if one person becomes terrifically poor, less wealth will flow
through links going away from him, as he or she will tend to purchase
fewer products. Overall, the flow of funds along links in the network
should act to wash away wealth disparities.
But it seems that this washing-out effect never manages to gain the
upper hand, for the random returns on investment drive a counterbalancing rich-get-richer phenomenon. Even if everyone starts out
equally, and all remain equally adept at choosing investments,
differences in investment luck will cause some people to accumulate
more wealth than others. Those who are lucky will tend to invest
more and so have a chance to make greater gains still. Hence, a string
of positive returns builds a persons wealth not merely by addition but
by multiplication, as each subsequent gain grows ever bigger. This is
enough, even in a world of equals where returns on investment are
entirely random, to stir up huge wealth disparities in the population.
This finding suggests that the basic inequality in wealth
distribution seen in most societies and globally as well, among
nations may have little to do with differences in the backgrounds
and talents of individuals or countries. Rather, the disparity appears
as a law of economic life that emerges naturally as an organisational
feature of a network. This finding suggests that the temptation to find
complex explanations behind the distribution of wealth may be
seriously misguided.
Altering inequality
However, this does not imply that there is no possibility for
mitigating inequities in wealth. There is some further subtlety to the
picture. From an empirical point of view, Pareto found (as many
other researchers have found later) that the basic mathematical form
of the wealth distribution is the same in all countries. One always
finds that each time you double the amount of wealth, the number of
people having that much falls by a constant factor. This is the pattern
that always leads to a small fraction of the wealthy possessing a large
110

Demos

Networks collection

4/1/04

4:35 PM

Page 111

The science of inequality

fraction of everything. Nevertheless, the constant factor can be

somewhat different from one case to another. The degree of inequity
can vary from country to country, and, socially speaking, theres a
huge difference between the richest 5 per cent owning 40 per cent of
the wealth or their owning 95 per cent.
An additional strength of Bouchauds and Mzards network
model is that it shows how the degree of inequity in an economy can
be altered. They found two general rules. First, the greater the volume
of wealth flowing through the economy the greater the vigour of
trading, if you will then the greater the equality. Conversely, the
more volatile the investment returns, the greater the inequity. This
has some curious practical implications some obvious and some
not so obvious.
Take taxes, for instance. The model confirms the assumption that
income taxes will tend to erode differences in wealth, as long as those
taxes are redistributed across the society in a more or less equal way.
After all, taxation represents the artificial addition of some extra
transactional links into the network, along which wealth can flow
from the rich towards the poor. Similarly, a rise in capital gains taxes
will also tend to ameliorate wealth disparities, both by discouraging
speculation and by decreasing the returns from it. On the other hand,
the model suggests that sales taxes, even those targeted at luxury
goods, might well exaggerate differences in wealth by leading to fewer
sales (thus reducing the number of transactional links) and
encouraging people to invest more of their money.
The model also offers an excellent test of some arguments that
politicians use to justify policies. In Britain and in the United States,
for example, the 1980s and 1990s were dominated by free-market
ideology, much of it defended by the idea that wealth would trickle
down to the poor. Everything was done to encourage investment
activity, regardless of the risks involved. This was the era of junk
bonds, the savings and loan debacle, and the dot-com boom, now
capped off by the Enron-led wave of corporate scandals. As we know,
the wealth did not trickle down and the distribution of wealth in both
countries is today significantly less equitable than it was three decades
Demos

111

Networks collection

4/1/04

4:35 PM

Page 112

Network logic

ago. Under the network model, this is just what one would expect a
dramatic increase in investment activity, unmatched by measures to
boost the flow of funds between people, ought to kick up an increase
in wealth inequality. (Indeed, taxes were also generally lowered during
this era, thus removing some of the links that could have helped to
redistribute wealth.)
What about globalisation? From the perspective of this model,
international trade should offer a means to create a better balance
between the richer and poorer nations. Leaving aside legitimate
concerns over a lack of environmental regulations, protection for
child labourers and so on, Western corporations setting up manufacturing plants in developing nations and exporting their computing
and accounting to places like India and the Philippines should help
wealth flow into these countries. In some cases, this promise of
globalisation has been realised. But, in view of the potential benefits,
it is easy to understand the anger of the poorer nations at measures
designed to skew the trading network in favour of the richer
countries. As Stiglitz comments:
The critics of globalization accuse Western countries of
hypocrisy, and the critics are right. The Western countries have
pushed poor countries to eliminate trade barriers, but kept up
their own barriers, preventing developing countries from
exporting their agricultural products and so depriving them of
desperately needed export incomeThe West has driven the
globalization agenda, ensuring that it garners a
disproportionate share of the benefits, at the expense of the
developing world.6
As Bouchauds and Mzards model illustrates, free trade could be a
good thing for everyone, but only if it enables wealth to flow in both
directions without bias.
But lets go back to the model, for it also reveals another rather
alarming prospect. In further investigations, Bouchaud and Mzard
found that if the volatility of investment returns becomes sufficiently
112

Demos

Networks collection

4/1/04

4:35 PM

Page 113

The science of inequality

great, the wealth differences it churns up can completely overwhelm

the natural diffusion of wealth generated by transactions. In such a
case, an economy whether within one nation, or more globally
can undergo a transition wherein its wealth, instead of being held by a
small minority, condenses into the pockets of a mere handful of
super-rich robber barons.
It is intriguing to wonder if some countries, particularly
developing nations, may already be in this state. It has been estimated,
for example, that the richest 40 people in Mexico have nearly 30 per
cent of the wealth. It could be, also, that many societies went through
this phase in the past. Long-term economic trends during the
twentieth century lend some credence to this idea, as the total share of
the richest individuals in England, for example, has fallen over the last
century.
In Russia, following the collapse of the USSR, wealth has become
spectacularly concentrated; inequality there is dramatically higher
than in any country in the West. The model would suggest that both
increased investment volatility and lack of opportunities for wealth
redistribution might be at work. In the social vacuum created by the
end of the Soviet era, economic activity is less restricted than in the
West, as there are few regulations to protect the environment or to
provide safety for workers. This not only leads to pollution and
human exploitation but also generates extraordinary profits for a few
companies (the politically well connected, especially a popular pun
in Russia equates privatisation with the grabbing of state assets).
Economists have also pointed out that Russia has been slow to
implement income taxes that would help to redistribute wealth.
This simple model is, of course, not the final word in explaining
the distribution of wealth or how best to manage it. But it does offer a
few basic lessons. Although wealth inequity may indeed be inevitable
to a certain extent, its degree can be adjusted. With proper regulation
to protect the environment and workers rights, free trade and
globalisation should be forces for good, offering better economic
opportunities for all. But we will achieve such happier ends only if
global integration is carried out sensibly, carefully and, most of all,
Demos

113

Networks collection

4/1/04

4:35 PM

Page 114

Network logic

honestly. If it is not, and if the disparity between the haves and havenots continues to grow, then one might expect countervailing social
forces to be stirred up, as they have throughout history.
By starting with remarkably simple assumptions and studying the
patterns that emerge in a network of interacting agents, Bouchaud
and Mzard have gained an important insight into one of the most
basic and contentious patterns of economic life. Unfortunately,
their model by itself cannot help us make wise use of this insight.
Mark Buchanan is a science writer. His most recent book is Small
World: uncovering natures hidden networks. A version of this essay
first appeared in New Statesman.
Notes
1
2
3
4
5
6

114

J K Galbraith, A History of Economics (London: Penguin, 1991).

J Flemming and J Micklewright, Income distribution, economic systems and
transition, Innocenti Occasional Papers, Economic and Social Policy Series, no
70 (Florence: UNICEF International Child Development Centre, 1999).
P Krugman, An unequal exchange in The Accidental Theorist (New York:
Norton, 1998).
J Stiglitz, Globalization and its Discontents (London: Allen Lane, 2002).
J-P Bouchaud and M Mzard, Wealth condensation in a simple model of
economy, Physica A, 282 (2000).
Stiglitz, Globalization and its Discontents.

Demos

Networks collection

4/1/04

4:35 PM

Page 115

Network Logic 9

Womens networks
represent a force
for change and
social agency with
the potential to
tackle persistent
workplace
inequalities . . .
Old boys and new girls
Helen McCarthy

Networks collection

4/1/04

4:35 PM

Page 116

Networks collection

4/1/04

4:35 PM

Page 117

9. Old boys and new girls

womens networks and diversity in
the workplace
Helen McCarthy

Hightech-Women meets on the first Wednesday of every month at

the Womens University Club, located in an elegant nineteenthcentury brick building in Mayfair. It is an unusual mix of capsule
wardrobes, mobile phones, wood-panelled walls and floral-patterned
carpets. And yet, the union of these two womens organisations with
very different styles could not be more fitting. The club was
established in 1886 for the growing numbers of women entering
higher education and the professions; today, it also provides a
meeting place for a network of ambitious women in technologyrelated occupations, first brought together by venture capitalist Lucy
Marcus in 2000. Hightech-Womens members might be very different
from their Victorian and Edwardian forebears, but they exhibit every
bit as much appetite for pooling their collective resources in a much
changed but still unequal world. Or, for what an early advocate of the
womens networking movement simply describes as getting together
to get ahead.1
Hightech-Women is one of a growing movement of networks
creating professional communities of women across many sectors. Yet
this phenomenon is little understood and sparsely documented. This
essay, based on emerging findings from a Demos research project, sets
out to understand how womens networks function and to explore the
contribution they might make to equality and diversity goals in a socalled post-feminist era.
Demos

117

Networks collection

4/1/04

4:35 PM

Page 118

Network logic

A brief history of womens networks

Women in Britain have associated in various ways for many centuries,
often forming strong personal ties through the relationships that
characterise womens traditional, domestically oriented roles. These
sorts of relationships between women still exist today to varying
degrees. However, what is analytically distinctive about the peer-topeer networking activities described in this essay is the way in which
they give a formal expression and visibility to previously informal and
loosely organised relationships, and focus them around a gendered
professional identity. Womens networks thus can complicate
professional identities by constructing alternative narratives of and
generating new perspectives about organisations. In doing so, they
represent a force for organisational change and a form of social
agency with the potential to tackle persistent workplace inequalities
between women and men.
Communities of professional women began to grow at the end of
the nineteenth century when various membership associations and
institutions were founded for the rising number of women entering
Britains universities and professions. However, it was not until the
1970s, the period in which sexual politics were put firmly back on the
political agenda by the Womens Liberation Movement, that the
dynamic peer-to-peer networking model recognisable among
professional women today began to take shape. Small groups of
businesswomen began to meet together in cities in the UK and the
US, initially through low-key gatherings for breakfast or lunch. These
grew rapidly into large corporate networks based within companies,
or external business networks with a nationwide reach.2
These early networks set the trend for those that followed. The
activities ranged from purely social gatherings to workshops,
seminars, mentoring programmes and opportunities for voluntary
work, supplemented with access to general advice, information and
mutual support through the network. This model was adopted in the
UK by women in professions beyond the business world steadily
throughout the 1980s. However, over the last five to ten years, Britain
has witnessed what might be understood as a new wave of the
118

Demos

Networks collection

4/1/04

4:35 PM

Page 119

Old boys and new girls

womens networking movement. This coincided with indeed

constituted part of a new self-confidence experienced by women,
especially among young professional women, in the 1990s. This
decade witnessed growing evidence of a generational shift in attitudes
towards more flexible and fluid gender roles, as well as the disruption
of traditional feminist narratives of womens collective consciousness
and political mobilisation.3 The new girl networks, as they were
light-heartedly described, reflected elements of this new gender
politics. Networks bubbled up in sectors as diverse as technology
(DigitalEve), journalism (Women in Journalism), small business
(Women into the Network) and corporate social responsibility (CSR
Chicks). Many of these have youthful memberships, are highly
proactive in marketing themselves externally with sophisticated
brand identities, funky websites and catchy names, and make effective
use of the internet as a communications and community-building
tool.
Over the same period, large employers have begun to explore
networks as an alternative to conventional equal opportunities and
centrally owned diversity initiatives, the limitations of which are
becoming more widely understood. Corporate networks provide
opportunities for women in organisations to lead their own
professional development and contribute to the strengthening of
gender diversity in the workplace. Lloyds TSB, GE, Citigroup and
Shell have all launched corporate womens networks in the last five
years, and diversity networks are a growing presence too within the
civil service and the wider public sector. These internal networks not
only connect women within the organisation, but provide important
advice and intelligence to the management on womens needs, and in
many cases help to design and implement diversity policies.
Selection of UK womens networks
Women in Management 1969
City Womens Network 1978
Women in Business and Finance 1980

Demos

119

Networks collection

4/1/04

4:35 PM

Page 120

Network logic

Women in Medicine 1980

Network 1981
Women in BP 1982
Women in Dentistry 1985
BT Womens Network 1986
Women Chemists Network 1988
Through the Glass Ceiling 1990
Women in Journalism 1995
Berwin Leighton Paisner/The Adelaide Group 1997
Everywoman 1999
CSR Chicks 1999
Lloyds TSB Womens Network 1999
Busygirl 2000 (renamed Aurora Womens Network in 2003)
Hightech-Women 2000
DigitalEve 2000
Financial Mail Womens Forum 2001
Citiwomen (Citigroup) 2002
Thinkingwomen 2002
Senior Civil Service Womens Network 2001
Shell Womens Network UK 2003
BT Senior Womens Network 2003
Dynamic Asian Womens Network (DAWN) 2003

New opportunities and old inequalities

So what accounts for this current flurry of interest around networks?
It is undoubtedly in part a reflection of the greater clout that women
wield in their working lives. The rising numbers of corporate
networks suggests growing recognition among employers of the case
for developing womens talent as a core business strategy. This
perspective is particularly well advanced in the US, where research
shows how corporate womens networks have helped large companies
to decrease turnover and increase productivity as well as enhance
equality and diversity in their organisations.4
120

Demos

Networks collection

4/1/04

4:35 PM

Page 121

Old boys and new girls

Nevertheless, womens networks are also a symptom of another,

less celebratory aspect of womens working lives. Alongside the
unprecedented opportunities now available to them, women continue
to experience gender-related disadvantage in their status in the
workforce. For example, women working full-time earn on average
18.8 per cent less than the average hourly earnings of male full-time
employees. They make up just 9 per cent of directors in FTSE 100
companies, 13 per cent of small business owners and 23 per cent of
top managers in the civil service,5 yet account for 79 per cent of
administrative and secretarial occupations.6 At the same time women
continue to work a second shift by taking on the lions share of
domestic work and childcare.7
In their modus operandi womens networks hold a mirror up to this
phenomenon of growing opportunity and persisting inequality.
While celebrating womens achievements and advancements, many,
either explicitly or implicitly, hold womens continuing underrepresentation and lack of progression in their chosen field as major
points of reference. Underlying this position is the normative
assumption that women are disadvantaged relative to men in building
successful careers, and this basic narrative of inequality is what thus
justifies the existence of women-centred initiatives, such as networks.
To this extent, womens networks might be accurately described as
feminist organisations, although few would choose to describe their
activities in these terms. Indeed, this distancing from the ideological
legacy of second-wave feminism may well be one of the greatest
strengths of womens networks, in that they provide a model of
affiliation and mutual support which has wide appeal, especially to
younger women.
Old boys and new girls
A concern with the position of women in the labour market is not
unique to womens networks, and is shared with any number of
policy-makers, academics and pressure groups working in this field. It
remains to be asked then: what is special or distinctive about peer-topeer networks as an organisational form for achieving change in this
Demos

121

Networks collection

4/1/04

4:35 PM

Page 122

Network logic

area? Or, to put it another way, why do networks matter when it

comes to inequality?
The answer lies in the role of networks in distributing power,
resources and opportunity across a wider system. Historically,
patterns of social connectivity have been instrumental in the
production and reproduction of gender disadvantage, an idea most
commonly expressed in the phrase the old boy network (OBN).
When first coined in the mid-nineteenth century, this phrase referred
specifically to the alumni of an elite educational institution, whose
shared educational backgrounds and common affiliations naturally
translated into an informal system of favours and mutual support
throughout later life.
Although these types of networks still exist and retain some
influence in certain quarters, as a concept, the OBN has gained a
wider meaning as familiar shorthand for all forms of male power in
public life and the workplace. Today, the patterns of social
connectivity among men that are of most concern for equality in
Britains workplaces are less likely to be tied to the public school or
the gentlemans club than ever before. Instead, they tend to reflect
embedded forms of sociability that do not formally exclude women
but, combined with unequal caring roles in the home and womens
often asymmetrical working patterns, contribute to the concentration
of power in male hands. In other words, these processes often create
organisational cultures that value and reward qualities or behaviours
that women are less likely to exhibit than men; that require, as US law
professor Susan Estrich puts it, womens adaptation to a male
comfort zone.8
In this sense, the new girl networks are not mirror images of their
male counterparts, as is sometimes supposed. While many supporters
believe that women can and should employ some of the same tactics
as the OBN, particularly with regard to using ones contacts
instrumentally, few networks aspire to achieving the same hegemony
for women that men have traditionally enjoyed in the workplace.
Womens networks have transparent structures and formal status, and
exist primarily for the purpose of supporting womens networking
122

Demos

Networks collection

4/1/04

4:35 PM

Page 123

Old boys and new girls

practices. In contrast, the institutions that nurture social ties among

men do not tend to hold networking as their formal or explicit aim.
Furthermore, womens networks should be understood as being
fundamentally concerned with altering social relations, rather than
preserving them. Just as it is useful to view patterns of social
connectivity among men through the filter of power relations, it is
also helpful to see womens networks as a disruptive force impacting
upon those relations. By actively creating opportunities for women to
extend their personal and professional networks, womens networks
alter the dynamics of the patterns of social ties established by men,
and create alternative spaces in which male norms and behaviours
can be challenged. Directed in the right way, womens networks thus
have the potential to create more inclusive working environments for
all. For example, the GE Womens Network allows men to attend
meetings as a way of drawing them into conversation about the
dynamics of gender in the workplace and how managers can treat
employees of both sexes fairly. In its early years of existence, the
Lloyds TSB Womens Network developed a set of professional
development tools for its members, which were eventually adopted by
Human Resources and taken into the mainstream of the core
programme offered to all employees.
The missing link in equality strategies?
These insights have much to offer the current debate on the future of
gender equality. For much of the last 30 years, the legislative and
policy instruments of the state have been regarded as the most
effective way to protect individuals from discrimination and to
compensate for historical disparities of power. Over the past decade
this framework has been challenged by the business case for equality
an approach that attempts to link equality goals with workplace
diversity and business objectives. Thus business imperatives, in
theory at least, remove the need for coercion or heavy regulation by
the state. While the business case approach is widely recognised by
policy-makers and employers, concerned voices from the equality
lobby point to slow rates of progress and the failure of soft tools to
Demos

