2017 - Rhodes - Network Governance and The Differentiated Polity Selected Essays, Volume I

NETWORK GOVERNANCE
A N D T H E D I F F E R E N T I A T E D P O L I TY
Network Governance
and the Differentiated
Polity
Selected Essays, Volume I
R. A. W. RHODES
1
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© R. A. W. Rhodes 2017
except where indicated in the first footnote of each chapter
The moral rights of the author have been asserted
First Edition published in 2017
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2017932298
ISBN 978–0–19–878610–8
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
OUP CORRECTED PROOF – FINAL, 26/6/2017, SPi
Preface
This collection of essays is in two volumes. Volume I is a retrospective. It

collects in one place for the first time the main articles I wrote on policy
networks and governance between 1990 and 2005. The introductory section
provides a short biography of my intellectual journey. Part I focuses on
policy networks. Part II focuses on governance. The conclusion provides
the critical commentary, both replying to my critics and reflecting on theor-
etical developments since publication. With the exceptions of Chapters 6, 7,
and 10, none of these articles and chapters appeared in my other books.
Chapter 5 has not been published before in English and Chapter 12 has not
been published before. The volume complements my other publications on
networks and governance. In effect, it updates my Understanding Govern-
ance, which was published twenty years ago. Finally, where necessary, I have
written an afterword to a chapter setting out the context in which it was
written, and identifying what has changed empirically. I have reserved my
discussion of both the continuing relevance of my argument and the perspi-
cacity of my critics to Chapter 12.
Volume II is prospective in that it looks forward and explores the
‘interpretive turn’ and its implications for the craft of political science,
especially public administration. It draws together articles from 2005
onwards on the theme of ‘the interpretive turn’ in political science. In Part
I, I provide a summary statement of the interpretive approach. It provides
the context for what follows. Part II develops the theme of blurring genres
and discusses a variety of research methods common in the humanities,
including: ethnographic fieldwork, life history, and focus groups. Part III
shows how the genres of thought and presentation found in the humanities
can be used in political science. It presents four examples of such blurring
‘at work’ with studies of: applied anthropology and civil service reform;
women’s studies and government departments; and storytelling and local
knowledge; and area studies and comparing Westminster governments. The
book concludes with a summary of what is edifying about an interpretive
approach, and why this approach matters. I revisit some of the more common
criticisms before indulging in plausible conjectures about the future of
interpretivism.
Volume II differs from my work with Mark Bevir in two significant ways.
First, it is not a book about interpretive theory. Briefly, I summarize the
theoretical case for interpretivism but my main concern is to make the case
for the approach by showing how it refreshes old topics and opens new
empirical topics. I seek new and interesting ways to explore governance,
vi Preface
high politics, public policies, and the study of public administration in general.
So, my emphasis falls on methods, and providing several examples of the
approach ‘at work’.
Second, with the exception of Volume II, Chapter 2, none of the articles was
co-written with Mark Bevir, although I acknowledge freely his influence
throughout Volume II. These essays complement but do not duplicate our
joint publications. None of these articles is in any of my single or co-authored
books, and Chapters 1 and 12 have not been published before. Volume II draws
together in one place for the first time my recent work applying interpretivism
to political science, especially public administration.
As the articles and chapters in Volume II are recent, I have not written
an afterword updating each chapter. Rather, I gather my reflections on the
chapters, with replies to my critics, in Chapter 12.
For Volume I, I have not changed the arguments in any chapter. However,
I have corrected factual mistakes and updated, standardized, and consolidated
the references. I am afraid I could not resist tinkering with my prose. Over the
years, I have acquired a growing aversion to the ‘hanging this’ and to long
sentences. I have pandered to both aversions. Inevitably with articles and
chapters written over 30 years, there is some duplication and overlap. At the
time, I could not assume that readers were familiar with earlier work. I have
eliminated most of the ‘catch-up’ passages in my previous work. For Volume
II, because all the chapters are linked by the twin themes of ‘blurring genres’
and ‘the interpretive turn’, I gave myself license to revise thoroughly and
rewrite to ensure internal and thematic consistency.
When writing, I do so to music, mainly folk, jazz, and rock. It is the ever-
present backcloth to my working life. Occasionally, I succumb to the conceit
that in another life I was in a rock band, playing air guitar of course. The
articles are the singles. The books are the CDs. These two volumes and edited
collections are compilation CDs. The lecture tours are the gigs. The hotels are
the motels of rock’s road songs. Song titles and phrases seep into my con-
sciousness and onto the printed page. You will find echoes of Bob Dylan,
Jethro Tull, Prince, and many more throughout these pages. I enjoy listening
to them, and now they are part of the backdrop to your reading.
Acknowledgements
Volume I, Chapter 4 was written with Ian Bache and Stephen George. I thank
them for their generosity in allowing me to include the paper in this collection.
Many colleagues have given me the benefit of their comments and advice
over the years and the following list is an inadequate way of acknowledging my
debts and saying thank you.
Claire Annesley (University of Sussex)
Chris Ansell (University of California, Berkeley, USA)
Ian Bache (University of Sheffield)
Mark Bevir (University of California, Berkeley, USA)
Karen Boll (Copenhagen Business School, Denmark)
John Boswell (University of Southampton)
George Boyne (University of Cardiff)
Judith Brett (LaTrobe University, Melbourne, Australia)
Dominic Byatt (Oxford University Press)
Neil Carter (University of York)
Louise Chappell (University of New South Wales, Sydney, Australia)
Jack Corbett (University of Southampton)
Charlotte Sausman (née Dargie) (University of Cambridge)
Carsten Daugbjerg (Australian National University, Canberra, Australia)
Patrick Dunleavy (LSE)
The late Andrew Dunsire (University of York)
Jenny Fleming (University of Southampton)
Francesca Gains (University of Manchester)
Andrew Gamble (Emeritus, University of Cambridge)
Stephen George (University of Sheffield)
Michael Goldsmith (formerly University of Salford)
Bob Goodin (Australian National University, Canberra, Australia)
Brian Hardy (formerly Nuffield Institute for Health, University of Leeds)
Richard Harrington (Manchester Statistical Society)
Carolyn Hendriks (Australian National University, Canberra, Australia)
Susan Hodgett (University of Ulster)
Liesbet Hooghe (University of North Carolina, Chapel Hill, USA)
Ingi Iusmen (University of Southampton)
Lotte Jensen (Copenhagen Business School, Denmark)
Bob Jessop (Lancaster University)
The late George Jones (LSE)
Josie Kelly (Aston Business School)
The late Adrian Leftwich (University of York)
viii Acknowledgements
David Levi-Faur (Hebrew University of Jerusalem)