123

Networks collection

4/1/04

4:35 PM

Page 124

Network logic

embed equality and diversity perspectives throughout organisational

cultures and working practices. As a result, today the debate often
becomes trapped within an unhelpful dichotomy of compulsion
versus voluntarism.
The major problem with this dichotomy is that it cannot
accommodate any approach to equality that doesnt take either the
state or the market as its starting point. Yet policy-makers are
increasingly recognising that long-term problems cannot be
addressed using the traditional instruments available to them, or by a
blind faith in market forces. Instead, as Bentley and Wilsdon have
recently argued, they require changing the dominant assumptions
governing models of organisation, and the development of greater
adaptive capacity at every level of the system.9 In this emerging
framework, gender inequality should be understood as a problem
involving complex, dynamic processes that cannot be easily tackled by
reference to existing models or off-the-peg solutions. Instead, it
requires open-ended capacity-building efforts, involving multiple
stakeholders, across the wider system, which produces and reproduces
gender inequalities. Womens networks might not provide the whole
picture here, but they offer some important clues as to what an
adaptive approach towards gender equality could look like.
An approach that is participatory

Equality cannot be gifted to individuals or groups by the state using

a top-down model, nor can it be bestowed by diversity initiatives
driven by employers. These approaches are least likely to win
widespread legitimacy and most likely to attract backlash and charges
of tokenism. In contrast, the starting point for womens networks is
women themselves. Whether corporate or sector-wide, networks rely
on the active participation of their members to succeed. They
facilitate access to a supportive and enabling community, but
individuals have the ultimate responsibility for converting the ties
they make through the network into concrete opportunities and
outcomes. Womens success is thus generated by womens own
agency.
124

Demos

Networks collection

4/1/04

4:35 PM

Page 125

Old boys and new girls

An approach that is self-organising

In the same way that equality and diversity initiatives that are
imposed in a top-down manner are unlikely to result in whole system
change, womens networks cannot be prescribed by managers. The
first corporate networks were created on the initiative of individual
women who perceived a need, and succeeded because there was a
groundswell of women within the company who shared that view.
The same principle applies to networks outside the workplace, where
the range and scope of activities tend to be member-led, often
evolving through series of experiments and listening exercises. This
self-organising quality gives networks their dynamic quality, but it
may also explain why some networks do not survive.
An approach that accommodates diversity

The fragmentation of the womens movement during the 1980s is

often attributed to the challenge of diversity. Womens identities and
experiences, it was felt by many, were far too complex and diverse to
sustain any notion of collective interests, and the search for consensus
thus became an illegitimate political project. Today, womens
networks offer an alternative model of solidarity, which balances the
interests of individuals (What can I get out of this? How can I
develop as a person?) with the desire for reciprocity, mutual support
and, if appropriate, collective action (How can I help others? What
can we achieve together?). Womens networks are highly flexible,
allowing members to pick and choose how and when they connect
with each other. Many women are members of several networks
simultaneously, or switch from one to another as their needs change.
While networks create shared identities among women, they do not
demand political consensus.
An approach that operates on several levels simultaneously

Gender inequality cannot be tackled by breaking the problem down

into smaller components and treating each one separately. Rather, a
fabric of interrelated factors must be addressed at the same time.
Womens networks operate on at least three levels. First, they meet the
Demos

125

Networks collection

4/5/04

3:51 PM

Page 126

Network logic

needs of individuals for professional development and community.

Second, by developing womens talent, they contribute to diversity
within workplaces and promote the progression of women across
sectors and, through this, womens networks have the potential to
strengthen the overall performance of an organisation or sector.
Emerging questions
This initial exploration of womens networks points to a cluster of
questions that warrant further consideration. The first centres on the
processes of change within organisations, and the sort of
contribution that diversity networks, with their dynamic, peer-topeer structures, can make to learning and managing knowledge,
capacity-building and developing trust within communities of
practice. The second focuses on the interface between womens
networks and other social institutions. Viewed through the lens of
their campaigning and voluntary work and, in many cases, their
strong grassroots presence in local communities, to what extent are
womens networks agents of social capital and civic participation?
And what do they tell us about the role of gender in communitybuilding processes? The final set of questions concerns the nature of
the policy challenge: how can we identify a role for policy-makers in
supporting and nurturing womens networks as part of a wider,
adaptive approach to gender equality? As such, this essay represents
an overture to a much richer, ongoing conversation about the
contribution of diversity networks in creating inclusive workplaces
for all.
Helen McCarthy is a researcher at Demos, and is co-founder of womens
network thinkingwomen. She is also author of the forthcoming Demos
publication Girlfriends in High Places, which will be published in April
2004.
Notes
1

126

M Scott Welch, Networking: the great new way for women to get ahead (New
York: Harcourt Brace Jovanovich, 1980).

Demos

Networks collection

4/5/04

3:51 PM

Page 127

Old boys and new girls

2
3
4
5
6
7
8
9

For an overview of these developments, see Scott Welch, Networking.

H Wilkinson, M Howard and S Gregory, Tomorrows Women (London: Demos,
1997); H Wilkinson, No Turning Back: generations and the genderquake
(London: Demos, 1995).
Catalyst Women in Business, Creating Womens Networks: a how-to guide for
women and companies (San Francisco, CA: Jossey-Bass, 1998).
Equal Opportunities Commission, Sex and Power: who runs Britain?
(Manchester: EOC, 2004).
Equal Opportunities Commission, Facts about Women and Men in Great Britain
(Manchester: EOC, 2003).
A Jones, About Time for Change (London: Work Foundation, 2003).
S Estrich, Sex and Power (New York: Riverhead Books, 2000).
T Bentley and J Wilsdon, Introduction in The Adaptive State (London: Demos,
2003).

Demos

127

Networks collection

4/1/04

4:35 PM

Page 128

Networks collection

4/1/04

4:35 PM

Page 129

Network Logic 10

Almost every
aspect of life that
citizens care about
is affected by the
patterns, nature
and distribution of
social ties between
people . . .
Your friendship
networks
Dr Perri 6

Networks collection

4/1/04

4:35 PM

Page 130

Networks collection

4/1/04

4:35 PM

Page 131

10. Your friendship

networks
are they any of the governments
business?
Dr Perri 6
Networks of friendship and acquaintance among citizens matter to
government.1 Almost every aspect of life that citizens care about and
want government to tackle is affected by the patterns, the nature and
the distribution of social ties between people. Your chances of
catching the common cold as well as many other aspects of health
status are significantly affected by the extent of your social support.
Getting out of unemployment is most often achieved using informal
ties to find work. The pathways into crime are best traced along the
connections young people have to those already involved in criminal
activities. Educational attainment is hugely affected by the culture of
attainment among your peers. We get through the grief of
bereavement better for being supported. Remaining independent into
old age rests greatly on being able to draw upon friends and
neighbours and not just on close relatives.
So it is hardly surprising that policy-makers are interested. But can
government do anything deliberately to influence our patterns of
friendship and acquaintance? And indeed, should it, or is a step too
far towards the intrusive, authoritarian state? Should friendship and
acquaintance be off-limits to policy-makers?
Government shapes whom we meet
Certainly, almost everything that government does has an
unavoidable impact on our personal social networks. Housing design,
Demos

131

Networks collection

4/1/04

4:35 PM

Page 132

Network logic

slum clearance and transport policy bring some people together and
keep others apart, and make it easier or harder for them to reach each
other. The now 50-year-long debate about tower blocks and
communities is really a debate about the ways in which government
shapes patterns of friendship and acquaintance. Education famously
creates ties between pupils and students that can sometimes last for a
lifetime. Whether social services are provided in ways that bring
people with similar problems together (think of special day centres
for people with mental health problems, or lunch clubs for older
people) or whether they are organised around providing services to
people individually, these decisions greatly affect the chances of
forming and sustaining certain types of bonds. When government
offers job clubs and special training programmes to unemployed
people to help them seek work, they tend to meet mainly other
unemployed people, who may be the least useful to them in seeking
work by informal means. Nonetheless, these services can significantly
affect whom users get the chance to meet.
So it is hard to see how government could do other than have a
huge effect on our social networks. Even the minimal or nightwatchman state advocated by neoliberals would have a huge impact,
both in the process of dismantling the apparatus of civil government, and in the ways in which people would have to adapt. As
Polanyi argued 60 years ago, free markets are only ever created by
government action, which itself brings about massive change to social
networks.2
Networks as an objective of policy?
Public services inevitably and vastly influence our networks in
unintended ways. But may government legitimately, and can it
feasibly, deliberately and directly pursue specific policy goals to
influence friendship and acquaintance? Or, in other words, does
government do better or worse, and does it violate fewer rights or
more, when it tries consciously and with care to achieve something
that it will affect massively in any case?
These are questions that ought to be addressed seriously before
132

Demos

Networks collection

4/1/04

4:35 PM

Page 133

Your friendship networks

governments rush to develop social capital building programmes.

One key problem is that all the good things in social life do not go
together, and different types of networks generate different sorts of
outcomes, which must be better understood. For example, the kinds
of social networks that conduce to thriving in the labour market tend
to be quite open, stressing acquaintance with people in situations
different from ones own. By contrast, the networks that help people
in later life tend to be more bounded, dense networks where all those
who provide support know each other and share linked lives. In
another context, the network forms which characterise neighbourhood renewal efforts are often marked by dense ties among the
residents of a tightly defined neighbourhood. These may well not be
conducive to the kinds of local economic development processes that
are most important for improving an areas connections with the
surrounding travel-to-work area. If governments are to try to
influence their citizens friendships and acquaintances deliberately,
they must first of all acknowledge the need for trade-offs between
different types of networks. Then policy-makers have the choice of
focusing on the special cases where only one type of network is
beneficial, or else on trying to balance different forms.
Network types
It is helpful to offer a framework for classifying the basic types of
social networks, so that we can at least identify the elements between
which trade-offs might have to be struck. It has been argued that
there are four basic types of networks (see Figure 1):3
Individualistic networks, which are sparse but open, allow for the
kinds of entrepreneurial and instrumental use of ties that link one to
people very different from oneself. This, then, is the freewheeling
world of the promotion-hungry networker. Enclaved networks, by
contrast, are dense but strongly bounded and tend to reinforce ties to
those similarly situated. Here is where we find the mutual support
clubs and some inward-looking communities.
Hierarchical networks are also bounded, but link people with very
different powers and resources in more rule-bound ways. This
Demos

133

Networks collection

4/1/04

4:35 PM

Page 134

Network logic

describes the kinds of ties that link people to those formally allocated
to mentor or counsel them, or the informal ties that link many chairs
of tenants associations with the town hall professionals who make
decisions about investments in their housing estates.
Isolate networks, the final type, are not necessarily those in which
people know literally no other people, but rather the mix of sparse
and casual ties to others with a few very close ties perhaps to
immediate family members, but which admit of very little reliable
support beyond immediate needs and afford little scope for collective
action.
Each of these network types has its strengths and its weaknesses.
Individualism is useful in many labour market situations; hierarchical
networks are valuable in some educational settings; and enclaves can
be very supportive for people who find themselves rejected by
mainstream institutions. Even the isolate form has its uses, for it
provides a way of coping during adversity.
Do governments know how to make a difference?
Assuming (a very big if) that governments can know better than
citizens themselves what network forms ought to be promoted, what
tools could they deploy through public services to cultivate among
citizens some beneficial mix of these types? And how could these
tools be deployed without violating rights such as liberty and privacy?
Past measures used in various public services to influence social
networks have a mixed record of success. Comprehensive schooling
and mixed tenure schemes are examples of interventions that have
not been terribly effective in promoting the kinds of social network
structures that policy-makers have hoped from them,4 although it is
possible that they might be more efficacious when used in social
contexts which are initially more communitarian in their institutions.
The evidence (to the extent that we have been able to interpret it) is
equivocal regarding the efficacy of excluding pupils from schools,
funding voluntary organisations and setting up buddying schemes in
influencing social networks in Western countries. This may be either
because the effects may be modest, because the effects do not last, or
134

Demos

Networks collection

4/1/04

4:35 PM

Page 135

Your friendship networks

Figure 1 Network signatures of the basic institutional

forms of social organisation
Social regulation
ISOLATE

Acquainted with
each other, share
general sentiment
Strong tie: kin, close
or old friend, etc.

INDIVIDUALISM

Act together:
coalitions of the
willing
Know each other

HIERARCHY

Weaker, sparser bilateral

ties among subalterns
Denser, stronger
multilateral ties between
high status individuals
Social
integration
ENCLAVE

Act together
Boundary of general
sentiment, cooperation
and shared identity

The network forms in the top half of this diagram exhibit more bilateral
ties, with those in the bottom half exhibiting more multilateral ties.
Similarly, those to the left involve a lower ongoing mutual dependence
for material resources and support, with those to the right involving a
higher mutual dependence.

Demos

135

Networks collection

4/1/04

4:35 PM

Page 136

Network logic

because the intervention may provoke significant counterorganisation towards other network forms. The impact of other types
of government action, such as life skills training and job clubs, we
simply do not know. Figure 2 shows the distribution of a range of
initiatives identified by the kind of solidarity that they might have
promoted if they were effective, and what we currently know about
their actual effectiveness.5
The evidence available in the literature6 suggests that, so far at
least, public services have yet to develop very sophisticated tools on
which to build any grand strategy for deliberate network shaping. The
evaluative literature is very thin indeed. It hardly considers the
interaction effects of the combinations of multiple measures as they
affect the same groups of people and is weak in examining
unintended consequences. In addition, it does not really examine the
extent to which privacy concerns are being respected or the extent to
which professionals are using these tools to gain greater discretionary
power. Some evaluative instruments have been developed, especially
in the field of care for frail older people (by Clare Wenger and her
collaborators7), that attempt to capture the impact of services upon
clients social networks, but they are still not being widely used.
Very often people advocate fashionable measures for which the
evidence of sustained impact on the social networks of their clients is
largely missing. Robert Putnam has famously argued for much more
generous public subsidy for voluntary organisations in the belief that
they will conduce to social capital by which he means almost any
kind of network other than the isolate form; he does not seem to
accept that there are incompatibilities between these network forms
that require trade-offs and even tough choices between them.8
However, there is really very little evidence that the social networks of
clients of voluntary bodies are influenced in any lasting way by using
their services, and the few studies that have been conducted actually
suggest that, if there are effects, they are short-lived.9 Even religious
bodies, said by some to be better at stimulating ties, actually turn out
in the few studies done to be no more impressive than comparable
government services.10
136

Demos

Networks collection

4/1/04

4:35 PM

Page 137

Your friendship networks

Figure 2 Interventions by solidarities they are

designed to promote
Social regulation
HIERARCHY

ISOLATE

Segregation of
individual
offenders within
institutions
School exclusion
without
readmission to
another school

?
?

Counselling
Life skills training
Mentoring
Funding voluntary organisations
Circles of support
Designing out crime, e.g. closing
streets to block criminals escape
routes
City centre pedestrianisation; zoning
for street cafs and clubs
School exclusion into speciality units
Social
integration
ENCLAVE

INDIVIDUALISM
?

*
*

Life skills training

Entrepreneurs
networks
Business
borrowers clubs
Comprehensive
schooling
Mixed tenure
schemes

?
?
*?

Segregated allocation of social rented

housing
Buddying schemes
Befriending schemes
Funding voluntary organisations
Drop-in centres
Job clubs
Communal spaces for residents in
defined small areas
City centre pedestrianisation; zoning
for street cafs and clubs
Creating local non-profit
organisations

Interventions which the evidence suggests have not been effective in

promoting the kinds of social network structures that policy-makers had
hoped from them.
Interventions where the evidence about their efficacy in changing social
networks is equivocal. Effects may not last, may be modest, or the
intervention may provoke counter-organisation towards other solidarities.
Interventions that may or may not in practice conduce to that solidarity.

Demos

137

Networks collection

4/1/04

4:35 PM

Page 138

Network logic

Certainly, if we are interested in the promotion of ties between the

worse off and the better off which is important in stimulating
labour market mobility then there is no evidence that voluntary
bodies are any more effective in this than are comparable public
bodies. Furthermore, social services agencies are less effective than
education bodies, irrespective of sector. While membership of
voluntary bodies may be associated with attitudes such as willingness
to trust others,11 it is far from clear that membership causes any
change in attitudes: associations may well typically recruit people
who are already readier to trust others.
Interestingly, the evaluative studies do suggest that the most
effective interventions are the least direct and the least coercive. For
example, support for voluntary social networks of local small
business entrepreneurs has been found effective in stimulating the
kinds of typically individualistic networks that can help in local
economic development. In addition, providing life skills training to
those least able to form friendships and acquaintances for various
reasons appears often to be capable of making a significant difference
at the individual level. Least effective in influencing social networks in
any lasting way appear to be the many interventions whereby public
services provide an artificial friend such as a mentor, or a specially
created group (these measures may have other merits, of course).
Physical measures to change architecture and town layout tend to be
effective mainly in the negative sense that certain kinds of crime can
be made more difficult, and certain spaces more easily surveyed by
residents. However, their effects on actual social networks seem not to
be very great, and least significant in respect of forming new ties that
would not otherwise have been created.
The policy challenge
Moreover, implementing such measures can be very challenging. The
skills required are not always available. Indeed, some public service
professions have been reformed in recent years precisely to shift their
work away from influencing patterns of social networks. Social work
has increasingly focused on practical support for older people and
138

Demos

Networks collection

4/1/04

4:35 PM

Page 139

Your friendship networks

protective interventions for children. Probation has been directed

towards more supervisory work, such as risk assessment, cognitive
behavioural interventions at the individual level and enforcement,
and away from its traditional role in shaping the social aspects of
rehabilitation. The pastoral role of teaching has been squeezed by the
need to focus on curriculum delivery and maintaining discipline.
While new skills for helping people to cultivate networks may be
emerging in neighbourhood renewal work, and perhaps even among
some employment advisers, they are hardly formalised, trained or
valued as such.
It is also clear that the tools for shaping social networks are better
developed in those services that work with the least advantaged. For
those who think that public services should always be targeted upon
the worst off, and for those who think that problems about social
networks mainly affect the poorest, perhaps this is not a problem. But
government should be concerned about the extent to which the least
well-off can use their networks to access people who are better off
than themselves, not only for instrumental reasons to do with seeking
work, but also to avoid the deepening mutual enclaving of the social
classes in respect of social ties that tends to follow whenever any
society goes through a period of growing income and wealth
inequality combined with high levels of fear of crime. In this context,
the limited nature of the toolkit with which government can address
the social ties of the better-off might well be a matter of some
concern.12
This suggests that government ought to be very cautious before
advocating ambitious policies for promoting social capital, as some
enthusiastic researchers and even some policy advocates in the World
Bank have been doing.13 This is not only out of suspicion of grand
social engineering projects in general but, more practically, because
policy-makers ought to recognise how little we really know about
how public services can develop sets of practices that might have
lasting influence on the social networks of service users and nonusers alike. In addition, they must recognise how far we are from
understanding how to strike trade-offs or make intelligent choices
Demos

139

Networks collection

4/1/04

4:35 PM

Page 140

Network logic

between the different types of social networks that government might

consider trying to promote.
Conclusion
In this situation it is much more sensible for governments to focus on
more modest goals. Specifically, they should surely make it a priority
to try to limit the clear harms their interventions do to peoples social
networks. For example, training programmes for the unemployed
that facilitate the creation of ties with other unemployed people only
do little to extend those individuals access to informal routes into
work, and may do some harm by way of peer effects on aspirations.
Again, it should be clear from recent inquiries that care in the
community quickly becomes neglect in the community and
domiciliary care can readily become a humane form of house arrest
if little or no attention is paid to the social networks of those who use
these services. Community development programmes that focus all
effort on building ties within a community and fail to address the
importance of links outside can quickly reinforce enclaving.
It makes far more sense at this stage in our knowledge for
governments to be trying to develop piecemeal strategies with which
to tackle these network harms. Along with this must go the most
careful attention to respecting the privacy of much of the highly
sensitive personal information that public bodies and their staff
collect about the friendships and acquaintance of their clients. For
without reassurance on these matters, the public will rightly be
reluctant to trust in governments that seek to influence their social
networks.
So, do governments do better by trying to influence the networks
of citizens deliberately? Or do they actually do better when they
simply provide material services and leave the network consequences
where they lie? At the moment, the only answer seems to be that if
policy-makers are to do better by trying to coordinate policies and
measures deliberately, then they had better begin by being very
modest in ambition; by prioritising to limit network harms that flows
from public service provision; by recognising that all the good things
140

Demos

Networks collection

4/1/04

4:35 PM

Page 141

Your friendship networks

do not go together; and by accepting that they must care about all
not just one or two types of networks.
Dr Perri 6 is a senior research fellow at the Health Services Management
Centre, University of Birmingham.
Notes
1

2
3

6
7
8
9

10

This article is based on work for a forthcoming book, Perri 6, The Politics of
Social Cohesion; see also Perri 6, Can government influence our friendships?
The range and limits of tools for trying to shape solidarities, in C Phillipson,
G Allen and D Morgan (eds), Social Networks and Social Exclusion: sociological
and policy issues (Aldershot and London: Ashgate, 2003); Perri 6, Governing
friends and acquaintances: public policy and social networks, in V Nash (ed.),
Reclaiming Community (London: Institute for Public Policy Research, 2002).
K Polanyi, The Great Transformation: the political and economic origins of our
time (Boston, Mass: Beacon Press, 1944).
Perri 6, Can government influence our friendships? The range and limits of
tools for trying to shape solidarities, in C Phillipson, G Allen and D Morgan
(eds), Social Networks and Social Exclusion: sociological and policy issues
(Aldershot and London: Ashgate, 2003). The taxonomy is based on that of M
Douglas, Cultural bias, in M Douglas, In the Active Voice (London: Routledge
and Kegan, 1982) and ultimately on Durkheim, Suicide: a study in sociology,
tr. J Spaulding and G Simpson (London: Routledge, 1951 [1897]).
This is based on a review of the available literature, but it should be noted that
rather little of the research has been principally concerned with evaluating
impacts on social networks, and there are many limitations and weaknesses in
the studies.
Figure 2 refers to the Western world, with its relative aggregate weighting
towards greater individualism rather than either of the strongly integrated
solidarities. Interventions that seem ineffective in this context might be more
efficacious when used in social contexts which are initially more
communitarian in their institutions.
The full review of the literature will appear in Perri 6, The Politics of Social
Cohesion, forthcoming.
GC Wenger, Social networks and the prediction of elderly people at risk, Aging
and Mental Health 1, no 4 (1997).
RE Putnam, Bowling Alone: the collapse and revival of American community
(New York: Simon and Schuster, 2000).
D Stolle, Getting to trust: an analysis of the importance of institutions,
families, personal experiences and group membership, in P Dekker and EM
Uslaner (eds), Social Capital and Participation in Everyday Life (London:
Routledge, 2001).
M Chaves and W Tsitos, Congregations and social services: what they do, how

Demos

141

Networks collection

4/1/04

4:35 PM

Page 142

Network logic

11
12
13

142

they do it, and with whom, Nonprofit and Voluntary Sector Quarterly 30, 4
(2001).
M Hooghe and D Stolle (eds) Generating Social Capital: civil society and
institutions in comparative perspective (Basingstoke: Palgrave, 2003).
Perri 6, Profiles, networks, risk and hoarding: public policy and the dynamics
of social mobility and social cohesion, paper for the Performance and
Innovation Unit seminar on social mobility, 20 March 2001.
Available at www.worldbank.org/poverty/scapital/; see also S Aldridge, D
Halpern and S Fitzpatrick, Social capital: a discussion paper, Strategy Unit,
Cabinet Office, 2002, available at www.number10.gov.uk/su/social%20capital/socialcapital.pdf.

Demos

Networks collection

4/1/04

4:35 PM

Page 143

Network Logic 11

If the rhetoric of
community-led
regeneration is to
be translated into
real change, then
informal networks
are crucial to the
journey . . .
Developing the wellconnected community
Alison Gilchrist

Networks collection

4/1/04

4:35 PM

Page 144

Networks collection

4/1/04

4:35 PM

Page 145

11. Developing the wellconnected community

Alison Gilchrist

Over the recent past, society has become both more informal and
more complex. We prefer to live without status and hierarchy,
abandoning bureaucratic rules and conventions in favour of more
fluid notions of identity, and becoming more flexible about how, and
by whom, decisions are made.
These trends are recognised in many current debates about
community. Nonetheless, government policies to promote stronger
communities and active citizenship have tended to emphasise the
role of individuals within formal structures, and, until recently,
have overlooked the significance of informal activities within
community settings. The related concepts of community and
networks must be better understood by policy-makers if they are to
avoid the risk of masking common experiences of inequality and
discrimination, based on enduring power imbalances and social
exclusion.
Governments love affair with community
Strategies for public participation in decision-making date back
several decades. The Labour government has simply accelerated this
trend with its current emphasis on subsidiarity and partnership
working based on community involvement and leadership.1 Policy
debates have often been premised on an image of community as a
homogenous and harmonious dimension of social life, securely

Demos

145

Networks collection

4/1/04

4:35 PM

Page 146

Network logic

located within geographical boundaries and anchored by shared

values and assumptions. Policy-makers have regarded the idea of
community as a panacea for societys ills a haven where people can
find, and express, solidarity and compassion. The term community is
used in policy debates to soften the edges of unpalatable programmes
and to harness the unpaid energy, expertise and enthusiasm of local
volunteers and activists.
The governments interest in community can be broken down in
relation to three different but not mutually exclusive goals.
Improving governance

Broad swathes of government policy require public and private

agencies to work in partnership arrangements, and to engage with
communities in developing and delivering programmes that meet
needs and aspirations. Decision-making procedures, led by elected
politicians and implemented by public servants, are augmented or
displaced by consultative forums and mechanisms for encouraging
greater participation and influence among groups who have hitherto
been marginalised or alienated by formal political systems. New
forms of governance are being devised to support more effective
community representation in a concerted attempt to modernise
democracy and renovate civil society.
Yet there is still a long way to go in learning how these complex
multi-agency structures can manage the interface between
communities, programme managers and policy-makers. One key
challenge is to understand better the impact of informal networks on
what should be transparent and democratic processes.2 Community
representatives report that they feel disempowered by the culture of
many partnerships, finding themselves out of the loop of much of
the decision-making. While often being excluded from the informal
networks where crucial discussions take place, their attendance at
partnership meetings also means that they have less time to attend to
their own community networks, which provide support and ensure
accountability. The resulting frustrations and suspicions can lead to
derailment and delays because of the lack of trust among partners
146

Demos

Networks collection

4/1/04

4:35 PM

Page 147

Developing the well-connected community

and a consequent unwillingness to delegate to delivery agents on the

part of policy-makers.
Strengthening social capital

Over the past few years, there has been a burgeoning interest in the
concept of social capital, a term coined by social scientists to
highlight the collective value of the networks of personal
relationships and organisational connections. There is a growing
body of empirical evidence suggesting that robust and diverse social
networks enhance the health and happiness of individuals, and
contribute to the well-being of society as a whole. In his well-known
book, Bowling Alone, Robert Putnam sets out a range of studies
linking measures of social capital to educational attainment,
economic regeneration, crime reduction and employability.3 All of
these are key themes in the governments neighbourhood renewal
agenda and, as a consequence, policy-makers have enthusiastically
embraced social capital theory as a possible framework for increasing
social inclusion and community cohesion hence current
programmes supporting volunteering, active citizenship and the work
of social entrepreneurs. The government is particularly interested in
bridging and linking social capital, and recognises the important
role that voluntary and community organisations play in this respect.
However, sceptics have raised issues around the social capital
approach to strengthening communities, pointing to inequalities
operating within networks and arguing that norms can be oppressive
for some, while empowering for others. Putnams studies relied on
survey data on civic engagement and membership of voluntary
organisations. It has proved much harder, however, to capture the
connectivity that this is supposed to reflect, mainly because it has
been difficult to collect and analyse evidence on the nature and extent
of informal community networks, let alone make an assessment of
how these might have changed over time. This is beginning to be
addressed through the use of participatory appraisal techniques,
where communities undertake their own research into local social
capital.4
Demos

147

Networks collection

4/1/04

4:35 PM

Page 148

Network logic

Enhancing service delivery

Communities have always provided a range of services to their own

members, through self-help groups and mutual aid. Interpersonal
networks, based on trust and need, enable people to access services
and resources without an exchange of money or formal contract.
These arrangements work when people can be confident that such
exchanges are likely to be reciprocated, either directly through
returned favours or indirectly through a shared commitment to the
common good. Local exchange trading schemes (LETS) and
TimeBank initiatives have sought to replicate this networking
through the explicit use of tokens or time credits.
Voluntary organisations regard communities as their natural
constituency, claiming that they are innovative and sensitive to the
requirements of disadvantaged sections of the population that have
been neglected by mainstream services. In seeking to tackle social
exclusion, the government has looked to strengthen the capacity of
the voluntary and community sector to deliver a wider range of
services, acknowledging its particular strengths in reaching the
hardest to reach. However, even though compacts have been
developed to validate and protect the autonomy of voluntary and
community organisations, greater partnership working still runs the
risk of eroding the independent and pioneering nature of many
voluntary organisations. Recent reviews have recommended a
targeted increase in resources to the sector, but at the same time called
for a rationalisation of intermediary and support bodies in favour of
front-line service providers. These trends in funding arrangements
may make it less easy for the sector to maintain its flexible, boundaryspanning networks because of the pressure to deliver services and
provide formal consultation mechanisms.
These new policy directions acknowledge that communities can
contribute to democratic decision-making by adding value to the
design, development and delivery of services, and in so doing
promote social inclusion and cohesion. However, the rhetoric of
participation and partnership has proved difficult to turn into a
reality. There has been an over-reliance on voluntary sector
148

Demos

Networks collection

4/1/04

4:35 PM

Page 149

Developing the well-connected community

professionals or self-appointed community leaders acting as representatives for the sector, and a failure to allow enough time for
relationships of trust, respect and understanding to develop between
partners and communities. Partnership boards have been under
pressure to spend budgets and find solutions to hitherto intractable
social problems. Unrealistic timescales set by electoral cycles and a
tradition of short-termism have created additional pressures. It is not
surprising that many are experiencing major difficulties in achieving
the ambitious targets set by government.
Developing community networks
Despite this focus on service delivery there has been a growing
realisation that community engagement is not the same as voluntary
sector liaison. Community leaders have often been unpaid and undervalued activists who are constantly asked to convey the (sometimes
unknown) views of their communities and to defend decisions over
which they feel no sense of ownership. Effective community
involvement in cross-sectoral partnerships can only be sustained,
therefore, on the basis of sufficient organisational capacity and strong
interpersonal connections. Communities are themselves complex and
dynamic, comprising overlapping but shifting networks and alliances,
used by communities to promote or defend their interests in an everchanging environment. Communities exist where there is interaction
and mutual influence. Social networks express and reinforce a sense
of belonging, of mutuality, based on somewhat transient notions of
us and them. To varying extents, community networks comprise
neighbours, work colleagues, fellow activists, those who might share a
faith or hobby, or people who have faced similar experiences, for
example, through migration or discrimination. If this is the case, then
how can networks be shaped to contribute to stronger, more inclusive
communities?
The well-connected community
In the first instance, it is important to acknowledge that communities
have always contained differences and divisions, even where these
Demos

149

Networks collection

4/1/04

4:35 PM

Page 150

Network logic

have been hidden under a veneer of conformity or respectability.

Tensions are an inescapable part of community life, and the capacity
to resolve conflicts is essential for a cohesive and inclusive society. As
Fritjof Capra recognises in his model of the web of life, ethnic and
cultural diversity plays the same role in human societies as that of
ecological diversity in ecosystems. A diverse community is resilient,
with the capacity to adapt to changing situations: However, diversity
is a strategic advantage only if there is a truly vibrant community,
sustained by a web of relationships. If the community is fragmented
into isolated groups and individuals, diversity can easily become a
source of prejudice and friction.5
We actively and symbolically construct communities according to
the conditions we find, the challenges that we encounter and the
choices we make. Many communities organise themselves along lines
that perpetuate injustice and segregation. Traditions and structures
are overly rigid and social connections simply repeat patterns of
informal interaction among people who are broadly similar or who
obey conventional mores. Such communities are often isolated and
unwilling to adapt to change. They become stagnant, defending
themselves against newcomers rather than seeking opportunities for
engagement and learning.
Organisation studies and complexity theory provide useful insights
into the important functions played by network-type linkages in
helping complex systems to manage themselves within turbulent
(ever-changing) environments. Complex systems that are sufficiently
well connected demonstrate a tendency to self-organise over time,
with clusters and groups emerging from the pattern of interactions
among the participating members. In some cases these configurations evolve as formal structures that survive so long as they fit
a niche in the organisational environment. Network modes of
organising are said to reduce transaction costs, though in reality these
are simply transferred to less public arenas, where invisible emotional
work builds trust and mutuality. Networks operate as highly effective
communication channels, gathering intelligence from a range of
sources and processing it to become collective wisdom or common
150

Demos

Networks collection

4/1/04

4:35 PM

Page 151

Developing the well-connected community

sense. Community networks convey information and ideas,

recruiting potential collaborators and supporters for self-help groups
and collective action.
Communities can be thought of as complex systems comprising
overlapping networks of social and organisational relationships.
Their boundaries are difficult to define; they consist of multiple,
interdependent elements; there exists no centralised control
mechanism; and interactions between members are governed by
localised and relatively simple conventions. Well-functioning
communities possess a range of capacities for absorbing or adapting
to change, managing internal tensions and generating (and
dissolving) a variety of forms for collective action.
These capacities are organic and owned by communities themselves. In many instances, they has been built up over years of
collective action among a fairly stable population who accumulate
skills, knowledge and confidence from a shared experience of
successful organising. Social networks with diverse, boundaryspanning contacts are particularly useful, especially where these
provide links to other communities and powerful agencies. Loose
networks are often the precursors to more formal organisational
arrangements, especially where these involve people or organisations
from a diversity of backgrounds and interests.
Social movements depend on activist networks to organise
collective action, to raise awareness and to lobby for change. For
example, the first Bristol Festival Against Racism was organised
entirely through the political, professional and personal networks of
the main coordinators. Having no resources of its own, it used
connections to mobilise organisations, galvanise individuals, generate
ideas, gain access to funding and distribute its publicity. Trust
and shared values were vital to these achievements and the
organisers particularly encouraged groups to work together where
there had previously been only limited contact. A subsequent
evaluation of the initiative revealed that participation in the Festival
had reinforced peoples commitment and confidence around antiracism. A framework had been established on which stronger, more
Demos

151

Networks collection

4/1/04

4:35 PM

Page 152

Network logic

diverse networks could grow. The success of the Festival Against

Racism provided a model for subsequent RESPECT events and has
important implications for strategies to develop greater community
cohesion.
However, some areas and some sections of the population lack
what has been termed community capacity and are disadvantaged as
a result. Informal networks might be cliquey, perpetuating power
imbalances and fostering segregation rather than integration across
different groups. Consequently, such communities find it difficult to
represent their interests to public decision-making bodies and react
to conflict and crises in ways that are seen by outsiders as chaotic or
apathetic.
Networking for community development
Tackling such situations involves strategic and well-resourced
interventions, including support for individuals who take on
representative or leadership roles, and help with setting up and
managing formal organisations.
Community development has traditionally emphasised these two
approaches, now the focus of several government initiatives, for
example, Community Champions and FutureBuilders. While these
are making important contributions to strengthening voluntary and
community sector capacity, they are not the most effective means of
building social capital.
Community development has been defined as [being] about
building active and sustainable communities based on social justice
and mutual respect. It is about changing power structures to remove
barriers that prevent people from participating in the issues that
affect their lives.6 A vital, but underestimated aspect of this work is
the practical, psychological and political support that nurtures
informal networks. Community workers often find themselves acting
as interpreters and mediators within communities, helping people to
talk and work together when there are difficulties relating to
language, assumptions and the occasional antagonism. They may
themselves provide the boundary-spanning mechanisms, brokering
152

Demos

Networks collection

4/1/04

4:35 PM

Page 153

Developing the well-connected community

joint ventures between organisations or staying in touch through

informal conversation.
Ideally, effort and resources should be used to create the conditions
where communities can grow and maintain their own networks,
recognising that extra guidance and assistance may be needed to
ensure that these are inclusive, diverse and extrovert. This means
helping with the boundary-spanning connections and addressing
power differentials and obstacles to effective cooperation. There are
many ways in which community development activities can
encourage and enhance networking. Indeed this is a core purpose for
community development.7 Community workers provide contact
details and information for the communities they work with and
other practitioners. They direct people to appropriate agencies or
funding bodies. Community workers work with communities to
arrange activities such as street festivals or meetings, where
networking happens fairly naturally, though they may have a role in
ensuring that such events are fully accessible to all members of the
relevant communities. Communal buildings such as community
centres or village halls provide space for casual interaction as well as
more purposeful joint working. These need to be safe, welcoming and
affordable, so the dcor, location and publicity may need careful
consideration to avoid the space being dominated by certain groups.
Experience suggests that this will require continuous monitoring and
occasional interventions to deal with inevitable conflicts and
misunderstandings.
Conclusion
The idea of the well-connected community should not be seen as a
nostalgic, communitarian model of urban neighbourhoods or village
life. It is a radical approach that celebrates the dynamic and multifaceted nature of our lives while asserting the value of social
connectivity. Community development strategies that strengthen and
extend informal networks are essential to tackling social exclusion
and building community cohesion. They improve and sustain
partnership working by building trust between partners and ensuring
Demos

153

Networks collection

4/1/04

4:35 PM

Page 154

Network logic

more representative community involvement. These processes are not

amenable to the current audit regimes of targets and indicators, and
progress is better measured using techniques of participatory
appraisal and network analysis. It is difficult to predict, let alone
control, outcomes and it is therefore likely that this model will be
resisted by policy-makers and politicians who are interested in
tangible and short-term gains. However, if the rhetoric of civil
renewal and community-led regeneration are to be translated into
real changes in peoples lives, then informal networks, supported
where necessary by community workers, are crucial components of
the journey towards a fairer, more sustainable vision of social justice.
Alison Gilchrist is Director of Practice Development at the Community
Development Foundation and the author of The Well-connected
Community: a networking approach to community development.
Notes
1
2
3
4
5
6
7

154

G Chanan, Searching for Solid Foundations: community involvement and urban

policy (London: ODPM, 2003).
M Mayo and M Taylor, Partnership and power in community regeneration in
S Balloch and M Taylor (eds), Partnership Working (Bristol: Policy Press, 2001).
R Putnam, Bowling Alone: the collapse and revival of American community (New
York: Simon and Shuster, 2000).
See for example, P Walker et al, Prove It: measuring the effect of neighbourhood
renewal on local people (London: New Economics Foundation, 2000) and
Bassacs BEDE programme, available at www.bassac.org.uk/info_bede.htm.
F Capra, The Web of Life: a new synthesis of mind and matter (London:
HarperCollins, 1996).
Standing Conference for Community Development, Strategic Framework for
Community Development (Sheffield: SCCD, 2001).
A Gilchrist, The well-connected community: networking to the edge of
chaos, Community Development Journal 35 (2000).