Rodney Lowe (University of Bristol)
Fiona MacKay (University of Edinburgh)
David Marsh (University of Canberra, Australia)
Janice McMillan (Edinburgh Napier University)
Mick Moran (Manchester Business School)
Mirko Noordegraaf (Universiteit Utrecht, The Netherlands)
Johan Olsen (ARENA Centre for European Studies, Oslo, Norway)
The late Nelson Polsby (University of California, Berkeley, USA)
Alison Proctor (Australian National University, Canberra, Australia)
David Richards (University of Liverpool)
Ella Ritchie (University of Newcastle)
The late Jim Sharpe (Nuffield College, Oxford)
Martin Smith (University of York)
John Stewart (Formerly INLOGOV, University of Birmingham)
Richard J. Stillman II (University of Colorado at Denver)
Gerry Stoker (University of Canberra, Australia)
Paul ‘t Hart (Universiteit Utrecht, The Netherlands)
Anne Tiernan (Griffith University, Brisbane, Australia)
Nick Turnbull (University of Manchester)
James Walter (University of Monash, Melbourne, Australia)
John Wanna (Australian National University, Canberra, Australia)
Georgina Waylen (University of Manchester)
Patrick Weller (Griffith University, Brisbane, Australia)
David Wilson (De Montfort University)
The late Vincent Wright (Nuffield College, Oxford)
Tamyko Ysa (ESADE, Universitat Ramon Llull, Barcelona, Spain)
Many institutions have helped over the years—too many to thank—but I must
single out the Institute of Public Administration Australia (IPAA) and its state
divisions for repeat invitations to speak, hospitality, questions, and unfailing
good humour. I hope they enjoyed my visits as much as I did.
I am grateful to the following publishers for their permission to reprint in
whole or in part the following articles and book chapters.
Oxford University Press for: ‘Policy Network Analysis’. In M. Moran,
M. Rein and R. E. Goodin (eds), The Oxford Handbook of Public Policy.
Oxford: Oxford University Press, 2006: 423–45; ‘Policy Networks and Policy
Making in the European Union: A Critical Appraisal’. In L. Hooghe (ed.),
Cohesion Policy and European Integration. Oxford: Clarendon Press, 1996:
367–87; ‘What is Governance and Why Does It Matter?’ In J. E. S. Hayward
and Anand Menon (eds), Governing Europe. Oxford: Oxford University
Press, 2003: 61–73; and ‘Waves of Governance’. In David Levi-Faur (ed.),
The Oxford Handbook of Governance. Oxford: Oxford University Press,
2012, 33–48.
Acknowledgements ix
Palgrave Macmillan for: ‘From Prime Ministerial Power to Core Executive’.
In R. A. W. Rhodes and P. Dunleavy (eds), Prime Minister, Cabinet and
Core Executive. London: Macmillan, 1995: 11–37.
Sage for: ‘Policy Networks: A British Perspective’, Journal of Theoretical
Politics, 2, 1990: 292–316; ‘Bureaucracy, Contracts and Networks: The
Unholy Trinity and the Police’, Australian and New Zealand Journal of
Criminology, 38, 2005: 192–205; ‘The New Governance: Governing without
Government’, Political Studies, 44, 1996: 652–67; and ‘Understanding
Governance: Ten Years On’, Organization Studies, 28 (8), 2007: 1243–64.
Ant. N. Sakkoulas, Athens, Greece for: ‘Analysing Networks: From
Typologies of Institutions to Narratives of Beliefs’, Science and Society,
No. 10, Spring 2003: 21–56.
Taylor & Francis/Routledge for: ‘Putting the People Back into Networks’,
Australian Journal of Political Science, 37 (3), 2002: 399–415.
John Wiley for: ‘From Marketization to Diplomacy: It’s the Mix that
Matters’, Australian Journal of Public Administration, 56, 1997: 40–53;
‘The Hollowing Out of the State’, Political Quarterly, 65, 1994: 138–51;
‘Thinking On: I Was So Much Older Then’, Public Administration, 89,
2011: 196–212.
There were too many conferences and workshops at which colleagues com-
mented on drafts of these several papers, and it is not feasible to list them all.
So, this general thank you must suffice. I should also thank the many an-
onymous referees. I obeyed the ‘rules of the game’, even when convinced the
revised version was no improvement; for example, there is no advantage in
using the third person over the first person. It proliferates passive verbs in
pursuit of a spurious detachment.
I will not try to describe the stultifying claustrophobia of Bradford in the
1950s. In 1958, aged 14, I preferred Lonnie Donegan to Elvis Presley if only
because the latter was so exotic he seemed to be from another planet;
untouchable. At least Lonnie was one of us. Everybody knew someone in a
skiffle group. My mother, Irene Rhodes (née Clegg), loathed the confines of
the Rhodes’ extended family, of chapel, and of the narrow horizons of a textile
town. She insisted I think beyond the confines of provincial Yorkshire,
knowing the journey would take me away from her. It was a precious and
still valued gift. These two volumes are dedicated to her memory.
Contents
List of Figures xiii

List of Tables xv
INTRODUCTION
1. What a Long Strange Trip It’s Been 3
PART I. POLICY NETWORKS

2. Policy Networks in Britain: The Early Years 15
3. Policy Networks: The Historical Moment 37
4. Policy Networks and Policy-making in the European Union 57
5. How to Manage Your Policy Network 74
6. Putting the People Back into Networks 87
7. Analysing Networks as Narratives of Beliefs and Practices 103
PART II. GOVERNANCE

8. The Hollowing Out of the State 119
9. From Prime Ministerial Power to Core Executive 137
10. The New Governance: Governing without Government 158
11. It’s the Mix That Matters: From Marketization to Diplomacy 173
CONCLUSION
12. What Is New about ‘Network Governance’ and Why Does
It Matter? 199
Appendix: Bibliography on the Anglo-Governance Debate 225

References 229
Author Index 265
Subject Index 271
List of Figures
2.1 Approaches to networks 16

5.1 When to build a network: ten lessons 77
5.2 How to manage your network: ten commandments 82
5.3 Strategic storyteller 86
List of Tables
3.1 Types of policy networks 39

3.2 Approaches to network management 46
5.1 Characteristics of networks 75
5.2 The sour laws of network governance 85
11.1 The characteristics of governance 176
Introduction
1
What a Long Strange Trip It’s Been
I WAS SO MUCH OLDER THEN
It is intimidating to look back over a 45-year career as a political scientist. Has

it been that long? Can I remember ‘the gangling youth of the prominent
Adam’s apple variety’—as one of my referees expressed it back then? It is
tempting to claim I had a rationale to cover an unfolding research agenda but
I find myself reading the work of someone I struggle to remember. Any
overarching rationale would be a patina. Often I was lucky in the people
I met. The journey had many twists and turns. Looking back imposes a logic
that was not clear at the time. As Bob Dylan’s evocative line from his song ‘My
Back Pages’ suggests, I did seem older then but the certainties of a young
academic did not last; old beliefs gave way to new ideas. Life myths were
rewritten. And I told myself, the harder I worked, the luckier I got.1
In the Beginning, 1970–1976
The study of public administration in the 1970s was shaking off the old order. Its
grand old men were William Robson (1895–1980), Norman Chester (1907–86)
and W. J. M. (Bill) Mackenzie (1909–96). All were on the cusp of retirement. For
me, they represented traditional public administration, which was essentially
institutional and concerned to analyse the history, structure, functions, powers,
and relationships of government organizations (see Mackenzie 1975; Rhodes
1979a: ch. 5; Robson 1975). Robson represented that blend of institutional
description and Westminster reformism so typical of the British school. ‘His
great ability was to assemble a huge mass of data, to analyse order out of the
complexity, and to argue a coherent case for change.’ He was ‘one of the
Olympian Fabians, worthy company to the Webbs’ (Jones 1986: 12). Norman
Chester’s best books were the official history of the nationalized industries (1975)
1
Sections of this chapter appeared in R. A. W. Rhodes (2011c) ‘Thinking On: I Was So Much
Older Then’, Public Administration, 89 (1): 196–212. Reprinted with permission of John Wiley
and Sons.
4 Network Governance and Differentiated Polity
and a history of the English administrative system between 1780 and 1870
(1981). Bill Mackenzie (1975) was admired for his lucid, nuanced essays on
both British government and the study of public administration. All were
prominent in my undergraduate education. Robson’s Nationalised Industries
and Public Ownership (1962) was a birthday present—yes, I was delighted, and
still have it.
Like many a young scholar, my horizons were confined by my academic
training and employment opportunities. I had an undergraduate degree in
business and administration from Bradford Business School and a yet-to-be-
completed research degree from Oxford. I applied for jobs at Trinity College,
Dublin, under Basil Chubb, and Aberdeen, under Frank Bealey, but both in
their wisdom decided they could survive without my talents. John Stewart and
Richard Chapman at the Institute of Local Government Studies (INLOGOV),
University of Birmingham, were more discerning! So, I had ten years of
teaching and research on British local government. To put no finer point on
it, I floundered. I never intended to be a consultant for local government or
train local government officers. I don’t think I knew what I wanted to do. I had
no individual voice, just boundless, ill-directed enthusiasm. So, I wrote on the
reform of English local government, Anthony Trollope and the nineteenth-
century civil service, developments in the study of public administration, and
the impact of membership of the (then) European Economic Community
(EEC) on local government. From the vantage point of 2016, I can think of no
reason to be interested in competition for public works contracts, but I read
and wrote about these EEC regulations, and kept an interest in EU matters for
many years afterwards (Rhodes 1973, 1986c; Rhodes, Bache, and George 1996;
and Chapter 5, this volume).
INLOGOV expected applied work relevant to its local government audi-
ence, and micro-specialization was ever the lot of the novitiate academic, more
so today than then. Still, I had to prove myself. Some of my scribbling might
have had passing value, but are best classed as juvenilia. I made no lasting
contribution until I was commissioned by the Committee of Inquiry into Local
Government Finance (Layfield) to review the academic literature on the
relationship between central departments and local authorities (Rhodes
1976). This work led me to submit evidence to the (then) Social Science
Research Council (SSRC) Panel on Research into Local Government
(Rhodes 1977) and my appointment to the SSRC Panel on Central–Local
Government Relationships. For the first time, I had an intellectual agenda.
A Professional Political Scientist at Last, 1976–1988
During the 1970s, change was also afoot in the wider world. The young lions
were at public administration’s door. I experienced the change first-hand at
What a Long Strange Trip It’s Been 5
the Public Administration Committee’s (PAC) Conference on 13–15 September