Demos

Networks collection

4/1/04

4:35 PM

Page 155

Network Logic 12

Not all networks

are created equal.
The mistake has
been to equate
their existence
with mechanisms
of effective social
action . . .
Networks and
neighbourhoods
Robert J Sampson

Networks collection

4/1/04

4:35 PM

Page 156

Networks collection

4/1/04

4:35 PM

Page 157

12. Networks and

neighbourhoods
the implications of connectivity
for thinking about crime in the
modern city
Robert J Sampson

The idyllic notion of local communities as urban villages

characterised by dense networks of personal ties is a seductive image,
and one that pervades theoretical perspectives on neighbourhood
crime. The idea seems to be that tight-knit neighbourhoods are safe
because of their rich supply of social networks. Yet such ideal typical
neighbourhoods appear to bear little resemblance to contemporary
cities where weak ties prevail over strong ties and social interaction
among residents is characterised more by instrumentality than
altruism. Moreover, the dark side of community is often neglected
social networks can and often are put to use for illegal or violent
purposes. In short, a deceptive conflation of networks and
neighbourhoods characterises the current scene.
The urban village model of cities is further compromised by the
assumption that networks of personal ties map neatly on to the
geographically defined boundaries of neighbourhoods, such that
neighbourhoods can be analysed as independent social entities. In
fact, social networks in the modern city frequently traverse traditional
ecological boundaries, many of which are permeable and vaguely
defined. Living in close proximity to high-crime neighbourhoods
may increase the risk of crime no matter what the density of social
networks in an adjacent neighbourhood. It follows that neighbourDemos

157

Networks collection

4/1/04

4:35 PM

Page 158

Network logic

hoods themselves need to be conceptualised as nodes in a larger

network of spatial relations.
In this essay I explore these issues by considering new ways of
thinking about the relevance of neighbourhoods and networks that
nonetheless build on important work of the past. Tradition needs to
be transcended, not discarded. Neighbourhoods, after all, show
remarkable continuities in patterns of criminal activity. For at least a
hundred years criminological research in the ecological tradition has
confirmed the concentration of interpersonal violence in certain
neighbourhoods, especially those characterised by poverty, the racial
segregation of minority groups, and the concentration of singleparent families. The challenge, then, is to incorporate new urban
realities into our understanding of crime and ultimately our policy
responses.
From social disorganisation to networks
The intellectual history of situating social networks in local
neighbourhoods is a venerable one. In the classic work of the Chicago
School of Urban Sociology in the early twentieth century it was
thought that density, low economic status, ethnic heterogeneity and
residential instability led to the rupture of local social ties, which in
turn accounted for high rates of crime and disorder. Later in the
century, the concept of social disorganisation came to be defined as
the inability of a community to realise the common values of its
residents and maintain effective social order. This theoretical
definition was formulated in systemic terms the allegedly
disorganised community was viewed as suffering from a disrupted or
weakened system of friendship, kinship and acquaintanceship
networks, and thus ultimately of processes of socialisation.
More recently, the intellectual tradition of community-level
research has been revitalised by the increasingly popular idea of
social capital. Although there are conflicting definitions, social
capital is typically conceptualised as being embodied in the social ties
among persons. In an influential version, Robert Putnam defines
social capital as the networks, norms, and trust that facilitate
158

Demos

Networks collection

4/1/04

4:35 PM

Page 159

Networks and neighbourhoods

coordination and cooperation for mutual benefit.1 The connection of

social disorganisation and social capital theory can be articulated as
follows: neighbourhoods bereft of social capital (read social networks) are less able to realise common values and maintain the social
controls that foster safety.
Despite the popular appeal of social capital, there are good reasons
to question the translation of strong social ties into low crime rates.
First, in some neighbourhood contexts strong ties may impede efforts
to establish social control. William Julius Wilson, for example, has
argued that residents of very poor neighbourhoods tend to be tightly
interconnected through network ties but without necessarily
producing collective resources such as social control. He reasons that
ties in the inner city are excessively personalistic and parochial in
nature socially isolated from public resources.2
Second, networks connect do-gooders just as they connect drug
dealers. In her study of a black middle-class community in Chicago,
Pattillo-McCoy specifically addresses the limits of tight-knit social
bonds in facilitating social control.3 She argues that although dense
local ties do promote social cohesion, at the same time they foster the
growth of networks that impede efforts to rid the neighbourhood of
organised drug- and gang-related crime. In this way, dense social ties
have both positive and negative repercussions, reminding us that in a
consideration of networks it is important to ask what is being connected
networks are not inherently egalitarian or prosocial in nature.
Third, shared expectations for social control and strategic
connections that yield action can be fostered in the absence of thick
ties among neighbours. As Granovetter argued in his seminal essay,
weak ties less intimate connections between people based on more
infrequent social interaction may be critical for establishing social
resources, such as job referrals, because they integrate the community
by way of bringing together otherwise disconnected subgroups.4
Consistent with this view, there is evidence that weak ties among
neighbours, as manifested in middle-range rather than either nonexistent or intensive social interaction, are predictive of lower crime
rates.
Demos

159

Networks collection

4/1/04

4:35 PM

Page 160

Network logic

Collective efficacy
Research on dense social ties reveals a paradox of sorts for thinking
about crime. Many city-dwellers have only limited interaction with
their neighbours and yet appear to generate community-specific
social capital. Moreover, urban areas where strong ties are tightly
restricted geographically may actually produce a climate that
discourages collective responses to local problems. To address these
urban realities, in recent work I and my colleagues have proposed a
focus on mechanisms of social control that may be facilitated by, but
do not necessarily require, strong ties or associations.5 Rejecting the
outmoded assumption that neighbourhoods are characterised by
dense, intimate, emotional bonds, I define neighbourhoods in
ecological terms and highlight variations in the working trust and
shared willingness of residents to intervene in achieving social
control. The concept of neighbourhood collective efficacy captures
the importance of this link between trust and cohesion on the one
hand and shared expectations for control on the other. Just as selfefficacy is situated rather than general (one has self-efficacy relative to
a particular task), a neighbourhoods efficacy exists relative to specific
tasks such as maintaining public order.
Viewed through this theoretical lens, collective efficacy is a taskspecific construct that draws attention to shared expectations and
mutual engagement by residents in local social control. To measure
the social control aspect of collective efficacy, we have asked residents
about the likelihood that their neighbours could be counted on to
take action under various scenarios (for example, children skipping
school and hanging out on a street corner, or the fire station closest to
home being threatened with budget cuts). The cohesion and working
trust dimension has been measured by items that capture the extent
of local trust, willingness to help neighbours, a close-knit fabric, lack
of conflict and shared values. Published results show that after
controlling for a range of individual and neighbourhood characteristics, including poverty and the density of friendship ties, collective
efficacy is associated with lower rates of violence. Neighbourhoods high
in collective efficacy predict significantly lower rates of violence even
160

Demos

Networks collection

4/1/04

4:35 PM

Page 161

Networks and neighbourhoods

where earlier experience of violence may have depressed collective

efficacy because of fear.
Moving away from a focus on private ties, use of the term
collective efficacy is meant to signify an emphasis on shared beliefs
in a neighbourhoods capability for action to achieve an intended
effect, coupled with an active sense of engagement on the part of
residents. Some density of social networks is essential, to be sure,
especially networks rooted in social trust. But the key theoretical
point is that networks have to be activated to be ultimately meaningful.
Collective efficacy therefore helps to elevate the agentic aspect of
social life over a perspective centred on the accumulation of stocks of
social resources (or what some call social capital). This is consistent
with a redefinition of social capital in terms of expectations for action
within a collectivity.
Distinguishing between the resource potential represented by
personal ties, on the one hand, and the shared expectations for action
among neighbours represented by collective efficacy, on the other,
helps clarify the dense networks paradox: social networks foster the
conditions under which collective efficacy may flourish, but they are not
sufficient for the exercise of control. So the theoretical framework
proposed here recognises the transformed landscape of modern
urban life, holding that while community efficacy may depend on
working trust and social interaction, it does not require that my
neighbour or the local police officer be my friend.
Exclusive and non-exclusive social networks
As noted above, recent writing on social capital tends to gloss over its
potential downside namely that social capital can be drawn upon
for negative as well as positive goals. After all, resources can be put to
many uses, and therefore some constraints on goals are theoretically
necessary. For example, we would not consider racial exclusion, as
practised in many a defended neighbourhood, to be a desirable
result of networking. Many neighbourhood associations in American
cities have been so exploited by whites to keep blacks from moving to
white working-class areas. Although often resisted by social scientists,
Demos

161

Networks collection

4/1/04

4:35 PM

Page 162

Network logic

I therefore believe there is a need to invoke a normative or goaldirected dimension when evaluating social networks and collective
efficacy.
To judge whether neighbourhood structures serve collective needs
I apply the non-exclusivity requirement of a social good does its
consumption by one member of a community diminish the sum
available to the community as a whole? For example, I would argue
that safety, clean environments, quality education for children, active
maintenance of intergenerational ties, the reciprocal exchange of
information and services among families, and the shared willingness
to intervene on behalf of the neighbourhood all produce a social
good that yields positive externalities potentially of benefit to all
residents especially children. As with other resources that produce
positive externalities, I believe that collective efficacy is widely desired
but much harder to achieve, owing in large part to structural
constraints. Ultimately, then, I view the role of social networks in the
production of collective efficacy not as a simple panacea but as
dependent on specific normative and structural contexts.
The natural question that follows is: what are the kinds of contexts
that promote collective efficacy and non-exclusive social networks?
Although it is beyond the scope of this essay, I would argue that the
infrastructure and cohesion of organisations help sustain capacity for
social action in a way that transcends traditional personal ties. In
other words, organisations are at least in principle able to foster
collective efficacy, often through strategic networking of their own.
Whether garbage removal, choosing the site of a fire station, school
improvements, or police responses, a continuous stream of challenges
faces modern communities, challenges that no longer can be met (if
they ever were) by relying solely on individuals. Action depends on
connections among organisations, connections that are not
necessarily dense or reflective of the structure of personal ties in a
neighbourhood. Our research supports this position, showing that
the density of local organisations and voluntary associations predicts
higher levels of collective efficacy, controlling for poverty and the
social composition of the population.6
162

Demos

Networks collection

4/1/04

4:35 PM

Page 163

Networks and neighbourhoods

Figure 1 Neighbourhood inequality, social processes

and safety
Structural
resources
Social
networks

Collective
efficacy

Safety

Organisational
or institutional
factors
Prior crime or
violence

Inequality in other resources nonetheless still matters for

explaining the production of collective efficacy. Concentrated
disadvantage and lack of home ownership, for example, predict lower
levels of later collective efficacy, and the associations of disadvantage
and housing instability with violence are significantly reduced when
collective efficacy is controlled. These patterns are consistent with the
inference that neighbourhood constraints influence violence in part
through the mediating role of neighbourhood efficacy. Our work
suggests that social resources and social networks create the capacity
for collective efficacy, but it is the act of exercising control under
conditions of trust that is the most proximate to explaining crime
(see Figure 1).
Spatial networks
Networks need not be conceptualised only in personal terms. I would
argue that neighbourhoods are themselves nodes in a larger network
of spatial relations. Contrary to the common assumption in
criminology of analytic independence, my contention is that
Demos

163

Networks collection

4/1/04

4:35 PM

Page 164

Network logic

neighbourhoods are interdependent and characterised by a

functional relationship between what happens at one point in space
and what happens elsewhere.
Consider first the inexact correspondence between the
neighbourhood boundaries imposed by census geography and the
ecological properties that shape social interaction. One of the biggest
criticisms of neighbourhood-level research to date concerns the
artificiality of boundaries; for example, two families living across the
street from one another may be arbitrarily assigned to live in different
neighbourhoods even though they share social ties. From the
standpoint of systemic theory, it is thus important to account for
social and institutional ties that link residents across neighbourhoods.
The idea of spatial dependence challenges the urban village model,
which implicitly assumes that neighbourhoods represent intact social
systems, functioning as islands unto themselves.
Second, spatial dependence is implicated by the fact that offenders
are disproportionately involved in acts of violence near their homes.
From a routine activities perspective, it follows that a neighbourhoods risk of violence is heightened by geographical proximity to
places where known offenders live or to places characterised by risk
factors such as concentrated poverty or low collective efficacy.
A third motivation for studying spatial dependence relates to the
notion that interpersonal crimes such as homicide are based on social
interaction and thus subject to processes of diffusion where knockon effects may be felt far from the initial point of impact. Acts of
violence may themselves instigate a sequence of events that leads to
further violence in a spatially channelled way. A key insight, for
example, is that many homicides are retaliatory in nature, such that a
homicide in one neighbourhood may provide the spark that
eventually leads to a retaliatory killing in a nearby neighbourhood. In
addition, most homicides occur among persons known to one
another, usually involving networks of association that follow a
geographical logic.
There are good reasons, then, to believe that the characteristics of
surrounding neighbourhoods are crucial to understanding violence
164

Demos

Networks collection

4/1/04

4:35 PM

Page 165

Networks and neighbourhoods

in any given neighbourhood. Our findings support this notion by

establishing the salience of spatial proximity and the inequality of
neighbourhood resources that are played out in citywide dynamics.
The mechanisms of racial segregation reinforce spatial inequality,
explaining why it is, for example, that despite similar income profiles
black middle-class neighbourhoods are at greater risk of violence
than white middle-class neighbourhoods. In short, violence is
conditioned by the characteristics of spatially proximate neighbourhoods, which in turn are conditioned by adjoining neighbourhoods
in a spatially linked process that ultimately characterises the entire
metropolitan system. Policies that focus solely on the internal
characteristics of neighbourhoods, as is typical, are simply
insufficient.
Conclusion
We live in a network society, or so we are widely told. But not all
networks are created equal, and many lie dormant. The mistake has
been to equate the existence of networks with mechanisms of
effective social action. As Arthur Stinchcombe put it in a useful
analogy, just as road systems have their causal impact through the
flow of traffic, so systems of links among people and organisations
(and in our case, neighbourhoods) have their causal impact through
what flows through them.7 The problem then becomes obvious
through neighbourhood networks flows the full spectrum of lifes
realities, whether despair, criminal knowledge, friendship or social
control.
In this essay I have considered one small slice of the problem. My
basic position is that collective action for problem-solving is a crucial
causal mechanism that is differentially activated under specific kinds
of contextual conditions. The density of social networks is only one
and probably not the most important characteristic of neighbourhoods that contributes to effective social action. I have also
argued that neighbourhoods themselves are part of a spatial network
encompassing the entire city. To use an overworked term, not only are
individuals embedded, so too are neighbourhoods.
Demos

165

Networks collection

4/1/04

4:35 PM

Page 166

Network logic

Nothing in the logic of my approach is limited to the United States,

or any country for that matter. Our current research is seeking to
examine the role of spatial inequality and neighbourhood efficacy in
several cities around the world. To date we have explored a
neighbourhood-level, cross-national comparison of Chicago and
Stockholm. Although Chicago and Stockholm vary dramatically in
their social structure and levels of violence, this does not necessarily
imply a difference in the processes or mechanisms that link
communities and crime. Indeed, our analyses suggest that rates of
violence are predicted by collective efficacy in Stockholm as in
Chicago. Furthermore, collective efficacy is promoted by housing
stability and undermined by concentrated disadvantage again
similarly in both cities and in accord with our general theory. The
data are thus consistent with a general approach to social policy that
emphasises ameliorating neighbourhood inequality in social
resources and enhancing social conditions that foster the collective
efficacy of residents and organisations.
Robert Sampson is Henry Ford II Professor of the Social Sciences at
Harvard University.
Notes
1
2
3
4
5
6
7

166

R Putnam, The prosperous community: social capital and community life,

American Prospect, Spring (1993).
WJ Wilson, The Truly Disadvantaged (Chicago: University of Chicago Press,
1978).
ME Pattillo-McCoy, Black Picket Fences: privilege and peril among the black
middle class (Chicago: University of Chicago Press, 1999).
MS Granovetter, The strength of weak ties, American Journal of Sociology 78
(1973).
RJ Sampson, SW Raudenbush and F Earls, Neighborhoods and violent crime: a
multilevel study of collective efficacy, Science 277 (1997).
J Morenoff, RJ Sampson, SW Raudenbush, Neighborhood inequality, collective
efficacy, and the spatial dynamics of urban violence, Criminology 39 (2001).
A Stinchcombe, An outsiders view of network analyses of power in R Perrucci
and H Potter, Networks of Power (New York: Aldine De Gruyter, 1989).

Demos

Networks collection

4/1/04

4:35 PM

Page 167

Network Logic 13

Delegating
judgement and
individualising
service would be a
radical departure
for most of the
private and all of
the public sector . . .
Organising for success
Diane Coyle

Networks collection

4/1/04

4:35 PM

Page 168

Networks collection

4/1/04

4:35 PM

Page 169

13. Organising for success

how network technologies are
changing organisations
Diane Coyle

Swimming in the vast tides of information that are the result of

new(ish) network technologies it is easy to overlook a simple
consequence of cheap communications and information processing.
There is more of both, of course none of us could miss that. But all
that information is also flowing into different places. It isnt a
question of having the same connections as before, only faster:
entirely new connections are developing.
That is easy enough to understand in the abstract. What does it
mean in a specific context, such as a particular company or public
sector agency? Organisations take the shape they have for many
reasons, including historical accidents, political outcomes and
cultural context. However, one key explanation for organisational
structure is the ease or difficulty of exchanging information. The
reason companies exist is because there are compelling reasons for
some transactions not to take place in a market between individuals.
An important one is the existence of economies of scale. But others
concern the lack of information: the difficulty of monitoring quality
and effort or the impossibility of writing down all possible
contingencies in a watertight contract. In addition, companies take
certain internal forms because of problems like monitoring workers
productivity or keeping track of the large amount of information
relevant to day-to-day decisions.
If it is costly and difficult to exchange information, companies are

Demos

169

Networks collection

4/1/04

4:35 PM

Page 170

Network logic

more likely to be large, composed of many internal businesses, and

hierarchical. It will make sense to integrate essential supplies and
services in one single organisation. And it will make sense to channel
relevant information up a pyramid and have a set of rules for which
decisions need to be taken at each level of the business. Conversely, if
information flows more easily, outsourcing, breaking up conglomerates and flattening hierarchies will look more attractive.
And so it is been since the early 1990s, when the steep recession
triggered a substantial corporate reorganisation with exactly these
features. Although they might have looked like straightforward costcutting or management fads, each had a fundamental technological
rationale.
The pattern throughout the past decade has been a hollowing-out
of medium-sized companies. A huge merger wave created more very
big companies, increasingly with an internal network rather than a
hierarchical structure, taking advantage of economies of scale and
global market opportunities. The possibilities of ICT have allowed
more small companies to thrive by exploiting better and faster
information on specialist consumer demands or specialised sources
of supply, and operating in an external network of other companies.
The relative advantages of being in between have diminished. Indeed,
being medium-sized can combine all the disadvantages of lacking
scale with all the disadvantages of lacking flexibility.
There are no consistent data across countries, but the OECD has
estimated from different national sources that the average size of
companies has been shrinking. The number employing more than
500 people has declined since 1990 although the remaining big
companies are almost certainly much bigger than before. And in most
industrial countries (Japan being a significant exception) the number
of businesses employing up to 50 people has rocketed.1
It is hard to be sure how far this size redistribution will go, for
there will certainly always be some big firms around. But it is a safe
bet that the organisational and social ramifications are still in their
infancy.

170

Demos

Networks collection

4/1/04

4:35 PM

Page 171

Organising for success

Electricity and the flow of production

The radical possibilities new technologies might hold for the
organisational shape of British business are easier to appreciate from
a past example. Just how far can organisational change resulting from
a new technology actually go?
Take the case of electrification following the invention of the
electric dynamo in 1870. This innovation made it possible to run
every machine in a factory from a separate electric motor, rather than
all machines running from a single energy source, a steam engine. In
the latter case, all the machines had to be clustered around a central
drive shaft, so factories and mills were built on several storeys. If one
needed to be shut down for repair or maintenance, the whole factory
stopped. With an electric motor on each machine, they could be
halted individually and arranged in any order.
In fact, the machines could be much more efficiently laid out on
one storey according to the logical sequence of production at least,
once new factories were being built. Single-storey buildings could
have more windows even in the roof which increased productivity
further, and also improved safety. And once managers and engineers
started to think about altering the pattern of production, the way was
open for the assembly line.
The ramifications of the dynamo can be traced back further still.
Whereas factories used to be in town centres, the advantages of flat,
low buildings along with concerns about urban conditions created
an impetus towards building on greenfield suburban sites. Housing
followed the jobs, although of course the internal combustion engine
played a more central role in the development of suburbs and urban
sprawl.
The new style of factory needed machine operatives performing
more standardised tasks, the more so as the pattern of production
increasingly took on assembly line form. So instead of craft skills, the
demand was for a basic standard education. This in turn encouraged
the development of public primary and secondary schools, batchproducing assembly line workers. It is not too fanciful to see the

Demos

171

Networks collection

4/1/04

4:35 PM

Page 172

Network logic

culmination of the cluster of new technologies based on the electric

motor, developed in the late nineteenth century and diffused during
the early twentieth century, as culminating in the Fordist model of
production and mass consumer society.
The economies of scale and high initial capital requirements for
these mass-production plants favoured the emergence of very large
conglomerate companies. The early part of the twentieth century saw
a period of dramatic industrial consolidation, with the number of
firms in most industries shrinking from the hundreds to the handfuls
most dramatically in the automobile and electrical goods industries
themselves.
Todays new technologies and the flow of service
It would be an exaggeration to say that these earlier forces shaping
corporate organisation have been swept away. On the contrary,
economies of scale are more important than ever in many significant
industries more important because if they can be exploited on a
global scale they can give a business a near impregnable competitive
advantage. And more important too because of the network
externalities or initial research and development costs involved in
many important new goods consider software or biotech products.
In these cases spending on research and development is a kind of
strategic arms race in which it is getting harder and harder for new
competitors to enter the business. Big is definitely better in these
cases.
On the other hand, new opportunities are opening up due to the
fragmentation of demand. There has been an explosion in the choice
offered to consumers thanks to the increased flow of information.
Consumers are better able to communicate their specific preferences
to companies. Companies are better able to collect and act on that
information, through improved logistics and vastly more flexible
networked production systems.
Nobody ever thought quantity was the spice of life. When our basic
needs are met, we seek to meet additional wants in varied ways. So we
have seen in recent years an explosion in quality and variety. Henry
172

Demos

Networks collection

4/1/04

4:35 PM

Page 173

Organising for success

Ford famously said of the Model T: The customer can have any
colour he wants so long as it is black. The choice of new vehicle
models in US is now nearly 300; Ford offers 46 colours. The paradigm
now isnt Ford but Dell. It gives customers who order online 16
million theoretical combinations of specifications for a desktop PC.
Amazon has 2.3 million books available compared with 250,000 in
the biggest New York book superstore, and 40,000 to 100,000 in most
big bookstores. Nearly half the books ordered from Amazon.com are
titles not likely to be in stock in any physical store. In addition the
ability to search for titles and discover books online has also increased
special orders through physical stores by an unknown amount. MIT
Press estimates that online discovery has helped increase orders for its
backlist titles by 12 per cent. These spillovers, increasing market size,
improve the viability of publishing titles for a non-mass readership
and further increase variety in a virtuous circle. MIT economist Erik
Brynjolfsson estimates that the welfare gain to American consumers
from being able to choose books not available in stock in big
bookstores at about $1 billion a year, roughly five times the biggest
estimates of gains from lower prices online.2
Within this massively expanded array of choice in all kinds of
products, from toothpaste to computers, there are many niche
markets in which relatively small companies can compete effectively.
Even where they are selling tangible products, they are adding value
essentially through providing a service. The service can be thought of
as information-broking or more creatively as the satisfaction of
desires, the desires of individuals to step outside the mass market and
craft themselves.
This is a product of the fact that value added in the advanced
economies is increasingly weightless (literally so UK GDP weighed
roughly the same in 1999 as in 1990, although it had grown by a
quarter in real terms, according to the ONS).3 The way companies
can attract customers is through the service-like elements of what
they sell, not through processing stuff.
The essence of service is customisation something one person
does for another. While theres plenty of scope for standardisation in
Demos

173

Networks collection

4/1/04

4:35 PM

Page 174

Network logic

many sectors we categorise as services, such as processing insurance

claims or sorting cheques, these are low-value activities. Value added
in service and manufacturing sectors alike lies more and more in
individualisation, not standardisation.
Ultimately that means a transaction between individual customers
and individual employees, using the vastly increased flow of
information available to both. For all the hype about employee
empowerment and being a people business, very few companies have
delegated genuine decision-making authority to the level of the
individual employee. The parameters for making judgements without
referring up whats left of the hierarchy are usually tightly
circumscribed.
Implications
Delegating real judgement to permit individualised service would be
a radical departure from the organisational forms prevalent in most
private businesses and all of the public sector today. Indeed, public
services are still in the Model T Ford era, and the debate on
standards the adjective high is implicit indicates that nobody is
really thinking beyond standardisation. Yet many people are unlikely
to be satisfied for much longer with the Fordist model as citizens
when as consumers they are increasingly offered the Dell model. And
as consumers they are also increasingly dissatisfied with, say, a
standard call-centre script. They want an individual response. The
information is available to permit it it is the organisational form
that lags behind.
And understandably so, for this kind of radical organisational
devolution demands a high level of mutual trust between employees
and managers and ultimately investors and owners. Remember, the
transactions that take place within organisations are those it is too
difficult to transact in a market because of the informational
constraints. Yet it is extremely difficult to monitor the productivity
and quality of a non-standard service inside an organisation too. The
companies that add the most value will be the ones with the highest
levels of internal trust.
174

Demos

Networks collection

4/1/04

4:35 PM

Page 175

Organising for success

This is a profound management challenge. In theory, the new

information and communications technologies permit any employee
of a business to have real-time access to all kinds of information that
would allow her or him to make better more profitable decisions.
But very few companies indeed take advantage of this possibility. The
evidence is, though, that all the productivity gains from the
technologies are concentrated among companies that have undergone
organisational change. No change, no gain but if work is
reorganised, the level of productivity in the business can jump by 25
or 30 per cent according to case study evidence.4 Such figures are a
measure of how much consumers are willing to pay for improved
quality and customisation.
The transition from the type of company and employment
relationship we have now to the type suggested by the logic of a freeinformation world is going to be at least as radical and as slow as
the shift from Victorian entrepreneurialism to 1950s corporatism.
And yet the logic is there. With the average company life now down to
under 20 years, any business that can trace its roots back to the
nineteenth century and which hopes to survive through the twentyfirst century should start struggling with it now.
Diane Coyle runs Enlightenment Economics and is a visiting professor at
Manchester Universitys Institute for Political and Economic
Governance. She is the author of, among other books, Paradoxes of
Prosperity and The Weightless World.
Notes
1
2
3
4

B van Ark and E Monnikhof, Size distribution of output and employment: a

data set for five OECD countries 1960s1990, Economics Department Working
Paper no 166 (Paris: OECD, 1996).
E Brynjolfsson, MD Smith and Y Hu, Consumer surplus in the digital
economy: estimating the value of increased product variety at online
booksellers, MIT Working Paper, April (2003).
C Sheerin, UK material flow accounting, Economic Trends 583, June (2002).
See, for example, T Bresnahan, E Brynjolfsson and L Hitt, IT, workplace
organisation and the demand for skilled labor: a firm-level analysis, MIT
Working Paper (2002); E Brynjolfsson and L Hitt, Paradox lost? Firm-level

Demos

175

Networks collection

4/1/04

4:35 PM

Page 176

Network logic

evidence on the return to information systems spending, Management Science,

42 no 4 (1997); E Brynjolfsson and L Hitt, Computing productivity: are
computers pulling their weight?, MIT Working Paper (2000); E Brynjolfsson
and L Hitt, Beyond computation: information technology and organisational
transformation, Journal of Economic Perspectives 14 no 4, Fall (2000); McKinsey
Global Institute, US Productivity Growth 19952000: understanding the impact
of information technology relative to other factors, October (2001).

176

Demos

Networks collection

4/1/04

4:35 PM

Page 177

Network Logic 14

Smart devices will

be worn in your
clothing, jewellery
and accessories,
and will hop in
to local wireless
networks when
they need to . . .
The information utility
John Taylor

Networks collection

4/1/04

4:35 PM

Page 178

Networks collection

4/1/04

4:35 PM

Page 179

14. The information

utility
John Taylor

We are entering the era of the information utility. Billions of

computers, smart devices and huge data resources are inexorably
becoming societys indispensable infrastructure. Information utilities
are set to become as pervasive as printing, electricity, the automobile
and telephone and just as disruptive to existing social practices and
forms of organisation. Futurologists have produced innumerable
accounts of the importance of electronic networks such as the
internet in shaping our world. Social and management scientists have
written similarly about the importance of social networks of human
beings in characterising all aspects of society. From now on
networking means an intimate combination of the two: people and
technology.
It is helpful to think of the information utility as comprising what
goes on behind the wall, what is in front of the wall, and crucially
the two-way interactions that can go on through the wall between
people and things connected to the utility. Behind the wall there will
be huge amounts of connectivity, computing power and data
resources of all kinds. When you access the information utility from
any location in the world you will not have to worry about where the
resources are or how they work, any more than you worry about
where the electricity and water come from when you turn on your
dishwasher. In front of the wall, there will be billions of smart devices
of all kinds connected wirelessly to each other and to the utility,
Demos

179

Networks collection

4/1/04

4:35 PM

Page 180

Network logic

enabling billions of people to be online all the time they want to be.
These sets of information utilities evolving around the world
represent an emerging global information infrastructure. What
makes this emergence possible indeed, inevitable is partly the
continuing pace of technological change: Moores law will still enable
the underlying microelectronics technology to deliver a doubling of
price-performance every 18 months for the next 1020 years; storage
technology hard discs, CDs, flash memory and so on is getting
cheaper even faster than this, with really disruptive new technologies
in the pipeline such as atomic resolution storage and holographic
memory. But it is also the result of the demand-side pressures
unleashed by the new and radical capabilities that this infrastructure
will possess, and the useful (or simply delightful) applications to
which they may be put.
Behind the wall
We are already seeing the first generation of the worldwide
information utility in the form of e-science Grids building on the
internet and the World Wide Web. Already todays internet and
telecommunications infrastructure probably comprises at least ten
billion computers and 100 exabytes (100 billion billion bytes) of data.
At least 600 million people can currently access the internet, which
carries about 4 billion emails per day, and between 1 and 2 billion
people worldwide now have phones.
The information utility will be qualitatively unlike any previous
global communication system such as road, rail, air or telephone
networks. It will consist of huge amounts of interconnected data and
computing power, and it will be able to interact through the wall with
tens of billions of smart devices of all kinds, and through them with
billions of people. In this way it will progressively exhibit more and
more intelligence. We are already seeing the rise of intelligent
software and machine learning, as the ability of humans to design and
programme large systems reaches its limits. The rich interconnection
between elements of the infrastructure will mean that a software
object that turns out to be powerful and useful for some particular
180

Demos

Networks collection

4/1/04

4:35 PM

Page 181

The information utility

task will be able to replicate, spread and jump species around the
infrastructure in many fewer generations than we see in biological
systems. Evolution through incremental enhancement and (often
unpredictable) adaptation will become a key paradigm.
We have already reached the stage when the global communications infrastructure can no longer be thought of as a system
in the traditional, hard engineering sense of the word. It has no
specification. It was not designed by any one person or organisation,
nor is it implemented, owned, operated or maintained by any
recognisable single authority. It is never the same from one access to
the next so faults cannot be reproduced. The underlying engineering
principles and protocols are based on having to know as little as
possible about what is going on in the infrastructure, rather than
knowing as much as possible; the old IT tradition of full
documentation is long gone. Instead it may be more appropriate to
think of the utility as a system in the biological sense of a living,
complex, adaptive whole, continuously evolving through the
collaboration of many autonomous subsystems.
A crucial property of such large sets of cooperating elements is that
they collectively display emergent behaviour, which is not readily
predictable from a knowledge of the individual elements. Emergent
behaviours come from systems involving decision-making entities,
like the way ant colonies organise their collective activities to discover
and retrieve food. An emergent behaviour involving the internet itself
is spamming. About 50 per cent of all the email on the internet this
year is spam, up from 2 per cent a couple of years ago.
In front of the wall
Our understanding of emergent behaviours in complex systems is still
very primitive, but will have to become highly sophisticated if we are
to grasp the full implications of the world of the information utility, a
world populated not just by billions of people and smart things but
also by a huge cloud of agents of many different kinds. Some of these
will be designed to make our lives easier by mediating our
interactions with the infrastructure, carrying out tasks and
Demos

181

Networks collection

4/1/04

4:35 PM

Page 182

Network logic

negotiating with other agents. Sometimes these entities may take on

personalities that engage in natural dialogue with people, and will
have the ability to evolve, adapt and learn.
These agents will be enabled by the billions of smart devices linked
through wireless networks to the utility and sometimes each other.
These devices will be able to form ad hoc networks that come
together for a specific, temporary purpose. For example, several cars
in proximity may want to talk to each other to avoid having a
potential crash on a particular stretch of road; a group of people who
have come together for a meeting may want their agents to set up
arrangements for them to work together on a set of documents and
presentations; a shopper may want to interact with specific products
in the supermarket he or she happens to be in. Smart devices will be
worn in your clothing, jewellery and accessories, and hop into the
utility via local wireless networks whenever they need to.
Any thing that uses power today will use the information utility
tomorrow. Any thing can be made smart and networked, and its
physical location and status tracked using location services such as
Global Positioning System (GPS) and cellphone networks, satellite
imaging and tagging technologies like Radio Frequency Identification
(RFID) tags. We already know how to keep track of millions, often
billions, of people, animals, vehicles, personal belongings, electrical
appliances and household objects. We cant (yet) cope with trillions of
trillions of insects, weeds, birds, leaves, flowers, rocks or stones. But
vast digital archive networks are now burgeoning into existence with
every conceivable kind of information, from molecular structures and
astronomical observatories to records of individuals genetic material,
their medical histories, financial transactions and electronic
communications.
And the infrastructure can also be connected to a huge range of
instruments and devices, from earth surveillance satellites that
monitor crops, water and weather, to CCTV cameras, webcams and
astronomical telescopes. These, and many others, offer real-time data
to users, acting as the eyes, ears and sensors of the global
infrastructure. In a very practicable sense this will support the
182

Demos

Networks collection

4/1/04

4:35 PM

Page 183

The information utility

instrumentation of the planet in a wholly unprecedented way. It will

become very hard for countries and companies to keep most of their
activities from being visible in detail and in real time to the rest of the
world.
Through the wall
Perhaps the most subtle and far-reaching impact of the information
utility will be its ability to support and mediate interactions between
people through the wall. Today, accessing a web page is a solitary
activity undertaken between a person and a website. Teleconferencing
and videoconferencing remain primitive and cumbersome, and
personal networks among people get very little support from ICTs
beyond phones and email.
One of the key advances of the information utility will therefore be
to support fluent collaboration and cooperation between groups and
teams of people: at work, socially, in families, at play. These
collaborative environments or col-laboratories will be the
infrastructure for teams of knowledge workers and citizens who trust
each other to work together to create new intellectual property or
social goods. It will be invaluable for lawyers working on a case, say,
or for a design team working on a new product. It will also be ideal
for supporting outsourcing and collaboration between companies
and organisations for example, where a contract requires people
from several organisations to work securely together for a limited
period of time.
The signs that this is already happening are with us. Each year we
are becoming more dependent on an extremely complex interconnected infrastructure, which supports more and more of our
everyday lives. This dependence is becoming crucial and irreversible.
Gradually we will be more and more unable to function without it. It
is instructive to look at a few examples.
Work
E-business and the continued development of outsourcing already
demand unprecedented levels of dynamic, fleet-of-foot inter-firm
Demos

183

Networks collection

4/1/04

4:35 PM

Page 184

Network logic

collaborations. The advent of secure collaborative environments will

accelerate the ability of people to work as individuals in several
different teams and companies, and for companies to have temporary,
and often casual and exploratory relationships with each other.
This will drive the need for new business models, audit practices
and concepts of intellectual property (IP). If a new animated movie
or commercial is created by an ad hoc collaboration of people using
fragments of pre-existing characters, personalities, story lines, music
and synthetic locations obtained from sources in the utility and
added to by the creative work of the team, then who should get paid
for what? We will need new systems for tracking, banking and
remunerating knowledge work. There may be a need for new forms of
micro-IP and micro-payments, with audit trails trusted by the
various participants.
Education and learning
The students and knowledge workers of tomorrow will be
permanently and wirelessly online to the information utility and each
other. In place of textbooks and ringbinders of notes, each person will
build up a lifelong personal information space comprising millions of
personal items and millions of persistent reliable links to web pages,
other information resources and other people. Accessing these
resources casually and continually as a matter of course in almost
every aspect of life will become as natural as turning on the radio or
TV or picking up the phone. You will require super-Google search
and management engines to find your way around it, and a very
smart forgettery to prune and discard material continually from
your memory space if complete information overload is to be
avoided. It will become increasingly difficult and unthinkable for
people to function without continuous access to their personal
information spaces. Together with new forms of human capital
assessment and accreditation, these might even make old style exams
obsolete providing a perfect record of your lifelong learning and
performance, which could be audited by an authorised certification
agent, in place of todays examinations and paper qualifications.
184