1971, at the University of York. It was my first academic conference and I was
excited because it had such luminaries as Ron Brown (1971) extolling the
virtues of organization theory, John Stewart (1971) on public policy-making,
Lewis Gunn (1971) on public management, and Peter Self (1971), who
exorcised the evil spirits of economic efficiency. The conference explored
new ways of studying public administration. I was a spectator of the new
generation—the successors to Robson, Chester, and Mackenzie. I also saw the
future in the guise of the theory and methods of American social science. In
John Stewart, I had a mentor whose commitment to ideas, to INLOGOV, and
to local government was as admirable as it was infectious, even if I did not
share his enthusiasm for corporate management (Rhodes 1992b).
As a postgraduate, I read American social science avidly. I was an admirer
of the theoretically informed case studies of, for example, Michel Crozier
(1964) and Philip Selznick (1966). I saw this work as the intellectual challenge
to traditional public administration. Policy studies and organization theory
were the way forward (see also Hood 1990). The temper of the times encour-
aged me to apply the theory and methods of American social science in case
studies of British local government in its dealing with central government. Of
the distinguished speakers at the PAC conference, all are now retired and
several are dead. The generations pass. But, for a time, I was heir to their ideas
and enthusiasms; a modernist-empiricist in all but name. In other words,
I treated institutions such as central departments, local governments, and
policy networks as discrete, atomized objects to be compared, measured, and
classified. I sought to explain these institutions by appealing to ahistorical
mechanisms such as functional differentiation (see Bevir 2001).
In January 1978, I was invited to join the SSRC Panel on Central–Local
Government Relationships. The Panel commissioned me to write a review of
the existing literature on the subject and develop an analytical framework. My
work was completed in May 1978 and an article-length version was published
as an appendix to the Panel’s own report in January 1979 (Rhodes 1979b). The
full-length version of my report to the Panel was published as Rhodes 1981.
The work I did for the SSRC was modernist-empiricist: the subtitle of one
report was ‘the search for positive theory’ and gives the game away (Rhodes
1978a). The theory was ‘interorganizational analysis’ and my main influences
were Kenneth Benson (1975), Michel Crozier and Jean-Claude Thoenig
(1976), and James Thompson (1967). To this day, exchange theory lies at
the heart of policy network theory. Thus, ‘an organisation has power, relative
to an element of its task environment, to the extent that the organisation has
the capacity to satisfy needs of that element and to the extent that the organi-
sation monopolises that capacity’ (Thompson 1967: 30–1). I elaborated this
idea arguing that any organization is dependent on other organizations
for resources. To achieve their goals, the organizations have to exchange
resources. The organization’s dominant coalition employs strategies within

known rules of the game to regulate this exchange relationship (paraphrased
from Rhodes 1979b; 1981: 98–9).
So, I argued local authorities were embedded in sets of relationships and we
should analyse the patterns of interdependence, not just the links with central
departments. Following the lead of Heclo and Wildavsky, I suggested that
these networks were structured by policy area or function (Rhodes 1978b;
1981: ch. 5). So, the interorganizational links between central departments and
local authorities took the form of ‘policy communities’ of:
personal relationships between major political and administrative actors—some-
times in conflict, often in agreement, but always in touch and operating within a
shared framework. Community is the cohesive and orienting bond underlying
any particular issue (Heclo and Wildavsky 1974: xv).
I did not know it at the time but here were the roots of ten years’ work on
policy networks (see Chapter 3, this volume).
Policy Networks
As I began to explore policy networks, Margaret Thatcher was intent on

transforming the public sector about which I was writing. The age of man-
agerialism in its twin guises of performance measurement and marketization
was upon us. Mainstream public administration embraced the new public
management. There were a sceptical few. Christopher Hood (1990) argued
that the rise of managerialism meant the field had lost coherence. It had
fragmented into subdisciplines, still including, but not limited to, organiza-
tional studies and policy analysis. The challenge was to find a framework and a
language to compare and contrast these several paradigms. I argued for an
explicit multi-theoretic approach, methodological pluralism, and, above all,
the need to set our own research agendas (Rhodes 1991a). No matter how
individuals responded to the changes in the public sector, few would deny
managerialism was pre-eminent (see also Hood 1991; Pollitt 1993).
I spent the 1980s in the Department of Government at the University of
Essex. It set out to emulate American political science. It became, and remained,
among the best political science departments in the UK. Initially, I did not
prosper. The Department of Government rigorously pursued the highest stand-
ards of professional excellence in which research was the clear priority. Running
an undergraduate degree may be necessary, but it was a chore. The thrill lay in
your next grant, article or book, and building an international reputation. It was
a lesson to learn quickly if you wanted promotion. I learnt, but perhaps not as
quickly as I should. My pet project was a new undergraduate degree in public
administration, which grew from zero to 30 admissions a year. Pet projects can
slow you down. I did not publish enough. I was not promoted. So, I resigned as
degree director and inflicted two large, 400-pages-plus books on a world that
had done nothing to deserve such punishment.
My fieldwork on the local government peak associations and their linked
specialist, advisory bodies was part of the (now) Economic and Social Research
Council’s (ESRC) Research Programme on Central and Local Government Rela-
tionships. It was published in 1986 as The National World of Local Government.
Subsequently, I won an ESRC personal research grant to draw together the
findings of the 16 major research projects that formed the Research Programme.
It resulted in Beyond Westminster and Whitehall (1988). This book provided a
full-length treatment of policy networks and argued that Britain should be seen as
a differentiated polity.
In 1988 I became Head of Department and had the task of compiling the
department’s submission for the first Research Assessment Exercise (RAE),
now the Research Evaluation Framework (REF). I enjoyed the job—it was easy
because colleagues were not only productive but are among the best in the
country. We got our five stars. So, Essex in the 1980s was a department to
admire. It turned me into a professional political scientist and gave me my first
taste of university management.
From Government to Governance, 1988–1998
After a decade of Thatcherism, the 1990s were an inauspicious time for the
theory and practice of public administration. Managerialism was rife. The
civil service had been the butt of criticism and reform for over a decade. I had
just been appointed to my first chair at the University of York, and I did not
think I had inherited either a healthy department or discipline. I wrote a
couple of pessimistic pieces on the decline of public administration (for
example, Rhodes 1997a: ch. 8). I was not the first (Ridley 1975). I was not
alone among my contemporaries. Dunsire (1995: 34) noted that implementa-
tion theory and contingency theory had died. I set about doing something
to revive my field, and those things were the ‘Local Governance’ and the
‘Whitehall’ research programmes.
A senior Danish colleague once told me he had reached the summit of his
career when he became a full professor. I was surprised. I found becoming a
professor was the start. Now, I could do things that had been closed to a mere
lecturer. For example, I sat on the ESRC’s committee responsible for research
programmes. I argued for both a local government (Rhodes 1991b) and a
central government programme (Rhodes 1993). With Gerry Stoker, I set up
the local governance programme (Rhodes 1999b). I then stepped down from
the committee so I could be director of the central government programme
that became known as the Whitehall Programme.
I had always been told by my elders that researchers could not get access
to central government. Heclo and Wildavsky (1974) showed that claim to
be inaccurate. Of course, came the retort, it was because they were foreign-
ers. British academics could not penetrate the veil of secrecy. I had my
doubts. I suspected we said ‘no’ for the ministers and senior civil servants
instead of asking and letting them say ‘no’ for themselves. I drew a simple
lesson. I would ask. I was organizing the annual PAC conference at
University of York, so I invited the (then) Head of the Home Civil Service,
Sir Robin Butler, to give the Frank Stacey Memorial Lecture in which he
signalled his willingness to encourage research on central government
(Butler 1992). Subsequently, the Cabinet Office and the ESRC signed a
formal accord with the former participating in a joint steering and com-
missioning panel to develop the research programme. So, we had access.
Even more striking, the accord was to conduct ‘curiosity research’. It was
agreed by the ESRC and the Cabinet Office that the Research Programme’s
primary objective was not to provide policy relevant advice. Rather, it
would provide an ‘anthology of change’ in British government. To continue
with the language of the civil servants with whom I worked, the Programme
was ‘holding up a mirror to government’ and ‘learning each other’s
language’. The task was ‘to help one another understand the changes’.
According to Peter Hennessy, Sir Robin was every head teacher’s dream
of the perfect head boy. For me, he was the essential ingredient for getting
the ERSC Whitehall Programme off the ground, making his time and other
people available as necessary.
I make this process seem all sweet light and reasonableness. So it seemed
most of the time. My equanimity would have been disturbed had I seen the
advice given to Sir Robin at the time:
Having read the papers my own advice is that Sir Robin should treat this with a
long spoon. . . . There is a lot of excitement in the academic community at the
moment about ‘public sector organisation theory’, but it is never clear exactly
what it means, except a desire to be academic about essentially practical matters.
. . . it looks as if, in order to develop academic theories, the authors of this
proposal want to put a lot of senior civil servants and Ministers to a good deal of
bother in submitting to interviews, answering questionnaires and being mem-
bers of ‘Advice Workshops’.
. . . behind it seems to lie some jealousy of the skill with which Peter Hennessy
has got into and explained present changes in the Civil Service—there are . . .
some rather snide comments on the Peter Hennessy-style approach, i.e. ‘telling
the story of current events or descriptions of institutional and legal arrange-
ments’, because ‘such approaches are atheoretical’
(dated 26 August 1992; personal correspondence received 26 June 2016).
Even today my heart flutters on reading this assessment. And I was not jealous
of Peter Hennessy. I was a fan who wanted to follow in his footsteps, and to do
so I had to be different. My fortunes hung by a slender thread and I am ever

grateful to Sir Robin for preferring his own counsel.
The Programme’s main aims were: to describe, to explain, and to create a
better understanding of both recent and long-term changes in the nature of
British government; to develop new theoretical perspectives; and to encourage
the use of new research methods in the study of central government. The
Programme comprised 23 projects costing £2.1 million. The first project began
in March 1995. The last project finished in December 1998. At its peak the
Programme employed 49 people (for a short history see Rhodes 2000b).
My rationale for the Programme lay in two ideas; the core executive and
network governance. Instead of asking which positions are important in
British government, prime minister or Cabinet, the core executive idea asks
which functions define the heart of the machine. The core functions of the
British executive are to pull together and to act as final arbiters of conflicts
between different elements of the government machine. This notion directs
our attention to two key questions: ‘Who does what?’ and ‘Who has what
resources?’ (see Chapter 9, this volume).
In his review of administrative theory in Britain, Dunsire (1995: 34) specu-
lated that just as public administration had become public management in the
1980s, it could become governance in the 1990s. I first used the term ‘govern-
ance’ for the launch of the local governance initiative when I wrote a short
piece entitled ‘Beyond Whitehall: Researching Local Governance’ in Social
Sciences (Rhodes 1992a). This work on governance was a logical extension of
my previous work on policy networks. It came out of my reappraisal of Beyond
Westminster and Whitehall (1988), which was necessary after Thatcher’s
reforms. My reappraisal was published as Understanding Governance (1997
and Chapter 9, this volume), which developed over the next few years into
‘the Anglo-Governance School’ (Marinetto 2003). The notion of governance
became ubiquitous, and, as with any idea worth its salt, fuelled critical debates
(for a survey of the critics and a reply see Rhodes 2007b and Chapter 12,
this volume).
Apart from studying British government, a central aim of the Whitehall
Programme was to compare the changes in British government with those in
other member states of the European Union (EU) and other states with a
‘Westminster’ system of government.2 Until now, with Vincent Wright as my
patron, my comparative interests had been limited to writing the chapter on
Britain in edited collections of country studies (Peters, Rhodes, and Wright
2000). Vincent got me invited to various international workshops. Others
were irritated by the brusque Northerner in love with the chip on his shoulder.
Vincent just smiled and steered me in productive directions. The Whitehall
2
It is pedantic and tedious to switch between EEC, EC, and EU depending on the date. I refer
to the EU throughout.
Programme gave me the opportunity to branch out on my own and do

genuine comparative work. It fostered my collaboration with Patrick Weller
(Griffith University, Brisbane).
The initial product of our partnership was a collaborative project structured
around the ideas of the hollowing out of the state and the changing role of the
core executive (Weller, Bakviss, and Rhodes 1997). We covered Australia,
Britain, Canada, Germany, and the Netherlands but we did not write country
studies. Instead, everyone wrote on every country and we focused on the
functions of the core executive: winning and keeping support for government,
collective government, policy advice, resource allocation, coordination, and
reform. If there is a single conclusion it was that we told ‘sad stories of the
death of Kings’ as we identified the manifold shackles on leadership.
We then turned to the changing role of the public service (Rhodes and
Weller, 2001). It was a collaborative project again, although this time the
research was based around country studies. We covered Australia, Britain,
Denmark, France, Germany, The Netherlands, and New Zealand. However,
there was a shared framework and a set of agreed methods. We created a data
set on the characteristics of the administrative elite, covering such topics as
age, sex, education, recruitment, training, career paths, and departure. We
explored a common set of topics on what they did and how their roles were
changing. Finally, and most distinctive, we wrote short biographical portraits
constructed from lengthy interviews with the public servants. We tried to let
them speak for themselves. This work demonstrated that the social science
ideas of hollowing out, the core executive, and network governance have
purchase; they travel and illuminate governance practices in other countries
(see Bevir and Rhodes 2003a and 2003b; Elgie 2011; Eymeri-Douzans et al.
2015; Heilman and Stepan 2016).
CONCLUSIONS
In 1996, I had a downbeat view of the state of my discipline. Hood (1999:

288) noted, I was a pessimist who thought ‘an optimist would describe the
future as bleak. A pessimist would be living and working in America.’ Hood
(2011: 128) demurred and inclined cautiously to a ‘never had it so good’ view
of the state of the discipline. In fact, emigrating to America (or Australia for
that matter) was not the only option. I may have thought the discipline was
in a precarious condition but that did not stop me from trying to do
something about it—hence my involvement with the ESRC. I would date
the good times from the mid-1990s when I wrote my prophecy of doom! In
the 1990s, the ESRC funded major research programmes on local governance,
Whitehall, and devolution.
The discipline has survived, even thrived, because some of its leading
players mastered the ‘trick’ of linking policy and academic relevance. We
may specialize in central–local relationships, public service delivery or other
topics of the day, but we must link such topics to broader agendas in the social
and human sciences. Otherwise we become either mere technicians or loyal
servants of power or, of course, both. I have been fortunate. My field has
benefited from the work of many outstanding scholars throughout Europe
over the past 25 years, including, to name but a few, Christopher Hood, Erik-
Hans Klijn, Christopher Pollitt, Johan P. Olsen, Renate Mayntz, Fritz Scharpf,
Paul ‘t Hart, and Jean-Claude Theonig. Indeed, a significant trend over the
past 25 years is this shift to a European community of scholars known to one
another and engaging with one another’s work.
I get ahead of myself. All journeys have starting points and mine was the
study of policy networks and governance. These topics are the focus of the rest
of this volume. The interpretive leg of the journey is the subject of Rhodes
(2017, Volume II).
Part I
Policy Networks
2
Policy Networks in Britain

The Early Years
This chapter identifies and discusses the distinctive British contribution to the
study of policy networks up to 1990 before reviewing the problems of, and
possible developments in, the concept and its application. I excuse the paro-
chialism of the chapter because it will draw attention to two major research
initiatives in Britain, which might otherwise escape attention. Moreover, the
parochialism does not involve description of the practice of British govern-
ment. Details of the operation of policy networks are renounced for an
exploration of the concept itself. Such theoretical concerns transcend national
boundaries. The early stages of my journey began here.1
THE L I TE RATURE ON NE TW ORKS
The first problem is to provide an organizing format for the literature on

networks. A thematic format is ruled out by the diffuse nature of the subject.
A chronological approach is not helpful when the survey has to cover a
heterogeneous subject and to encompass the range of contributions, recog-
nizing that there will be some loss of detail. The various approaches have been
classified by academic discipline and by level of analysis (Figure 2.1).
The Micro-level of Analysis
Group Dynamics
The distinctive features of social psychology’s contribution to the study of
networks lie in the analysis of small-group dynamics, the use of laboratory
1
This is an edited version of: R. A. W. Rhodes (1990) ‘Policy Networks: A British Perspective’,
Journal of Theoretical Politics, 2: 292–316. © Sage Publication. Reprinted with permission.
Levels of Micro Meso Macro

analysis
Discipline
Sociology Group dynamics Interorganizational analysis Political
Social network sociology
analysis
Political Science Issue networks Subgovernment/intergovernmental Neo-pluralism
relations
Figure 2.1 Approaches to networks.
experiments as the preferred method, and the frequent resort to some species
of role theory. Thus, the concept of ‘role set’ is described by Katz and Kahn
(1978: 189) as a ‘vast fishnet’ and any one knot in the net is directly attached to
many others. These linkages make up the role set of an office in an organiza-
tion. This notion was adopted by Evan (1976a: 79) to develop the notion of
‘organization set’, an early contribution to interorganizational analysis (see
below p. 18).
Alternatively the ‘communication networks’ of small groups have been
analysed to determine their influence on group effectiveness (Handy 1985:
180–1). Much work has been carried out on groups in organizations and
the social psychology of organizations is a thriving field (for a summary see
Schein 1980). ‘Informal organization’ is no longer viewed simply as behaviour
that deviates from the managerial structure and expectations but covers the
network of social relationships within organizations. It is but a small step from
the analysis of group relationships and interpersonal communication within
an organization to the analysis of such relations between organizations. (For
useful collections of articles see Cartwright and Zander 1968 and Crosbie
1975; and for a survey of the field see Lorsch 1987.)
Social Network Analysis