Demos

Networks collection

4/1/04

4:35 PM

Page 185

The information utility

Goods and services

This year, eBay will help 30 million people buy $20 billion worth of
goods and services. As e-retail services develop further we will be able
routinely to send one of our personal agents to get the best deal for us
on almost anything we want to procure, from babyfood and clothes
to cars, training courses or holidays. When we choose to shop in
person perhaps as an enjoyable leisure activity the smart devices
on our bodies will be able to talk to the screens and displays around
the store, and directly to the smarts on every single item in the store.
Disappointingly, nothing like the same attention has been paid to
the opportunities for improved delivery of public services, even
though productivity here lags far behind the private sector and there
is potentially huge value to be added. Current attempts, from online
income tax returns to hospital scheduling and combating benefits
fraud, however, do not augur well for genuinely ambitious
networking approaches to public sector modernisation or indeed
e-democracy. Instant phone referenda may work for Pop Idol and
Fame Academy but serious public issues would not be well served by
the same treatment.
Regulating the utility
The information utility will also bring inevitable opportunities for
malicious exploitation, and serious vulnerabilities and pathologies,
which need to be anticipated and managed. For example, security,
privacy and integrity are going to be crucial in this new era. In a
world where our interaction with the infrastructure (and with other
people and organisations through the infrastructure) is increasingly
mediated by trusted agents, brands, identity and authentication will
be of great importance. Identity theft and impersonation will have
bigger consequences. Conversely, retaining anonymity will become
increasingly hard, especially as anonymous cash transactions are
replaced by e-transactions, which inevitably leave a trace. If every
banknote has its own RFID tag, even cash will lose its anonymity.
This creates unique challenges for regulation. For example, how
can you regulate and police a bank that exists only in a computer in a
Demos

185

Networks collection

4/1/04

4:35 PM

Page 186

Network logic

satellite, and delivers its services to you via a wireless link from the
infrastructure wherever you are on the planet? Who actually owns it?
And when things go wrong, to whom do you complain? Related to
this is the question of data accuracy and integrity. One of the few
certainties in this future environment is that some of the data held in
the world wide utility will be inaccurate, out of date, or just plain
wrong, and some will get lost or destroyed. In particular, some of the
personal information held about you will be wrong, and we will need
quite radical innovations to enable individuals to find out and check,
and then get it put it right.
Sometimes this will be due just to malfunctions and accidents. But
in other cases there will be deliberate and continuous attempts to
subvert and attack the infrastructure for all sorts of criminal, terrorist
and other malicious purposes. As a result, we will have to develop
processes for continually cross-checking, purging, repairing and
restoring data. Like immune systems in living organisms, we will need
to counter both ageing from accumulated defects and deliberate
infections (such as viruses), as well as the damage to particular
components of the system caused by external traumas.
In our contact with the e-world we will rely on trusted third parties
with trusted brands to validate the infrastructure for us, and to
validate us to it. These will play a key role as the super agents we trust
to look after our data and our identity in the face of whatever threats
confront the infrastructure. In a world where services can be delivered
by anyone from anywhere, trusted brands will be vital. Everything else
behind the wall all the companies, organisations and public
institutions will be virtual, fast changing, elusive and evanescent.
Living in the goldfish bowl
The promise of this new era is immense. The intelligent
infrastructure will enable us to manage everything from our personal
time to the resources of the whole planet more optimally and
effectively. But the threats are also great, not least the risk of complete
dependency on the information utility in our personal lives, at work,
and when we travel, or attempt to access education, healthcare and
186

Demos

Networks collection

4/1/04

4:35 PM

Page 187

The information utility

other public services. The price of this dependency is that, as

individuals, we will have to come to terms with a major surrender of
personal privacy. As employees and managers in companies, public
institutions and organisations of all kinds, we will face the same issue.
Our most confidential emails will be undeletable and the global
activities of the organisations we work for will be instantly visible to
almost anyone who wants to know. As things become more densely
interconnected, time constants and constraints will shrink and we
will have less and less chance to think, reflect and make balanced
decisions. In short, we are entering the era of transparency. This raises
important questions concerning how we will cope with living and
working in the global goldfish bowl. Will we welcome it and embrace
it, or just tolerate it most of the time? Will it be possible to opt out
will we see the emergence of a disaffected out-class on an
unprecedented scale?
We will have to use the information infrastructure itself to help us
discover, debate and resolve our responses to these challenges. As it
grows rapidly larger and more complex than human brains and
bodies, that infrastructure will require all of the defence mechanisms
evolved by living organisms in order to adapt to and protect themselves from their environments.
One thing is clear: a long wave of disruptive technological
innovation is coming. The question is what will it look like and when
will it emerge? This cannot be predicted in detail; there is always the
possibility of unexpected and even more powerful technologies
emerging. But that does not mean that they cannot be shaped by the
wider political, economic and social environment. How we
incentivise investment in building the infrastructure; how we influence public attitudes to the utility and its impacts; how we manage its
legal, ethical and social implications; how we craft the appropriate
regulatory framework; all are crucial questions facing developed and
developing countries in years to come. Policy-makers and business
leaders in the UK must begin to grasp the reality of the era of the
information utility if its opportunities, and its threats, are to be
appreciated and engaged.
Demos

187

Networks collection

4/1/04

4:35 PM

Page 188

Network logic

Professor Sir John Taylor completed his five-year term as Director

General of Research Councils and Chairman of Research Councils UK in
December 2003. He is currently Chairman of Roke Manor Research.

188

Demos

Networks collection

4/1/04

4:35 PM

Page 189

Network Logic 15

Smart mobs arent

a thing that you
can point to with
one finger or
describe with two
words, any more
than you could
the Internet . . .
Smart mobs
Howard Rheingold

Networks collection

4/1/04

4:35 PM

Page 190

Networks collection

4/1/04

4:35 PM

Page 191

15. Smart mobs

the power of the mobile many
Howard Rheingold

Smart mobs consist of people who are able to act in concert even if
they dont know each other. The people who make up smart mobs
cooperate in ways never before possible because they carry devices
that possess both communication and computing capabilities.
An unanticipated convergence of technologies is suggesting new
responses to civilisations founding question, How can competing
individuals learn to work cooperatively? Location-sensing wireless
organisers, wireless networks and community supercomputing
collectives all have one thing in common: they enable people to act
together in new ways and in situations where collective action was not
possible before.
The killer apps of tomorrows mobile infocom industry wont be
hardware devices or software programmes but social practices. The
most far-reaching changes will come, as they often do, from the kinds
of relationships, enterprises, communities and markets that the
infrastructure makes possible.
Netwar Dark and Light
On 20 January 2001, President Joseph Estrada of the Philippines
became the first head of state in history to lose power to a smart mob.
Following the abrupt ending of his impeachment trial by sympathetic
senators, Manila residents began to assemble in their thousands on
Epifanio de los Santas Avenue (known as Edsa), the site of the 1986
Demos

191

Networks collection

4/1/04

4:35 PM

Page 192

Network logic

People Power peaceful demonstrations that had toppled the Marcos

regime. Within 75 minutes, 20,000 people had converged on Edsa,
mobilised and coordinated by waves of text messages initiated by
opposition leaders: Go 2EDSA, Wear blck. Over four days, more than
a million people showed up, mostly dressed in black. The military
withdrew support from the regime; the Estrada government fell, as
the Marcos regime had fallen a decade previously, largely as a result of
massive non-violent demonstrations. The rapid assembly of the antiEstrada crowd was a hallmark of early smart mob technology, and the
millions of text messages exchanged by the demonstrators in 2001
were, by all accounts, a key to the crowds esprit de corps. The legend
of Generation Txt was born.
Bringing down a government without firing a shot was a momentous early eruption of smart mob behaviour. It wasnt, however, the
only one.

192

On 30 November 1999, autonomous but internet-worked

squads of demonstrators protesting at the meeting of the
World Trade Organisation (WTO) used swarming
tactics, mobile phones, websites, laptops and PDAs to win
The Battle of Seattle.
In September 2000, thousands of citizens in Britain,
outraged by a sudden rise in gasoline prices, used mobile
phones, SMS, email from laptop PCs and CB radios in
taxicabs to coordinate dispersed groups that blocked fuel
delivery at selected service stations in a wildcat political
protest.
A violent political demonstration in Toronto in the spring
of 2000 was chronicled by a group of roving
journalistresearchers who webcast digital video of
everything they saw.
Since 1992, thousands of bicycle activists have assembled
monthly for Critical Mass moving demonstrations,
weaving through San Francisco streets en masse. Critical
Mass operates through loosely linked networks, alerted by
Demos

Networks collection

4/1/04

4:35 PM

Page 193

Smart mobs

mobile phone and email trees, and breaks up into smaller,

tele-coordinated groups when appropriate.
The Battle of Seattle saw a more deliberate and tactically focused use
of wireless communications and mobile social networks in urban
political conflict, more than a year before texting mobs assembled in
Manila. A broad coalition of demonstrators who represented different
interests but were united in opposition to the views of the World
Trade Organisation planned to disrupt the WTOs 1999 meeting in
Seattle. The demonstrators included a wide range of different affinity
groups who loosely coordinated their actions around their shared
objective. The Direct Action Network enabled autonomous groups to
choose which levels of action to participate in, from non-violent
support to civil disobedience to joining mass arrests a kind of
dynamic ad hoc alliance that wouldnt have been possible without
a mobile, many-to-many, real-time communication network.
According to a report dramatically titled Black flag over Seattle by
Paul de Armond:
The cohesion of the Direct Action Network was partly due to
their improvised communications network assembled out of cell
phones, radios, police scanners and portable computers.
Protesters in the street with wireless Palm Pilots were able to link
into continuously updated web pages giving reports from the
streets. Police scanners monitored transmissions and provided
some warning of changing police tactics. Cell phones were widely
used.3
From Seattle to Manila, the first netwars have already broken out.
The term netwar was coined by John Arquilla and David Ronfeldt,
two analysts for the RAND corporation, who noticed that the same
combination of social networks, sophisticated communication technologies, and decentralised organisational structure was surfacing as
an effective force in very different kinds of political conflict:

Demos

193

Networks collection

4/1/04

4:35 PM

Page 194

Network logic

Netwar is an emerging mode of conflict in which the

protagonists ranging from terrorist and criminal organisations
on the dark side, to militant social activists on the bright side
use network forms of organisation, doctrine, strategy, and
technology attuned to the information ageThese networks are
proving very hard to deal with; some are winning. What all have
in common is that they operate in small, dispersed units that can
deploy nimbly anywhere, anytime.4
The swarming strategies noted by Arquilla and Ronfeldt rely on
many small units like the affinity groups in the Battle of Seattle.
Individual members of each group remained dispersed until mobile
communications drew them to converge on a specific location from
all directions simultaneously, in coordination with other groups.
Manila, Seattle, San Francisco, Senegal and Britain were sites of
non-violent political swarming. Arquilla and Ronfeldt cited the
non-governmental organisations associated with the Zapatistas
movement in Mexico, which mobilised world opinion in support
of Indian peasants, and the Nobel Prize-winning effort to enact an
anti-landmine treaty as examples of non-violent netwar actions.
Armed and violent swarms are another matter. The Chechen rebels
in Russia, soccer hooligans in Britain and the FARC guerrillas
in Colombia also have used netwar strategy and swarming tactics.5
The US military is in the forefront of smart mob technology
development.
Smart mobs engaging in either violent or non-violent netwar
represent only a few of the many possible varieties of smart mob.
Netwars do share a similar technical infrastructure with other smart
mobs. More importantly, however, they are both animated by a new
form of social organisation, the network. Networks include nodes and
links, use many possible paths to distribute information from any link
to any other, and are self-regulated through flat governance
hierarchies and distributed power. Arquilla and Ronfeldt are among
many who believe networks constitute the newest major social
organisational form, after tribes, hierarchies and markets. Although
194

Demos

Networks collection

4/1/04

4:35 PM

Page 195

Smart mobs

network-structured communications hold real potential for enabling

democratic forms of decision-making and beneficial instances of
collective action, that doesnt mean that the transition to networked
forms of social organisation will be a pleasant one with uniformly
benevolent outcomes. Arquilla and Ronfeldt note the potential for
cooperation in examples like the non-governmental organisations
that use netwar tactics for public benefit, but they also articulated a
strong caution, worth keeping in mind when contemplating the
future of smart mobs:
Most people might hope for the emergence of a new form of
organisation to be led by good guys who do the right thing and
grow stronger because of it. But history does not support this
contention. The cutting edge in the early rise of a new form may
be found equally among malcontents, neer-do-wells, and clever
opportunists eager to take advantage of new ways to manoeuvre,
exploit, and dominate.6
Lovegety and peer-to-peer journalism
In light of the military and terrorist potential of netwar tactics it
would be foolish to presume that only benign outcomes should be
expected from smart mobs. But any observer who focuses exclusively
on the potential for violence would miss evidence of perhaps even
more profoundly disruptive potential for beneficial as well as
malign purposes of smart mob technologies and techniques. Could
cooperation epidemics break out if smart mob media spread beyond
warriors to citizens, journalists, scientists, people looking for fun,
friends, mates, customers or trading partners?
Consider a few experiments on the fringes of mobile communications that might point towards a wide variety of non-violent smartmobbing in the future:

Interpersonal awareness devices have been evolving for

several years. Since 1998 hundreds of thousands Japanese
have used Lovegety keychain devices that signal when
Demos

195

Networks collection

4/1/04

4:35 PM

Page 196

Network logic

another Lovegety owner of the opposite sex and

compatible profile is within 15 feet.
ImaHima (are you free now?) enables hundreds of
thousands of Tokyo i-mode users to alert buddies who are
in their vicinity at the moment.
Upoc (universal point of contact) in Manhattan
sponsors mobile communities of interest: any member of
Manhattan celebrity watch, nyc terrorism alert, prayer of
the day or The Resistance, for example, can broadcast
text messages to and receive messages from all the other
members.
Phones that make it easy to send digital video directly to
the web make it possible for peer-to-peer journalism
networks to emerge; Steve Manns students in Toronto
have chronicled newsworthy events by webcasting
everything their wearable cameras and microphones
capture.
Researchers in Oregon have constructed social
middleware that enables wearable computer users to
form ad hoc communities, using distributed reputation
systems, privacy and knowledge-sharing agents, and
wireless networks.

In 2000 WearComp researcher, innovator and evangelist Steve Mann

launched ENGwear, an experiment in wearable news-gathering
systems conducted by students and researchers at the Humanistic
Intelligence Lab at the University of Toronto.7 In the spring of 2000
Mann and a group of his students, all wearing computers equipped
with EyeTaps, which broadcast everything they saw and heard to the
Web, showed up at a demonstration in Toronto called by the Ontario
Coalition Against Poverty (OCAP). Violence broke out. Mann
reported:
We, along with the journalists and various television crews, ran
for cover. However, unlike the reporters, my students and I were
196

Demos

Networks collection

4/1/04

4:35 PM

Page 197

Smart mobs

still broadcasting, capturing almost by accident the entire event.

Whatever we saw before us was captured and sent instantly in
real time to the World Wide Web, without our conscious thought
or effort.8
Swarm intelligence and the social mind
Massive outbreaks of cooperation precipitated the collapse of
communism. In city after city, huge crowds assembled in non-violent
street demonstrations, despite decades of well-founded fear of
political assembly. Although common sense leads to the conclusion
that unanimity of opinion among the demonstrators explained the
change of opinion, Natalie Glance and Bernardo Huberman, Xerox
PARC researchers who have studied the dynamics of social systems,
noted that a diversity of cooperation thresholds among the
individuals can tip a crowd into a sudden epidemic of cooperation.
Glance and Huberman pointed out that a minority of extremists can
choose to act first and, if the conditions are right, their actions can
trigger actions by others who needed to see somebody making the
first move before acting themselves at which point the bandwagonjumpers follow the early adopters who followed the first actors.9
Sudden epidemics of cooperation arent necessarily pleasant
experiences. Lynch mobs and entire nations cooperate to perpetrate
atrocities. Decades before the fall of communism, sociologist Mark
Granovetter examined radical collective behaviour of both positive
and negative kinds and proposed a threshold model of collective
behaviour. I recognised Granovetters model as a crucial conceptual
bridge that connects intelligent (smart mob) cooperation with
emergent behaviours of unintelligent actors, such as hives, flocks and
swarms.
Granovetter studied situations in which individuals were faced
with eitheror decisions regarding their relationship to a group
whether or not to join a riot or strike, adopt an innovation, spread a
rumour, sell a stock, leave a social gathering, migrate to a different
country. He identified the pivotal statistic as the proportion of other
people who would have to act before an individual decides to join them.
Demos

197

Networks collection

4/1/04

4:35 PM

Page 198

Network logic

One of Granovetters statements yielded a clue to smart mob

dynamics: By explaining paradoxical outcomes as the result of
aggregation processes, threshold models take the strangeness often
associated with collective behaviour out of the heads of actors and
put it into the dynamics of situations.10
Threshold models of collective action are about media for
exchange of coordinating knowledge. Understanding this made it
possible to see something I had not noticed clearly enough before a
possible connection between computer-wearing social networks of
thinking, communicating humans and the swarm intelligence of
unthinking (but also communicating) ants, bees, fish, and birds.
Individual fish and birds (and tight-formation fighter pilots) school
and flock simply by paying attention to what their nearest neighbours
do. The coordinated movement of schools and flocks is a dynamically
shifting aggregation of individual decisions. Even if there were a
central tuna or pigeon who could issue orders, no system of
propagating orders from a central source can operate swiftly enough
to avoid being eaten by sharks or slamming into trees. When it comes
to hives and swarms, the emergent capabilities of decentralised selforganisation can be surprisingly intelligent.
What happens when the individuals in a tightly coordinated group
are more highly intelligent creatures rather than simpler organisms
like insects or birds? How do humans exhibit emergent behaviour?
Kevin Kelly traced back the new theories regarding emergent
properties to William Morton Wheeler, an expert in the behaviour of
ants.11 Wheeler called insect colonies superorganisms and defined
the ability of the hive to accomplish tasks that no individual ant or
bee is intelligent enough to do on its own as emergent properties of
the superorganism. Kelly drew parallels between the ways both
biological and artificial vivisystems exhibit the same four
characteristics of what he called swarm systems:

198

the absence of imposed centralised control

the autonomous nature of sub-units
the high connectivity between the sub-units
Demos

Networks collection

4/1/04

4:35 PM

Page 199

Smart mobs

the webby non-linear causality of peers influencing

peers.12

Steven Johnsons 2001 book Emergence shows how the principles that
Kelly extrapolated from biological to technological networks also
apply to cities and Amazon.coms recommendation system: In these
systems, agents residing on one scale start producing behaviour that
lies on one scale above them: ants create colonies; urbanites create
neighbourhoods; simple pattern-recognition software learns how to
recommend new books. The movement from low-level rules to
higher level sophistication is what we call emergence.13 In the case of
cities, although the emergent intelligence resembles the ant mind, the
individual units, humans, possess extraordinary onboard intelligence
or at least the capacity for it.
At this point, connections between the behaviour of smart mobs
and the behaviour of swarm systems must be tentative, yet several of
the earliest investigations have shown that the right kinds of online
social networks know more than the sum of their parts: connected
and communicating in the right ways, populations of humans can
exhibit a kind of collective intelligence.
There have been various theories about the internet as the nervous
system of a global brain, but Bernardo Huberman and his colleagues
at Hewlett-Packards Information Dynamics research laboratory have
made clever use of markets and game simulations as computational
test beds for experiments with emergent group intelligence.
Huberman and his colleagues have used information markets to
perform experiments in emergent social intelligence. The Hollywood
Stock Exchange, for example, uses the market created from the
trading of symbolic shares to predict box office revenues and Oscar
winners. They have found that group forecasts were more accurate
than those of any of the individual participants forecasts.14 The HP
research team makes the extraordinary claim that they have created a
mathematically verifiable methodology for extracting emergent
intelligence from a group and using the groups knowledge to predict
the future in a limited but useful realm: One can take past predictive
Demos