This approach is commonly found in social anthropology, and Barnes (1954)
is perhaps the classic study employing the concept of network. In his account
of a Norwegian parish, the term ‘social network’ is described, employing
points and lines: ‘The points . . . are people, or sometimes groups, and the
lines indicate which people interact with each other . . . ’ (1954: 45).
Analyses of social networks multiplied afterwards and, as Whitten and
Wolfe (1972: 721) comment, the term was rarely given a precise definition.
Similarly, Barnes (1972: 3) pointed out that, as the term became fashionable,
Policy Networks in Britain: The Early Years 17
‘terminological confusion’ increased. As Whitten and Wolfe (1972: 729–32)

show, the theoretical underpinnings of networks are many and varied, includ-
ing role theory, exchange theory, and action theory if they are specified.
A flavour of the approach is provided by Mitchell (1969). He distinguishes
between the morphological and interactional characteristics of networks. The
former refers to the pattern or structure of the links. The latter refers to ‘the
nature of the links themselves’ (1969: 12). Various concepts are used to
describe the structure of networks; for example, anchorage, range, reachability,
degree, size, and density, and several measures have been developed; for
example, density is defined as:
S 100
½ N ðn1Þ
where S = the number of direct contacts and n = the number of actors.
The various elite studies of, for example, interlocking directorates share
many of the characteristics of social network analysis (see Stanworth and
Giddens 1974).
In sum, social network analysis is preoccupied with describing and meas-
uring linkages (see, for example, Mitchell 1969; Whitten and Wolfe 1972; and
more recently Scott 2012).
Issue Networks
This subsection deliberately omits British contributions; they are covered in
the next section. Jordan (1990) cannot identify a continuous intellectual
history for the concept of networks in political science. It is anticipated by
Truman (1951: 444) and in such phrases as ‘the inflexibility of the established
web’ of relationships. Heclo and Wildavsky’s (1974) analysis of the public
expenditure process in Britain is a prime example of the micro‐level analysis
of networks by political scientists, focusing on: ‘the personal relationships
between major political and administrative actors’ (Heclo and Wildavsky
1974: xv). Their focus on the ‘village’ community is an example of political
anthropology, looking at the patterns of interactions between senior admin-
istrators and politicians. A similar micro-level emphasis can be found in
Heclo’s (1978: 102) discussion of ‘issue networks’:
Looking for the few who are powerful, we tend to overlook the many whose webs
of influence provoke and guide the exercise of power. These webs, or what I will
call ‘issue networks’, are particularly relevant to the highly intricate and confusing
welfare policies that have been undertaken in recent years.
Heclo is challenging the prevailing emphasis on ‘iron triangles’ or ‘sub‐
governments’ (see p. 19) arguing that policy-making is fragmented,
with a large and unpredictable number of participants. Atomization or
‘fairly open networks’ have replaced ‘the closed circles of control’ (Heclo
1978: 88).
The distinctive features of these contributions are the way in which the
concept is used primarily as a metaphor for a fluid set of personal relation-
ships. However, the major contribution of political science to the study of
networks has been at the meso-level rather than the micro-level.
The Meso-level of Analysis
Interorganizational Analysis
I have reviewed this literature at length elsewhere (Rhodes 1980), so I
paraphrase the more relevant conclusions of that survey.
Elkin (1975: 175–6) provides a typical summary of the concerns of inter-
organizational analysis:
At the same time as the focal organisation attempts to manage its dependencies
by employing one or more strategies, other organisations in the network are
similarly engaged.
Of some significance for the present exercise, European sociology has made a
major contribution to the development of this subfield (see especially Crozier
and Thoenig 1976; Karpik 1978; Hanf and Scharpf 1978; for American
contributions see Thompson 1967; Evan 1976b).
However, this concern with the complex of organizational interactions is
not without its problems. First, interorganizational analysis has been preoccu-
pied with classifying and measuring interactions between organizations, putt-
ing the methodological cart before the horse of substantive theory. Second,
interorganizational analysis’s concern with the ‘figure’ or surface level of
interactions needs to be supplemented with an analysis of the ‘ground’ or
the structure of power, values, and interests that support the surface inter-
actions. Third, the distribution of power within and between organizations is
an essential element in any analysis of the ‘ground’. Fourth, the analysis of
power cannot be restricted to actors’ resources but must also encompass, for
example, perceptions of power and the rules governing interactions (Rhodes
1981: 60). Finally, the focus on networks of organizations, and on the power-
dependence relations between organizations, is essential for understand-
ing the patterns of organizational relationships, of personal relationships,
and the changing nature of each (see Rhodes 1981: ch. 5).
Subgovernment
Alternative phrases to subgovernment include subsystem politics, whirlpools,
iron triangles, and triple alliances. These terms are used in many accounts of
American policy-making to refer to ‘small groups of political actors, both

governmental and non-governmental, that specialise in specific issue areas’
(Ripley and Franklin 1980: 7). Typically they make ‘most of the routine
decisions in a given substantive policy area’ (1980: 8). The distinction between
government and non-government becomes blurred, and participation in a
subgovernment is the most effective form of access for private interests.
Provided the participants are prepared to compromise and can avoid contro-
versy, they can run ‘their’ policy area for a long time.
As Redford (1969: 102–6) argues subgovernments provide ‘stability for exist-
ing equilibriums among interests’: ‘continuous access and superior opportunities
for influence to high-quantity, aggregated interests’; ‘some access and represen-
tation to interests that are not dominant’ and only change ‘through macro-
political intervention that modifies the rules and roles operating in the systems’
(for a more detailed exposition and references see Jordan 1990; Ripley and
Franklin 1980; Freeman 1955; Freeman and Stevens 1987).
Although this section has focused on the American literature, there is a
literature applying the language of subgovernments to Britain, most notably
in studies of the subgovernment of education (see Kogan 1975; Manzer 1970).
More generally there are several relevant studies on Western Europe (see
Richardson 1982, and the citations there for the individual country chapters;
and on the ‘Nordic tradition’ and its studies of ‘structural co-optation’ or
‘corporate pluralism’ see Heisler 1979; Olsen 1983; Jordan and Richardson
1987b). The problems with this approach are fourfold. First, there is no agreed
terminology and Jordan (1990) warns against generalizations covering such
diverse literature.
Second, the popular alternative term, almost synonym, is ‘iron triangle’. This
metaphor and its attendant image of a closed, limited, and isolated—choose your
preferred adjective—policy-making process is altogether too rigid, and stereo-
types the literature on subgovernment (Freeman and Stevens 1987: 12–13).
Third, these several authors do not explicitly use the term ‘policy network’.
They have been classified as network approaches with the benefit of 20/20
hindsight. The various metaphors employed have an obvious affinity with
network ideas but affinity and explicit conceptualization are different.
Finally, subgovernment and related concepts are used to describe relations
between government and interest groups rather than to explain the growth
and development of such relations. Distinctively, there is a penchant for
prescription: iron triangles are seen as a distortion to be eradicated (see, for
example, Lowi 1969).
Intergovernmental Relations
Among the earliest proponents was Anderson (1960: 3), who defined intergov-
ernmental relations (IGR) as: ‘an important body of activities or interactions
occurring between governmental units of all types and levels within . . . the federal
system’. This general definition has been elaborated by Wright (1974: 1–16), who
identifies five distinct characteristics. First, IGR recognizes the multiplicity of
relationships between all types of government. Second, it emphasizes the inter-
actions between individuals, especially public officials. Third, these relationships
are continuous, day‐to-day, and informal. Fourth, IGR insists on the important
role played by all public officials, be they politicians or administrators. Finally, it
emphasizes the political nature of relationships. It focuses on substantive policies,
especially financial issues such as who raises what amount and who shall spend it
for whose benefit with what results (see also Wright 1978). In summary, Wright
(1974: 4) claims that IGR focuses on ‘the multiple, behavioural, continuous and
dynamic exchanges occurring between various officials in the political system’.
This ‘visual filter’, if ‘novel’, does not constitute a theory. It is more a checklist for
collecting data. However, Beer (1973, 1976, 1978) does provide an explanation
for some recent changes in American federalism that have been influential in the
IGR literature.
Beer (1978: 17) argues that two types of influence have become prominent
in American IGR: functional specialization, referred to as the ‘professional–
bureaucratic complex’, and territorial specialization, referred to as the ‘intergov-
ernmental lobby’. The emergence of the professional–bureaucratic complex
reflects the scientific advances of the post-war period. Professional specialisms
have abounded and developed a new role in policy-making. In alliance with the
interested legislators and affected interest groups, the professions have become
key actors in the subgovernments. They exercise a decisive influence in fields
such as health, housing, urban renewal, transportation, education, and energy.
Moreover, these programmes were not being implemented directly by the federal
government and, as a result, vertical bureaucratic hierarchies have emerged
facilitated by the shared discipline of professions.
The outcome of this rise of technocracy is not dictatorship by men in white
coats but the rise of ‘counter-vailing power in the form of the intergovern-
mental lobby’ (Beer 1978: 18). This ‘topocratic’ influence (from topos meaning
‘place’, and kratos meaning ‘authority’) has challenged the pre-eminence of
the technocrats because mayors and other executives ‘developed a heightened
interest in and increasing contact with federal policy making and administra-
tion’ (Beer 1978: 18).
The result of this development has been a complex pattern of both central-
ization and decentralization. The process has been centralizing because the
programmes were formed by federal technocrats but decentralizing in that
they were adopted at the subcentral level. It has drawn state and local officials
into the federal government but it has also exposed the federal government to
the topocratic perspective. Again, the American focus should not obscure the
European contribution (see, for example, Hanf and Scharpf 1978; Kaufman et al.
1986; Scharpf et al. 1976; Thoenig and Friedberg 1976; see also the individual
country chapters in Page and Goldsmith 1987 and Rhodes and Wright 1987).
Inevitably, intergovernmental theory has weaknesses. First, as Tarrow

(1978: 1–2) points out, intergovernmental theory fails to link the networks
and their policy outcomes to conflicts of interest and ideology (see also
Dunleavy 1984: 60). The question of who benefits is ignored.
Second, the focus on technological and organizational imperatives provides
an incomplete account of the role of government. Its role is seen as benign and,
if it is not seen as the referee of group conflicts, equally its relationships with
particular interests are not explored.
Finally, the focus on the similarities between advanced industrial societies
(that is, on the emergence of disaggregated policy networks) diverts attention
away from the differences between networks. Whether the causes of variation
are the type of policy or the social interests affected, intergovernmental theory
disregards overt differences between, for example, economic and social welfare
policy-making.
The Macro-level of Analysis
Political Economy
Interorganizational analysis has ‘a relatively narrow theoretical orientation’. It
‘has been guided largely by a restrictive practical concern with the coordin-
ation of public services’ (Benson 1982: 137). A broadening of scope would
seem in order. Benson (1975, 1982) has sought to reorient the field by
providing a political economy of interorganizational networks. He too sees
organizations pursuing scarce resources, in this case money and authority, but,
and crucially, he argues that this interorganizational network is linked to a
larger environment. The network has to be contextualized: that is, related to
‘the major structural problems of advanced societies, particularly those of the
state’ (Benson 1982: 147). His basic unit of analysis is the policy sector or
‘arena in which public policies are decided and implemented’. These arenas
are seen as ‘complexes of resource dependencies’ (Benson 1982: 148). The
difference with other conceptions of networks begins with the recognition that
‘the policy sector is a multi-levelled social structure’. He distinguishes between
surface level and deep structures. The surface-level structures are adminis-
trative arrangements, policy content, and interorganizational dependencies.
Deep structures are interest-power structures and rules of structure formation
(Benson 1982: 149). It is this latter category that is of most relevance here.
The term ‘interest-power structure’ refers to ‘those groups whose interests
are built into the (policy) sector, either negatively or positively’. Benson
employs a five-fold classification: demand groups (for example, recipients of
services), support groups (for example, resource providers), administrative
groups (for example, occupants of central administrative positions), provider
groups (for example, service deliverers), and coordinating groups (for example,
programme rationalizers). The interest-power structure restricts policy options.
It is the source of conflict in policy sectors and between interests; and it is central
to an analysis of the distribution of power in a policy sector.
However, at the deepest level are the rules of structure formation that set
the boundaries to policy sector operation. These rules are ‘generated by the
linkages between policy sectors’: ‘by the requirements of the larger social
formation’, especially by ‘the necessity to maintain the accumulation process
and to produce justifications for the order of things’ (1982: 161). Not only do
these rules set boundaries but they also generate contradictions or ‘fundamen-
tal structural inconsistency’ (1982: 164).
In short, therefore, Benson is seeking to provide an ‘interorganisational
theory of the state’ in which ‘the structure and contradictions of the [policy]
sector are linked to the developmental logic of the capitalist state’. As a result, a
‘complex of interorganisational resource dependencies constitutes a structure
of class domination’ (Benson 1982: 176; and, for other attempts to provide
a macro-theoretical context for organizational networks, see Crozier and
Friedberg 1980; Aldrich 1979; and Perrow 1986).
Neo-pluralism
The ability of neo-pluralist theory to provide a broader theoretical context for
analysing policy networks is discussed in the next section. Here, I provide an
introduction to some key themes of neo-pluralist theory. They act as an
antidote to some of the grosser misrepresentations of pluralist theory, too
many of which verge on caricature and presuppose the proponents of the
theory are suffering from terminal brain damage (see, for example, Dearlove
and Saunders 1984: 57–61).
In discussions of IGR, neo-pluralism explores the impact of professional
influence, the logic of technical rationality, the privileged position of a select
number of interest groups, and the complex interdependencies with decentral-
ized governmental structure. These themes are relevant in several advanced
industrial liberal democracies. Thus, Hanf (1978: 1–2) argues that the charac-
teristic problem of such countries is that ‘the problem solving capacity of
governments is disaggregated into a collection of subsystems with limited
tasks, competences and resources’. The central problem of government is,
therefore, to secure coordinated policy actions through networks of separate
but interdependent organizations.
The limits to rational policy-making, the factorizing and professionalization
of policy systems, the interdependence of governmental organizations and
the emergence of policy from network interaction are said to be recurrent
features of advanced industrial society. Oligopoly has replaced the free market
competition between groups said to characterize pluralism (for a more
detailed summary and citations, see Dunleavy and O’Leary 1987; and, for a
good example of the approach, see Kaufman et al. 1986).
Perhaps one of the major weaknesses of the neo-pluralist literature is the
tendency to develop specific themes rather than to provide a coherent state-
ment of the approach. Consequently, it is important to recognize that the
theory, like modern Marxist theory, comes in many guises. There are now
several reviews of the current state of play, which cover the several variants
(see McFarland 1987, and Alford and Friedland 1985, both of which contain
numerous additional references). Moreover, neo-pluralist theory is increas-
ingly concerned with the role of the state (see Dyson 1980; Nordlinger 1981;
and, for a brief summary and more citations, see Dunleavy and O’Leary 1987:
ch. 6). It is also important to note that the discussion of neo-pluralist theory in
this chapter does not seek to mount a defence of pluralism or to mount a
critique of corporatist theory (see Cawson 1986a, 1986b; Jordan 1981, 1984,
1990; Rhodes 1985a). Also, my coverage of the literature is limited to those
contributions of direct relevance to the study of policy networks (for a
discussion of the relationship between networks, pluralism, and corporatism,
see Marsh and Rhodes 1990).
BRITISH CONTRIBUTIONS TO THE S TUDY

OF PO LICY NETWORKS
The primary objective of this section is to provide a summary of a growing

body of work on policy networks in Britain by British academics. Briefly, the
contributions fall into three categories: the precursors of the approach, most
notably the work of Jeremy Richardson and Grant Jordan; the Social Science
Research Council (SSRC) research initiative on central–local government
relations; and the SSRC initiative on government–industry relations (GIR).
The contributions of particular individuals outside of these research initiatives
will be noted on the way.
Perhaps Richardson and Jordan made the earliest contribution to the study
of policy communities in the UK (Richardson and Jordan 1979; Jordan 1981;
Richardson 1982; Jordan and Richardson 1987a, 1987b). They argue that
policy-making is fragmented into subsystems, called policy communities,
and Richardson and Jordan (1979: 48–57, 103–5) identify the varieties of
consultation processes, for example, tripartism, clientelism, and the rules of
the game governing relationships; for example, conflict avoidance and secrecy.
They argue there is a dominant British policy style (generated by the policy
communities), termed ‘bureaucratic accommodation’ (Jordan and Richardson
1982). The aim is to create ‘a nexus of interests so that co-operation flows from
a sense of mutual advantage’ (Richardson and Jordan 1979: vii). The style is
‘deal seeking’ and consensual (Jordan and Richardson 1987b: 29–33).
Richardson and Jordan’s approach has four distinct features. First, it is
explicitly a defence of pluralism against caricaturing critics and corporatist
theory (Jordan 1981, 1984). Second, the term ‘policy community’ is used to
describe British policy-making processes; it is a metaphor. Third, following the
example of their acknowledged forerunners Heclo and Wildavsky (1974),
networks are seen as personal networks. Finally, the emphasis on a dominant
British policy style is a product of the search for a way of comparing the
policy-making processes in West European polities.
A major boost to the study of policy networks was provided by the (then)
SSRC research initiative on central–local government relationship (see SSRC
1979). The rationale for the initiative was provided by ‘the Rhodes model’ (see
Rhodes 1981), so this section focuses on it. Several participants in that initiative
also contributed to the study of policy networks. The publications of the various
projects are listed in Goldsmith and Rhodes (1986) and the research findings are
discussed at length in Rhodes (1988). To discuss each contribution would lead
to a substantial degree of repetition. Consequently, I simply note the body of
work produced by the SSRC initiative and commend, in particular, Barrett and
Fudge (1981); Goldsmith (1986); Gyford and James (1983); Jackson (1985);
Laffin (1986); and Ranson et al. (1985). In addition, there were several contri-
butions on policy networks, which were a reaction to the SSRC initiative but not
part of it (see, for example, Houlihan 1988; Marsh and Rhodes 1992a; and
Sharpe 1985). Their critical comments are noted at the relevant point.
The Rhodes Model
Rhodes (1981, especially chs 1 and 5) provided an explicit application of