199

Networks collection

4/1/04

4:35 PM

Page 200

Network logic

performance of participants in information markets and create

weighting schemes that will predict future events, even if they are not
the same event on which the performance was measured.15
Decades ago, computer scientists thought that some day there
would be forms of artificial intelligence but, with the exception of a
few visionaries, they never thought in terms of computer-equipped
humans as a kind of social intelligence. Although everyone who
understands the use of statistical techniques to make predictions
hastens to add a disclaimer that surprises are inevitable, and one of
the fundamental characteristics of complex adaptive systems is their
unpredictability, the initial findings that internet-worked groups of
human beings can exhibit emergent prediction capabilities are
potentially profound.
Another research group that takes emergent group intelligence
seriously is the laboratory at Los Alamos, where a group of artificial
life researchers issued a report in 1998, Symbiotic intelligence: selforganising knowledge on distributed networks, driven by human
interaction.16 The premise of this interdisciplinary team is based on
the view proposed by some in recent years that human society is an
adaptive collective organism and that social evolution parallels and
unfolds according to the same dynamics as biological evolution.17
According to this theory, new knowledge and new technologies have
made possible the evolution of the maximum size of the functioning
social group from tribes to nations to global coalitions. The
knowledge and technology that triggered the jump from clan to tribe
to nation to market to network all shared one characteristic: they each
amplified the way individual humans think and communicate, and
magnified their ability to share what they know.
The research conducted directly by the Los Alamos researchers
reinforced Huberman et als claims that groups of humans, linked
through online networks, can make collective decisions that prove
more accurate than the performance of the best individual predictors
in the group. If it isnt a dead end, the lines of research opened by
Hubermans team, the Los Alamos researchers and others could
amplify the powers of smart mobs into entirely new dimensions of
200

Demos

Networks collection

4/1/04

4:35 PM

Page 201

Smart mobs

possibility, the way Moores Law amplified the powers of computer

users.
Conclusion
Will self-organised, ad hoc networks of computer wearers, mediated
by privacy-protecting agents, blossom into a renaissance of new
wealth, knowledge and revitalised civil society, or will the same
technologicalsocial regime provide nothing more than yet another
revenue stream for Disinfotainment, Inc.?
Or is that the wrong question? Given the direction of the
technological, economic and political changes I have touched on, I
propose the following questions:

What do we know now about the emergent properties of

ad hoc mobile computing networks, and what do we need
to know in the future?
What are the central issues for individuals in a world
pervaded by surveillance devices in terms of what we
can do about it?
What are the long-term consequences of near-term
political decisions on the way well use and be affected by
mobile, pervasive, always-on media?

Smart mobs arent a thing that you can point to with one finger or
describe with two words, any more than the internet was a thing
you could point to. The internet is what happened when a lot of
computers started communicating. The computer and the internet
were designed, but the ways people used them were not designed into
either technology, nor were the most world-shifting uses of these
tools anticipated by their designers or vendors. Word processing and
virtual communities, eBay and e-commerce, Google and weblogs and
reputation systems emerged. Smart mobs are an unpredictable but at
least partially describable emergent property that I see surfacing as
more people use mobile telephones, more chips communicate with
each other, more computers know where they are located, more
Demos

201

Networks collection

4/1/04

4:35 PM

Page 202

Network logic

technology becomes wearable, and more people start using these new
media to invent new forms of sex, commerce, entertainment,
communion and, as always, conflict.
Howard Rheingold is the author of Smart Mobs: the next social
revolution, from which this essay is extracted.
Notes
1
2
3
4

5
6
7
8
9
10
11
12
13
14

15
16

202

V Rafael, The cell phone and the crowd: messianic politics in recent Philippine
history, 13 June 2001, available at
http://communication.ucsd.edu/people/f_rafael.cellphone.html.
R Lloyd Parry, The TXT MSG revolution, Independent, 23 Jan 2001, available at
www.independent.co.uk/story.jsp?story=51748 (accessed 1 Mar 2002).
P de Armond, Black flag over Seattle, Albion Monitor 72, Mar 2000, available at
www.monitor.net/monitor/seattlewto/index.html (accessed 1 Mar 2002).
D Ronfeldt and J Arquilla, Networks, netwars, and the fight for the future, First
Monday 6, 10 October (2001),
http://firstmonday.org/issues/issue6_10/ronfeldt/index.html (accessed 1 Mar
2002).
Ibid.
J Arquilla and D Ronfeldt (eds) Networks and Netwars: the future of terror,
crime, and militancy (Santa Monica, Ca: RAND, 2001).
ENGwear: wearable wireless systems for electronic news gathering, available at
www.eyetap.org/hi/ENGwear (accessed 1 Mar 2002).
S Mann and H Niedzviecki, Cyborg: digital destiny and human possibility in the
age of the wearable computer (Mississauga: Doubleday Canada, 2001).
N Glance and B Huberman, The dynamics of social dilemmas, Scientific
American, March 1994.
M Granovetter, Threshold models of collective behaviour, American Journal of
Sociology 83, no 6 (1978).
WM Wheeler, Emergent Evolution and the Development of Societies (New York:
WW Norton, 1928).
Ibid.
S Johnson, Emergence: the connected lives of ants, brains, cities and software (New
York: Scribner, 2001).
K Chen, L Fine and B Huberman, Forecasting uncertain events with small
groups, HP Laboratories, Palo Alto, California, 3 Aug 2001,
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=278601 (accessed 6 March
2002)
Ibid.
N Johnson et al, Symbiotic intelligence: self-organising knowledge on
distributed networks, driven by human interaction, in C Adami et al (eds)
Artificial Life VI: proceedings of the Sixth International Conference on Artificial

Demos

Networks collection

4/1/04

4:35 PM

Page 203

Smart mobs

17

Life (Complex Adaptive Systems, No 6) (Cambridge: Bradford Books/MIT

Press, 1998).
G Dyson, Darwin Among the Machines: the evolution of global intelligence
(Reading, MA: Addison-Wesley, 1997).

Demos

203

Networks collection

4/1/04

4:35 PM

Page 204

Networks collection

4/1/04

4:35 PM

Page 205

Network Logic 16

Social movements
have always had an
impact on politics.
In an increasingly
connected society,
a new breed the
network campaign
has emerged . . .
The rise of network
campaigning
Paul Miller

Networks collection

4/1/04

4:35 PM

Page 206

Networks collection

4/1/04

4:35 PM

Page 207

16. The rise of network

campaigning
Paul Miller

16 May 1998 was the day the network came alive. By coach, car, train,
boat, bike and foot, 70,000 people converged on Birmingham where
the G8 leaders, the most powerful men on the planet, were meeting in
a steel, glass and concrete building especially fortified for the occasion
by hundreds of CIA officers. At 3pm, the campaigners formed a
human chain some 10km long, encircling the security-cordoned
conference centre. But the men and women who had criss-crossed the
UK to join hands werent there to protest about a high-profile issue,
decision or event in the news. Instead they wanted to make their
feelings known about a complicated and, at the time, obscure matter
of economic policy. They were supporters of the Jubilee 2000
campaign to cancel the unpayable debts of the worlds poorest
nations.
Social movements have always had an important effect on our
political systems but in an increasingly connected society a new breed
the network campaign has emerged. On issues from the
environment and human rights to poverty eradication and debt
reduction, network campaigns have taken on some of the biggest and
most powerful institutions on the planet: from governments and
multinational companies to the World Trade Organisation and the
World Bank. Connecting non-governmental organisations, faith
groups and trade unions as well as individual campaigners, networks
have emerged that combine the resources, powers, skills and
Demos

207

Networks collection

4/1/04

4:35 PM

Page 208

Network logic

experience of diverse institutions in new and potent combinations.

While there has been no network theorist behind their strategies,
campaigners have learned that networks offer a number of
advantages over other institutional models. Over time, they have
become their weapon of choice.
This essay argues that politics has much to understand and learn
about network campaigning. It begins by examining the structure of
the network campaigns, then goes on to look at the reasons for the
use of networks by campaigners over other organisational forms
before looking at some of the downsides network campaigning
presents. It draws on examples from the Jubilee 2000 campaign,
perhaps the first fully fledged network campaign to reach public
consciousness.1
The characteristics of network campaigns
First we need to consider what makes a campaign a network
campaign. Simple definitions are difficult since it is partly the
looseness of institutional structure and the diversity of tactics that
make network campaigns different to traditional approaches. In
essence, network campaigning allows a diverse grouping of organisations and individuals to participate through commitment to a
shared purpose, while remaining autonomous individual agents. In
this way it is possible to gain additional leverage over decisionmaking bodies through the multiplier effect of a coherent message
and more efficient deployment of resources and effort, while
maintaining the flexibility and energy that more bureaucratic forms
of coordination tend to squander. The characteristics and common
features of network campaigns include the following features.
Having a shared goal

The main characteristic of campaigning networks is that they have a

goal shared by those who are part of the network. As David Ronfeldt
says, The network form offers its best advantages where the
members, as often occurs in civil society, aim to preserve their
autonomy and to avoid hierarchical controls, yet have agendas that
208

Demos

Networks collection

4/1/04

4:35 PM

Page 209

The rise of network campaigning

are interdependent and benefit from consultation and coordination.2

The meaning and goal of network campaigns is usually so simple to
express and communicate that they are often seen as single issue
campaigns. For Jubilee 2000, the shared goal was a debt-free start for
a billion people.
Being structure light

Network campaigns tend to be very light on traditional structure with

only a small secretariat or coordinating body compared to the size of
the network or the resources that it is able to mobilise. The role of the
secretariat becomes that of network nurturer. Secretariats must spend
their time creating connections between other people, and
channelling the energy and enthusiasm that is created in the network
towards increasing the cumulative impact on decision-makers. The
central secretariat can also help the network to interpret the overall
environment for the campaign by providing research and communicating feedback from political targets. Often networks have
some form of democratic involvement from members, again
coordinated by the secretariat. This can take the form of a
management board, with some positions elected by network
members and others filled by the founders of the campaign.
Mobilising a coalition of skills and resources

Each of the most prominent examples of networked campaigning

over the last decade has drawn on established organisations and
linked them together in a coalition of the wilful. Coalition members
provide resources to fund the secretariat and access to their
memberships and communication channels. However, individuals
dont have to be part of one of the coalition organisations to be part
of the campaign. There are always easy ways to get involved, from
simply signing a petition to show your support, through attending
demonstrations, to letting it take over your life as a full-time
member of staff or volunteer. The network will also include people
with a variety of different skills that can be drawn upon at short
notice.
Demos

209

Networks collection

4/1/04

4:35 PM

Page 210

Network logic

Use of network technology

The internet has been vital to network campaigners and novel uses of
the web, email or newsgroups are a common thread through several
of the network campaigns of the late 1990s. Technology also plays a
key role in recording events and important moments for the
campaign. Since it is likely that the central secretariat wont be well
enough resourced to be at every local demonstration or meeting,
small groups of activists can use technology to swap pictures, video or
recordings of events to give a rich picture of the level of activity
within the network. The Jubilee 2000 website provided between 8,000
and 12,000 people with up-to-date information about the campaign
every week. A webchat with Bono run by Jubilee 2000 in conjunction
with MSN received hundreds of thousands of hits.
Embracing diversity and openness

An approach used by network campaigners is to communicate using

a variety of different media, each allowing them to connect to new
constituencies. So network campaigns dont just rely on, say,
producing a magazine or holding a public meeting: they have a
variety of ways of communicating to match the variety of potential
participants. While it would be very unusual for organisations with
completely opposite goals to the rest of the network to be allowed to
join, it is not necessary for network members to agree on everything.
Network members need to be compatible in their views, not identical.
Jubilee 2000 brought together organisations as diverse as Friends of
the Earth and the Mothers Union.
Use of celebrity

The power of TV can reach the parts that social networks cannot. In
an era of broadcast media concentration and satellite news channels
that reach across the world, campaigners have needed a way of
accessing and harnessing these networks. The solution they have
identified has been to use something that is common currency in
these networks a bit of stardust. When the organisers of the Brit
awards decided to promote the Jubilee 2000 campaign after both
210

Demos

Networks collection

4/1/04

4:35 PM

Page 211

The rise of network campaigning

Muhammad Ali and Bono said they would attend, their message
promoting the campaign reached a global TV audience of over 100
million people as well as widespread coverage in the press.
Use of physical space

The tactic of network campaigners has often been to foster a large

and diverse network and then bring as many members of the
networks as possible together in one place. The vibrant atmosphere
on these occasions does not just look impressive for the TV cameras;
it has also given campaigns an additional injection of energy. It has
given people the motivation to go away and create new network
connections. Jubilee 2000 took the human chain as its symbol of
choice when bringing people together.
Personalisation of targets

While campaigners might want to include as many people as possible

in their network, they see the value of focusing their efforts on one or
two individuals who will feel the full weight of the networks efforts.
In the UK, it was Tony Blair and Gordon Brown who took the flak
from the Jubilee 2000 campaign. The mail rooms at Number 10
Downing Street and the Treasury discovered first-hand how large the
Jubilee 2000 network had become when postcards from the network
began to arrive in their tens of thousands.
Being time limited

A sense of urgency is the network campaigners best friend. Jubilee

2000 took the end of the millennium as a deadline for world leaders
to make a decision, taking their inspiration from the biblical idea of
jubilee, the time when debts would be forgiven and slaves freed every
50 years. While social links created through one particular network
campaign might live on to be used another day, individual network
campaigns disappear as quickly as they are created.
Network campaigning attractors
Why has the network become the organisational form of choice for
Demos

211

Networks collection

4/1/04

4:35 PM

Page 212

Network logic

campaigners? One way of understanding this is to look at the

advantages that this mode of organisation offers.
Its cheaper

Setting up new institutions is expensive. If you have a cause to fight,

the requirements to build an organisational infrastructure able to
communicate with millions of people are huge. Network campaigns
make up for their shortfall in terms of financial muscle by tapping
into the existing infrastructure that large NGOs or civil society
organisations already possess. Using these latent networks and
institutional resources as much as possible helps to keep costs to a
minimum. If they do have a central secretariat at all, network
campaigns do not have a large number of paid staff (Jubilee 2000 had
at most 25 paid staff and for much of its lifetime even fewer).
Its quicker

A campaign can be built up very quickly through a network when it

could have taken decades by other means. This is partly because
campaigners dont need to raise as much money, but also because the
agreement of a large number of people is not required before doing
anything decision-making structures are usually very light. And
once a network has reached a critical mass, the campaign can move
very quickly indeed. Like an epidemic, the message spreads
(sometimes uncontrollably) through social networks, new
connections reinforcing old connections. It took just four years for
Jubilee 2000 to grow from being a UK-based organisation with a
single employee working out of a shed on Londons South Bank to an
international force to be reckoned with, with 69 national campaigns
and 24 million signatures on its petition.
It provides eyes everywhere

Network campaigns provide an army of researchers sending each

other bits of information all the time. In particular, the internet has
revolutionised links between campaigners across the globe. On 13
April 2000, 63 campaigners were arrested in Kenya during a debt
212

Demos

Networks collection

4/1/04

4:35 PM

Page 213

The rise of network campaigning

march. News of their arrest was circulated via the internet, prompting
letters of protest from around the world. On 22 May of the same year
the charges were dropped. Andre Hotchkiss, one of the arrested
marchers said: Without the avalanche of email, fax, and letters that
poured into Kenya, this thing may have pushed on for a longer time.
Its more fun

Networks mean meeting new people and developing new friendships,

often with people who you would never have come into contact with
otherwise. Network campaigners make sure that there are plenty of
chances to communicate and meet with others who share their goal.
Physical gathering points are usually best, be they church meetings,
student activist groups or informal groups meeting in peoples
homes. The traditional image of campaigns delivering leaflets on a
soggy winters day, or hanging around at aggressive picket lines has
given way to a more colourful, eye-catching carnival atmosphere
where a sense of humour is always present.
Ultimately, its more effective

Described by Anthony Gaeta, a spokesman for the World Bank, as

one of the most effective global lobbying campaigns I have ever seen,
Jubilee 2000 surprised many of the institutions in the firing line with
the sheer energy and enthusiasm they were able to generate.3 The
network not only enabled Jubilee 2000 to bring together 70,000
people for one moment in 1998 or to gather 24 million petition
signatures but also put the issue of debt high up on the agenda of
subsequent international meetings and resulted in over $36 billion of
debts being cancelled. The fact that many of the organisations
involved in Jubilee 2000 in the UK are now part of the Trade Justice
Movement, a network campaign using many of the tactics of Jubilee
2000 to campaign for fairer trade rules between developed and
developing nations, perhaps also shows that they themselves see the
advantages of network campaigning for getting results.

Demos

213

Networks collection

4/1/04

4:35 PM

Page 214

Network logic

The challenges of networks

But networks do have downsides. Accountability, for example, is often
messy in networks, not easily corresponding to conventional ideas of
due process or democracy. The qualification for inclusion in a
network is enthusiasm and a willingness to work with others, but this
can develop to a point where the people who are the most
enthusiastic and most connected the hubs can dominate. These
are people who move between established groups, passing on
information, encouraging collaboration and fixing meetings. They
could be housewives or headmasters, accountants or Baptist ministers
and they are vital to the success of network campaigns. But since they
are often neither employed nor elected by the campaign they can be
very difficult to hold to account. While most network campaigns have
some democratic functions built into their very light structures, they
can be dogged by allegations that particular personalities are allowed
to dominate. Within international networks problems of strength of
voice can cause tensions, particularly between campaigners in the
South and those in the North. In extreme circumstances, this can lead
to a second challenge in campaigning networks forking. This is a
problem also seen in open source software programming, where
disagreement over the overall goal or aim of the network leads to it
splitting into two or more parts.
Another challenge is that once networks have grown they become
harder to maintain. Success can be expensive. As networks become
more effective and their activity more visible, traditional political
actors such as governments, corporate institutions or the media look
for a central point to ask questions and, if they are seeking to quash
the network, to focus their firepower. This leads to what some
campaigners call death by consultation, where the secretariat at the
centre is distracted from its central goal of nurturing the network by
the need to invest more and more time, people and resources in
servicing requests from the outside world. Just at the moment that
success in terms of profile has been achieved (a prerequisite for largescale networks) secretariats can grow very rapidly and demands for
the secretariat to become run like a normal organisation grow.
214

Demos

Networks collection

4/1/04

4:35 PM

Page 215

The rise of network campaigning

Network campaigns can also burn out before their goals are
achieved. Members of the network need to have something they can
individually achieve in order to see the value in their activity. But if
people see change happening quickly, and their cause getting
widespread coverage in the media, they think theyve done enough.
For this reason network campaigns tend to have a high-water mark
a point where they are at the peak of their influence, when they can
do most. After that they can find it hard to survive.
Finally and, looking to the future, perhaps most importantly, is the
question of whether network campaigns can become a constructive
force for change in their own right, or whether they will remain
essentially parasitic on existing institutional structures, policy tools
and power bases. Can they learn to deliver solutions to problems,
rather than just hoping that by shouting loud enough and long
enough they will get solved by someone else? Networks seem to be
very good at undermining more traditional organisational forms but,
so far, surprisingly few have made the transition to constructing new
positive institutions. The criticism levelled at campaigners by
governments or multilateral institutions is could they do any better?
The World Social Forum is an attempt by a huge swathe of civil
society groups to do just this. If successful, it could prove to be a
massive step forward for network-based civil society solutions.
The future of network campaigning
So what does the future hold for network campaigning? What will be
the longer-term effects of repeated use of network campaigning? Will
it influence levels of activity in civil society more broadly, for
instance, increasing (or decreasing) membership of more traditional,
vertically organised civil society institutions like the large NGOs? If
people see the network campaigns as more effective than the
institutions they are based upon, will they continue to support those
institutions? And are network campaigns getting more effective over
time as they learn from previous campaigns? These are all questions
to be researched and understood.
But the most important area of development will be in the political
Demos

215

Networks collection

4/1/04

4:35 PM

Page 216

Network logic

domain. Increasing use of network campaigning could pose a

significant threat to national governments. We have already seen the
potential political power of network campaigns in developing
countries.4 In 2003 the UK Stop The War Coalition brought more
than a million people on to the streets of London. If network
campaigns continue to grow in terms of their numerical power and
the sophistication of campaigning techniques, it could only be a
matter of time before a major Western government is brought down
by a network campaign. Networks have provided civil society
organisations with a way of handling organisational and logistical
complexity that governments have yet to embrace.
Political parties themselves will need to learn from the network
campaigners. As Manuel Castells has written: Mass political parties,
when and where they still exist, are empty shells, barely activated as
electoral machines at regular intervals.5 The rapid shrinkage in the
membership of political parties in the UK should be cause for
concern, and the simultaneous growth in the prominence of network
campaigns offers some lessons for their future. Howard Deans
campaign for the Democratic nomination for the US presidency is
the most prominent example to date of putting the techniques of
network campaigning to effect. Certainly the Dean network showed
many of the characteristics outlined above, and its influence both on
the tone and the issues that have defined the present contest, and on
the evolution of future campaign strategies, should not be
underestimated. That Deans networks were enough to catapult him
from obscurity to front-runner but not, ultimately, to deliver the final
prize perhaps also points to the next challenge for network politics:
navigating the delicate point where the new grassroots networks
come face to face with the old hierarchies of power, influence and
communication.
So the final question for the future is: what would happen if a
network campaign did successfully propel a government into office?
And when it got there, what would it mean for current models of
governance? For now, these remain what if questions. Yet the need to
shift power and decision-making away from the centre, and the
216

Demos

Networks collection

4/1/04

4:35 PM

Page 217

The rise of network campaigning

broader task of renewing the legitimacy of parties, government and

politics itself, are now an established part of the contemporary
debate. As UK Cabinet Office Minister, Douglas Alexander MP has
argued: Our challenge now is to build on the foundations of the
distributive democracy we have begun to fashion.6 The campaigns,
like Jubilee 2000, that have used networks so effectively in recent years
might just hold the key.
Paul Miller is a researcher at Demos.
Notes
1

2
3
4
5
6

The UK Jubilee 2000 Coalition closed its doors in December 2000. It has been
succeeded in the UK by the Jubilee Debt Campaign and Jubilee Research, both
of which continue to advocate the cancellation of unpayable debts of poor
countries. Examples in this essay are with thanks to Nick Buxton, formerly of
the UK Jubilee 2000 Coalition, and Jess Worth of People and Planet.
D Ronfeldt, Tribes, Institutions, Markets, Networks: a framework about societal
evolution (California: Rand Institute, 1996).
Quoted in PR Week, 16 April 1999.
For more information, see H Rheingold, Smart Mobs: the next social revolution
(Cambridge, MA: Perseus, 2002).
M Castells, The Internet Galaxy: reflections on the internet, business and society
(Oxford: Oxford University Press, 2002).
In foreword to D Rushkoff, Open Source Democracy: how online communication
is changing offline politics (London: Demos, 2003).

Demos

217

Networks collection

4/1/04

4:35 PM

Page 218

Networks collection

4/1/04

4:35 PM

Page 219

Network Logic 17

Power does not

reside in
institutions, not
even the state or
large corporations.
It is located in the
networks that
structure society . . .
Afterword: why
networks matter
Manuel Castells

Networks collection

4/1/04

4:35 PM

Page 220

Networks collection

4/1/04

4:35 PM

Page 221

17. Afterword
why networks matter
Manuel Castells

According to the insightful essays in this volume, networks appear to

be the organising form of life, including social life. If this is the case,
why is it only in recent years that networks have come to the forefront
of social practice? Why the network society now?
The answer is in the development of microelectronics and
software-based communication technologies. Of course, we know
that technology does not determine society. But we also know that
without specific technologies some social structures could not
develop. For example, the industrial society could not have emerged
without electricity and the electrical engine.
Thus only under the conditions of the recent wave of information
and communication technologies could networks (an old form of
social organisation) address their fundamental shortcoming: their
inability to manage coordination functions beyond a certain
threshold of size, complexity and velocity. Only under the electronicsbased technological paradigm can networks reconfigure themselves in
real time, on a globallocal scale, and permeate all domains of social
life. This is why we live in a network society, not in an information
society or a knowledge society.
Indeed, if by information or knowledge society we mean a society
in which information is an essential source of wealth and power, I
doubt there is any society in history that escapes this characterisation.
If by information society we mean a society in which the
Demos

221

Networks collection

4/1/04

4:35 PM

Page 222

Network logic

technological paradigm is the dominant medium for social

organisation, this is our society. But to characterise society only by its
technological dimension is reductionist and implicitly deterministic.
The proper identification of our society is in terms of its specific
social structure: networks powered by microelectronics and softwarebased information and communication technologies. If this is the
case, as a growing body of research seems to indicate, a number of
consequences follow.
First, the network society expands on a global scale. This is the
structural basis for globalisation. Networks know no boundaries. If
there is a material communication infrastructure (such as the internet
or an air transportation network) societies become interconnected
throughout the world on the basis of multidimensional networks.
Furthermore, the networking logic explains the features of the
process of globalisation. This is because, as Geoff Mulgan explained
to us in his pioneering work more than a decade ago, networks
communicate and incommunicate at the same time. So while the
network society is organised on a global scale, not all territories, or
people, are connected in this network society. But all countries are
influenced, shaped and ultimately dominated by the logic, interests
and conflicts of this network society the multidimensional network
of networks structuring peoples life around the planet while also
being shaped and modified by the codes and programmes inscripted
by peoples action.
Second, networked organisations outcompete all other forms of
organisation, particularly the vertical, rigid, command-and-control
bureaucracies. This is how networks expand, for instance, in the
business world. Companies that do not or cannot follow this logic are
outperformed and ultimately phased out by leaner, more flexible
competitors. Yes, we live in a world of mergers and conglomerates,
but the succesful ones are precisely those based on networks and
flexible partnerships. The image of networked firms is too often
associated with small and medium-sized businesses. In reality, it is a
much more complex world of large firms internally networked,
cooperating with networks of small and medium firms, and
222

Demos

Networks collection

4/1/04

4:35 PM

Page 223

Afterword

integrated in broader, strategic alliances in which cooperation and

competition alternate, often with the same actors, in different times
and spaces.
Third, the networking of political institutions is the de facto
response to the management crisis suffered by nation states in a
supranational world. The call for global governance has been
answered to some extent in the practice of governments and social
actors. Not under the utopian form of a world government led by
retired statesmen and noble intellectuals, but in the daily practice of
joint decision-making in a network state made of nation-states,
supranational associations, international institutions, local and
regional governments, and quasi-public non-governmental organisations.
Fourth, civil society is reconstructed at the local and global level
through networks of activists, often organised and debated over the
internet, which form and reconfigure depending on issues, on events,
on moods, on cultures. The network society does not cease to be a
contradictory structure, and a conflictual practice, as all societies in
history have been.
Fifth, sociability is transformed in the new historical context, with
networked individualism emerging as the synthesis between the
affirmation of an individual-centred culture, and the need and desire
for sharing and co-experiencing. Virtual communities and smart
mobs, hybrid networks of space and photons (as in the ME++ culture
conceptualised by William Mitchell) are redefining space and time
not in the terms of the science fiction writers but as the appropriation
of technology by people for their own uses and values.
Sixth, the whole range of social practices, both global and local,
communicates in the media space. The media, in the broadest sense,
are the public space of our time: the space in which, and by which,
societies exist as social forms of shared experience. Not just the
mainstream media, but all media, the hypertext formed by television,
radio, the print press, movies, music, videogames, art and the
internet as the communication medium of all the communication
media. The elasticity and interactivity of the media hypertext, its
Demos

223

Networks collection

4/1/04

4:35 PM

Page 224

Network logic

recombinant power, provide the media space with infinite capacity to

integrate and to exclude, thus defining the boundaries of society in
the material world of our minds and representations.
Finally, in this network society, power continues to be the
fundamental structuring force of its shape and direction. But power
does not reside in institutions, not even in the state or in large
corporations. It is located in the networks that structure society. Or,
rather, in what I propose to call the switchers; that is, the
mechanisms connecting or disconnecting networks on the basis of
certain programmes or strategies. For instance, in the connection
between the media and the political system. Or between the financial
markets and the regulatory agencies. Or between the criminal
economy and the same financial markets. Or between religious
apparatuses and government leaders. Or any multi-pronged
combination of any of the previous combinations.
Power elite? Precisely not. Elites change with each reconfiguration
of networks. Power is exercised by specific configurations of these
networks that express dominant interests and values, but whose
actors and forms can change. This is why to challenge a certain group
in government or in business does not alter the structural logic of
domination. This is why to counter networks of power and their
connections, alternative networks need to be introduced: networks
that disrupt certain connections and establish new ones, such as
disconnecting political institutions from the business-dominated
media and re-anchoring them in civil society through horizontal
communication networks. Networks versus networks. Domination
can hardly be exercised against self-configurating networks. And
democratic control is lost in a global network of multidimensional
domination hidden in the complexity of switches.
Networks matter because they are the underlying structure of our
lives. And without understanding their logic we cannot change their
programmes to harness their flexibility to our hopes, instead of
relentlessly adapting ourselves to the instructions received from their
unseen codes. Networks are the Matrix.

224

Demos

Networks collection

4/1/04

4:35 PM

Page 225

Afterword

Manuel Castells is Wallis Annenberg Chair of Communication

Technology and Society, Annenberg School of Communication,
University of Southern California, Los Angeles.

Demos

225

Networks collection

4/1/04

4:35 PM

Page 226

Copyright

DEMOS Licence to Publish

THE WORK (AS DEFINED BELOW) IS PROVIDED UNDER THE TERMS OF THIS LICENCE (LICENCE).THE
WORK IS PROTECTED BY COPYRIGHT AND/OR OTHER APPLICABLE LAW. ANY USE OF THE WORK OTHER
THAN AS AUTHORIZED UNDER THIS LICENCE IS PROHIBITED. BY EXERCISING ANY RIGHTS TO THE WORK
PROVIDED HERE,YOU ACCEPT AND AGREE TO BE BOUND BY THE TERMS OF THIS LICENCE. DEMOS
GRANTS YOU THE RIGHTS CONTAINED HERE IN CONSIDERATION OF YOUR ACCEPTANCE OF SUCH TERMS
AND CONDITIONS.
1. Definitions
a Collective Work means a work, such as a periodical issue, anthology or encyclopedia, in which
the Work in its entirety in unmodified form, along with a number of other contributions,
constituting separate and independent works in themselves, are assembled into a collective
whole. A work that constitutes a Collective Work will not be considered a Derivative Work (as
defined below) for the purposes of this Licence.
b Derivative Work means a work based upon the Work or upon the Work and other pre-existing
works, such as a musical arrangement, dramatization, fictionalization, motion picture version,
sound recording, art reproduction, abridgment, condensation, or any other form in which the
Work may be recast, transformed, or adapted, except that a work that constitutes a Collective
Work or a translation from English into another language will not be considered a Derivative
Work for the purpose of this Licence.
c Licensor means the individual or entity that offers the Work under the terms of this Licence.
d Original Author means the individual or entity who created the Work.
e Work means the copyrightable work of authorship offered under the terms of this Licence.
f You means an individual or entity exercising rights under this Licence who has not previously
violated the terms of this Licence with respect to the Work, or who has received express permission
from DEMOS to exercise rights under this Licence despite a previous violation.
2. Fair Use Rights. Nothing in this licence is intended to reduce, limit, or restrict any rights arising from
fair use, first sale or other limitations on the exclusive rights of the copyright owner under copyright
law or other applicable laws.
3. Licence Grant. Subject to the terms and conditions of this Licence, Licensor hereby grants You a
worldwide, royalty-free, non-exclusive, perpetual (for the duration of the applicable copyright) licence
to exercise the rights in the Work as stated below:
a to reproduce the Work, to incorporate the Work into one or more Collective Works, and to
reproduce the Work as incorporated in the Collective Works;
b to distribute copies or phonorecords of, display publicly, perform publicly, and perform publicly
by means of a digital audio transmission the Work including as incorporated in Collective Works;
The above rights may be exercised in all media and formats whether now known or hereafter
devised.The above rights include the right to make such modifications as are technically necessary to
exercise the rights in other media and formats. All rights not expressly granted by Licensor are hereby
reserved.
4. Restrictions. The licence granted in Section 3 above is expressly made subject to and limited by the
following restrictions:
a You may distribute, publicly display, publicly perform, or publicly digitally perform the Work only
under the terms of this Licence, and You must include a copy of, or the Uniform Resource
Identifier for, this Licence with every copy or phonorecord of the Work You distribute, publicly
display, publicly perform, or publicly digitally perform.You may not offer or impose any terms on
the Work that alter or restrict the terms of this Licence or the recipients exercise of the rights
granted hereunder.You may not sublicence the Work.You must keep intact all notices that refer
to this Licence and to the disclaimer of warranties.You may not distribute, publicly display,
publicly perform, or publicly digitally perform the Work with any technological measures that
control access or use of the Work in a manner inconsistent with the terms of this Licence
Agreement.The above applies to the Work as incorporated in a Collective Work, but this does not
require the Collective Work apart from the Work itself to be made subject to the terms of this
Licence. If You create a Collective Work, upon notice from any Licencor You must, to the extent
practicable, remove from the Collective Work any reference to such Licensor or the Original
Author, as requested.
b You may not exercise any of the rights granted to You in Section 3 above in any manner that is
primarily intended for or directed toward commercial advantage or private monetary

226

Demos

Networks collection

4/1/04

4:35 PM

Page 227

Copyright

5.

6.

7.

8.

compensation.The exchange of the Work for other copyrighted works by means of digital filesharing or otherwise shall not be considered to be intended for or directed toward commercial
advantage or private monetary compensation, provided there is no payment of any monetary
compensation in connection with the exchange of copyrighted works.
c If you distribute, publicly display, publicly perform, or publicly digitally perform the Work or any
Collective Works,You must keep intact all copyright notices for the Work and give the Original
Author credit reasonable to the medium or means You are utilizing by conveying the name (or
pseudonym if applicable) of the Original Author if supplied; the title of the Work if supplied. Such
credit may be implemented in any reasonable manner; provided, however, that in the case of a
Collective Work, at a minimum such credit will appear where any other comparable authorship
credit appears and in a manner at least as prominent as such other comparable authorship credit.
Representations, Warranties and Disclaimer
a By offering the Work for public release under this Licence, Licensor represents and warrants that,
to the best of Licensors knowledge after reasonable inquiry:
i Licensor has secured all rights in the Work necessary to grant the licence rights hereunder
and to permit the lawful exercise of the rights granted hereunder without You having any
obligation to pay any royalties, compulsory licence fees, residuals or any other payments;
ii The Work does not infringe the copyright, trademark, publicity rights, common law rights or
any other right of any third party or constitute defamation, invasion of privacy or other
tortious injury to any third party.
b EXCEPT AS EXPRESSLY STATED IN THIS LICENCE OR OTHERWISE AGREED IN WRITING OR
REQUIRED BY APPLICABLE LAW, THE WORK IS LICENCED ON AN AS IS BASIS, WITHOUT
WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT LIMITATION, ANY
WARRANTIES REGARDING THE CONTENTS OR ACCURACY OF THE WORK.
Limitation on Liability. EXCEPT TO THE EXTENT REQUIRED BY APPLICABLE LAW, AND EXCEPT FOR
DAMAGES ARISING FROM LIABILITY TO A THIRD PARTY RESULTING FROM BREACH OF THE
WARRANTIES IN SECTION 5, IN NO EVENT WILL LICENSOR BE LIABLE TO YOU ON ANY LEGAL THEORY
FOR ANY SPECIAL, INCIDENTAL, CONSEQUENTIAL, PUNITIVE OR EXEMPLARY DAMAGES ARISING OUT
OF THIS LICENCE OR THE USE OF THE WORK, EVEN IF LICENSOR HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.
Termination
a This Licence and the rights granted hereunder will terminate automatically upon any breach by
You of the terms of this Licence. Individuals or entities who have received Collective Works from
You under this Licence, however, will not have their licences terminated provided such individuals
or entities remain in full compliance with those licences. Sections 1, 2, 5, 6, 7, and 8 will survive any
termination of this Licence.
b Subject to the above terms and conditions, the licence granted here is perpetual (for the duration
of the applicable copyright in the Work). Notwithstanding the above, Licensor reserves the right
to release the Work under different licence terms or to stop distributing the Work at any time;
provided, however that any such election will not serve to withdraw this Licence (or any other
licence that has been, or is required to be, granted under the terms of this Licence), and this
Licence will continue in full force and effect unless terminated as stated above.
Miscellaneous
a Each time You distribute or publicly digitally perform the Work or a Collective Work, DEMOS offers
to the recipient a licence to the Work on the same terms and conditions as the licence granted to
You under this Licence.
b If any provision of this Licence is invalid or unenforceable under applicable law, it shall not affect
the validity or enforceability of the remainder of the terms of this Licence, and without further
action by the parties to this agreement, such provision shall be reformed to the minimum extent
necessary to make such provision valid and enforceable.
c No term or provision of this Licence shall be deemed waived and no breach consented to unless
such waiver or consent shall be in writing and signed by the party to be charged with such
waiver or consent.
d This Licence constitutes the entire agreement between the parties with respect to the Work
licensed here.There are no understandings, agreements or representations with respect to the
Work not specified here. Licensor shall not be bound by any additional provisions that may
appear in any communication from You.This Licence may not be modified without the mutual
written agreement of DEMOS and You.

Demos

227

Networks collection

4/1/04

4:35 PM

Page 228