intergovernmental theory to British central–local relations. In the initial
version of the model, central–local relations are viewed as a ‘game’. Both
central and local participants manoeuvre for advantage deploying the
resources they control to maximize their influence over outcomes. Everyone
tries to avoid becoming dependent on the other ‘players’. The relevant
resources include constitutional-legal, organizational, financial, political, and
informational resources.
This game is treated as the ‘figure’ or the micro-level of analysis. It is also
necessary to explain changes in the distribution of resources and the rules of
the game: to contextualize the patterns of interaction. This context or ‘ground’
is explored using Schmitter’s (1979) concept of corporatism. It is argued that
central–local relations are moving away from competition and bargaining
between local authorities and central departments towards a system in which
organizations are aggregated in policy communities limited to the accredited
spokespersons for local government (Rhodes 1981: 111–25). Competition is

limited and a system of interest intermediation—literally the group ‘comes
between’ its members and government—is created in which the membership
is regulated in return for influence over government policy. This shift to corpor-
atism is said to be related to the changing economic context of government.
In short, the game of central–local relations is shifting from pluralistic
bargaining to corporatism. This framework is clearly within the intergovern-
mental tradition and it has been subject to several criticisms (for a summary,
citations, and commentary, see Rhodes 1986b, and Chapter 12, this volume).
Perhaps the most significant failing of the power-dependence model is its
failure to distinguish clearly between the levels of analysis and, consequently, it
does not adequately explore the relationship between them. This problem
stems from the use of corporatist theory to analyse the ‘ground’. It is import-
ant to distinguish between corporatism as a form of government–interest
group relations and as a theory of the state (Marsh 1983: 1). As a form of
government–interest group relations, corporatism has been applied not only
to economic functional groupings but also to other types of groups and sub-
central government (Cawson 1978, 1982; Foster et al. 1980). Rhodes (1986b)
argues that corporatism provides a rigid metaphor of government–interest
group relations and that it cannot be extended to subcentral government.
The characteristics of corporatism such as the aggregation of interests, licensing
of groups, monopoly of representation, and regulation of members, are rarely
found in unsullied form. This state of affairs prompted adjectival proliferation—for
example, neo-corporatism, liberal corporatism, and bargained corporatism
(and for more detailed reviews of corporatism see Jessop 1978 and Jordan
1981). Clearly, a measure of reconstruction was necessary for ‘the Rhodes
model’, a task undertaken in Rhodes (1986a, 1986b).
The first step is to distinguish clearly between levels of analysis. The
macro-level of analysis necessarily involves an account of the changing
characteristics of British government during the post-war period. The
meso-level of analysis focuses on the variety of linkages between the centre
and the range of subcentral political and governmental organizations. The
concept of policy networks is appropriate at this level of analysis. The micro-
level of analysis focuses on the behaviour of particular actors, be they
individuals or organizations. In the following summary the emphasis falls
on the inter-relationship between the macro- and meso‐levels, which are
seen as crucial to any explanation of the changing pattern of network
relationships and their outcomes.
Following Benson (1982: 148) a policy network can be defined as:
a cluster or complex of organisations connected to each other by resource-

dependencies and distinguished from other clusters or complexes by breaks in
the structure of resource-dependencies.
Rhodes (1986a: ch. 2) elaborates this definition, arguing that networks have
different structures of dependencies, structures that vary along such dimen-
sions as membership (for example, professions, private sector), interdepend-
ence (for example, between levels of government), and resources. He also
distinguishes between five types of networks ranging along a continuum from
integrated policy communities to fragmented issue networks.
Policy communities are networks characterized by stability of relationships,
continuity of a restrictive membership, vertical interdependence based on
shared service delivery responsibilities, and insulation from other networks
and invariably to the public (including parliament). They have a high degree of
vertical interdependence and limited horizontal articulation. They are tightly
integrated. Policy communities are based on the major functional interests in
and of government; for example, education, the fire service (Richardson and
Jordan 1979; Rhodes 1986a: ch. 8).
Territorial communities encompass the major territorial interests, for
example, in Scotland, Wales, and Northern Ireland (Hunter and Wistow
1987; Rhodes 1986a: ch. 7).
Professional networks are characterized by the pre-eminence of one class of
participants in policy-making the profession. The most cited example of a
professionalized policy network is the National Health Service (see Ham
1981). The water service provides a further example in which the constraints
on water engineers seem particularly weak (Gray 1982; Saunders 1983: 34–7).
In short, professionalized networks express the interests of a particular pro-
fession and display a large degree of vertical independence while insulating
themselves from other networks.
Intergovernmental networks are the networks based on the representative
organizations of local authorities. Their distinctive characteristics are: topo-
cratic membership (and the explicit exclusion of all public sector unions); an
extensive constellation of interests encompassing all the services (and associ-
ated expertise and clients) of local authorities; limited vertical interdependence
because they have no service delivery responsibilities but extensive horizontal
articulation or ability to penetrate several other networks.
Producer networks are distinguished by the prominent role of economic
interests (both the public and the private sector) in policy-making; their
fluctuating membership; the dependence of the centre on industrial organiza-
tions for delivering the desired goods and for expertise; and the limited
interdependence among the economic interests.
The distinctive features of an issue network are its large number of parti-
cipants and their limited degree of interdependence. Stability and continuity
are at a premium and the structure tends to be atomistic (Heclo 1978).
Rhodes (1988: 48–77, 371–87) attempts to meet the criticism that the model
fails to provide an adequate analysis of the context within which policy
networks operate and consequently, is unable to explain how and why networks
change. It is argued that the ‘national government environment’ conditions the

operation of the policy networks. Six processes within the national government
environment have had a major effect on the networks: an unstable external
support system; the decline of the mixed economy; the growth of the welfare
state; the extension of the allied processes of functional differentiation and
professionalization; the development of a social structure characterized by
multiple (non-class) cleavages; and a political tradition characterized by a
two-party system, a unitary institutional structure, and a transitional central
elite ideology. Subsequently, the effects of the national government environ-
ment on policy-making are explored in several case studies of policy networks
‘in action’. For example, the case study of local government finance argues that
the search for control by the centre was impelled by the decline of the mixed
economy. It was confounded by a combination of the politics of inertia in
the guise of policy networks—those products of functional differentiation
and professionalization—and a central elite with a faulty operating code (see
Rhodes 1988: 371–87).
The GIR Initiative
The SSRC initiative on GIR was also based on ‘the Rhodes model’ and on the
analysis of interorganizational relations (SSRC 1983; Wilks 1989: 330). Some
of the publications reveal interesting differences of approach (Grant et al.
1988; Hancher and Moran 1989; Wilks and Wright 1987; Wright 1988a,
1988b; and Wilks 1989).
The general approach of the GIR initiative has three themes: ‘to break away
from system-level macro-generalisations’ in favour of ‘empirically based ana-
lysis’; a comparative focus; and the development of ‘a more productive theor-
etical approach’ (Wilks and Wright 1987: 275). Indeed, the first point to note
is that the GIR literature abounds with system-level macro-generalizations,
referred to by Wilks and Wright (1987: 282) as the strong state–weak state
orthodoxy. To oversimplify, the field of central–local relations focuses on
actors and their behaviour at the expense of analysing the context of such
behaviour, whereas GIR has developed broad theory that ignores, for example,
informal relationships and intra‐bureaucratic conflict. Wilks and Wright
(1987: 289) conclude that ‘in order to understand the operation of a frag-
mented bureaucracy within the GIR, we need a finer grained analytical schema’
(emphasis added).
The problem in GIR, therefore, is not to contextualize a substantial number
of case studies and meso-level theory but to provide both the meso-level
theory and the supporting, detailed case studies. Thus, Wilks (1989: 330)
describes the research strategy as the promotion of ‘intermediate level studies
of policies, processes and industrial sectors to provide an empirical basis for

medium level generalisations’.
Their preferred ‘fine grained analytical schema’ required a modified
interorganizational framework. It was ‘developed towards cross-sectoral and
cross-national comparisons with an emphasis on identifying, comparing and
classifying “policy networks”’ (Wilks 1989: 330). Which of the cross-national
or cross-sectoral variations were the most significant remained ‘an open but
important question’ (Wilks and Wright 1987: 290). These modifications lead
the GIR initiative to depart ‘in one important particular from Rhodes’ definition
and use of policy community’ (Wright 1988a: 604). The term ‘policy community’
refers to ‘actors and potential actors drawn from the policy universe who share a
common identity or interest’. The term ‘policy networks’ refers to the linking
process, the outcome of these exchanges ‘within a policy community or between
a number of policy communities’ (Wright 1988a: 606; see also Wilks and Wright
1987: 299). Thus industry is ‘neither monolithic nor homogeneous’, government
is ‘fragmented, differentiated and fissiparous’, and the key to understanding GIR
is to disaggregate to the subsectoral policy networks.
Wilks and Wright (1987: 295) see these distinctions as essential for the
comparative study of networks and argue that there are several advantages to
their formulation. Thus, their definition of policy community enables them to
distinguish between groups of actors within the community both at the
sectoral and subsectoral level; to identify actors excluded from a policy
network; and to compare the membership of networks within the same
community (1987: 300).
Grant et al. (1988) illustrate the usefulness of a subsectoral focus in their
study of the chemical industry. They seek ‘to identify the key actors making up
the “policy communities” centred in the chemical industry, to identify the
roles for those actors and the linkages between them, and to explore any
shared perceptions or informal “rules of the game” which might influence the
interactions taking place’ (Grant et al. 1988: 3). The concept of policy com-
munity refers to economic sectors that ‘have a political identity and life of their
own which is to some extent distinctive and insulated from that of other
sectors’ (1988: 10). The policy communities do not ‘conform to . . . adminis-
tratively imposed boundaries’ and the authors stress that they are looking
at personal relationships within a shared framework (1988: 11). A policy
community is said to have three characteristics: differentiation, specialized
organizations and policy‐making institutions, and interaction (1988: 55).
Employing several indicators to measure these characteristics, Grant et al.
(1988: 67–74) identify a core chemical policy community (at national and
the EU levels), and four subsectoral policy communities: pharmaceuticals,
agricultural chemicals (which is divided into fertilizers and agrochemicals),
the paint industry, and the soap detergent and toilet preparations industry.
Grant et al. (1988: 74) conclude that the concept of policy community is a
‘useful conceptual tool for ordering the material’ but they emphasize that ‘any
analysis which ignored the sub-sectoral level would be incomplete’.
Hancher and Moran (1989: 272) explore economic regulation, which is
viewed as ‘a process of intermediation and bargaining between large and
powerful organisations spanning . . . the public and private domains of
decision-making’. They reject the ‘capture theory’ of regulation and the
dichotomy of public authority versus private interest (1989: 276). They
focus on ‘regulatory arenas’ and ‘the outcomes of competitive struggles, the
resources used in these struggles, and the distribution of those resources
between different involved institutions’ (1989: 277). Moreover, a regulatory
arena contains a diversity of organizations and, as one might expect, their
interactions are conceptualized in terms ‘of networks or linkages of varying
density of formality’ (1989: 291).
The GIR initiative has produced already a distinct twist to the study of
policy networks. To such accepted characteristics as interdependence and
resource exchange, it has added disaggregation to the subsectoral level, a
concern with personal rather than organizational networks, and a rejection
of macro-level theory for middle-range theory. Indeed, Wilks and Wright
(1987: 298) begin their redefinition of policy networks with a discussion of
social network analysis (and references to interlocking directorships (1987:
313)), and Wright (1988b) is a micro-level analysis of the City and the
takeover panel. In sum, Wright’s reformulation seeks precise and operational
definitions of communities and networks to facilitate micro-level analysis. It is
exactly the opposite line of development pursued by Rhodes and yet reflects
the current ‘state of the art’ in GIR studies.
In short, there has been a substantial British contribution to the study of
policy networks. However, what, if anything, is distinctive about it?
First, American political science was not the major formative influence on
British developments. If there is a single key influence on the Rhodes model
(and, therefore, the GIR initiative), then it is European sociology and its
approach to interorganizational analysis. Any search for the American ante-
cedents of the network concept is mistaken because the roots of British
developments do not lie across the Atlantic.
Second, the UK literature has had an explicit concept of networks and it has
abjured other metaphors with the beneficial effect of avoiding the termino-
logical profusion and confusion of the American literature on subgovernment.
Third, the British contribution has had an explicit analytical and compara-
tive dimension from the outset. It did not focus on personal but organizational
networks. It sought to explain as well as describe the pattern of organizational
relationships. It did not focus on measuring either personal linkages or
organizational exchanges. Comparison has always been a central feature,
whether of national policy styles, the policy networks of the several welfare
state services, or of industrial sectors.
Fourth, and without suggesting that the theory is adequate, the British
contribution has sought to locate the concept of networks within a revised
pluralist theory. There is a marked contrast with, for example, that American
literature which views subgovernments as a distortion of the political process
to be corrected.
Finally, the British literature does not manifest much concern with the
coordination of public services. Whether at a practical or a theoretical level,
it has not sought to prescribe.
In short, the British contribution to the study of policy networks is dis-
tinctive, and not derivative of American political science, although it would be
instructive for others to consider the extent to which the British contribution
still betrays its origins in interorganizational analysis. Even if this assessment
of the British contribution is correct, however, the various contributions still
have their weaknesses.
CONCLUSIONS: P ROBLEMS AND DEVELOPMENT
The survey of the literature drew attention to its varied disciplinary origins,
increasingly varied terminology, mutually exclusive definitions, and, especially,
to the varying levels of analysis. In effect, it is possible to distinguish between
personal networks, organizational networks, and political networks (provided
this latter phrase is taken to refer to networks and their political context). When
both the unit and the level of analysis vary, it is clearly important to exercise care
in using (and evaluating) the concept.
The commonest criticisms made of the literatures on interorganizational
analysis, intergovernmental relations, and subgovernments are that the con-
cept of networks is used descriptively, and that the theoretical scope is narrow.
Given that the Rhodes model drew on these literatures, it too has been
criticized for failing to analyse the context within which policy networks
operate. Hopefully, the literature review demonstrates that such criticisms
are no longer accurate. There are now several efforts to ‘contextualize’
networks.
The discussion of the macro-level of analysis contrasted a political economy
approach with a neo-pluralist approach. Obviously, other approaches could
be employed: for example, is there a viable New Right approach to policy
networks? There are some obvious affinities with Henney’s (1984: 381) com-
plaint that sectors of the corporate state have ‘monopoly power and seek to
limit competition and change’. More formally, Ostrom’s (1986: 460) concept
of ‘action arenas’ with the attendant notions of, for example, rules and
resources, offers a potential bridge between the currently separate literatures
on networks and rational choice. Within the GIR initiative, Wilks (1989:
335–6) has noted that the framework has not prevented the development of
several ‘theoretical emphases’. Dunleavy (1980: 131) has argued that research
‘should not only be theoretically based . . . but should be multi-theoretical; that
is, it should draw on several or all of the theories relevant to the empirical
questions, using them as sources of competing hypotheses and interpretations
to guide the research’. ‘Policy network’ is a meso-level concept which needs to
be, and can be, located within different macro-theoretical approaches. It is
important that the concept be so located and that it be done in a self-conscious
manner. If Dunleavy’s injunction to draw on several theories is a counsel
of perfection, nonetheless the call for explicit theory must be a minimum
condition.
These general problems apart, there are problems with each of the three
British contributions. First, Richardson and Jordan use the concepts of policy
communities and networks to generalize across national policy-making sys-
tems. But this macro-level search for dominant policy styles also dilutes the
approach. It becomes impossible to identify policy variations within a country
because the characteristics of a given policy style are so broad (for a more
detailed critique, see Rhodes 1986a: 22–34). Yet the comparison of policy areas
at the subsectoral level is one of the more promising lines of development.
Some of the more common criticisms of the Rhodes model were touched on
in the earlier discussion of intergovernmental theory. Three weaknesses were
identified: it ignores social interests; it has an inadequate conception of the
state; and it does not explain the causes and consequences of variations
between policy areas (see, for example, Tarrow 1978; Dunleavy 1984:
58–60). Rhodes (1988: 254–5, 272–4, 284–6, 303–6, 325–7, 338–43, 354, and
387–406) discusses the consequences for social interests of the different types
of networks drawing on Parkin’s (1979) social closure model of social strati-
fication. Rhodes (1988: 97–8) follows Nordlinger (1981: 9) in defining the state
as ‘a complex set of institutional arrangements for rule operating through
continuous and regulated activities of individuals acting as occupants of
offices’. The differences between policy networks are discussed in detail in
Rhodes 1988 (ch. 4, and the summary on pp. 368–71). Whether the analysis of
these topics is adequate remains to be determined but the model warrants a
more extended discussion than the assertion that social interests are ignored.
There are other criticisms of the Rhodes model that may limit its utility.
Deliberately, it focuses on welfare state services involving subcentral govern-
ments and, usually, a profession or semi-profession. Such a definition of
network interests and membership is narrow and may well limit the utility
of the concept in, for example, industrial policy-making. This point is amply
illustrated by Grant et al. (1988: 58–67). They show that the government has
not involved itself with the chemical industry and the power-dependence
relationship favours the industry. Informational resources are the key cur-
rency in the chemical industry policy community and most of the important
information is concentrated in the companies. The rules of the game are fluid,
although ‘trust’ is one basic rule. Strategies are deployed to manage relation-
ships within the industry, not relationships between the government and
industry. In short, government–industry relations in the chemical industry
are markedly different from the intergovernmental relations studied by
Rhodes (1988: 67). Although Rhodes (1988: 327) recognized government
dependence on firms and the pre-eminence of economic interests and
argued that producer networks were markedly different from other types
of policy network, nonetheless his argument that producer networks were
loosely integrated is misleading. Such a conclusion may be accurate at the
industry level but, as Grant et al. (1988: 314) conclude, sectoral analysis is
crucial to understanding GIR, ‘sectoral variations do at least modify national
characteristics’, and the degree of integration at the subsectoral level can be
considerable.
The differences between the Rhodes model and the other British contribu-
tions to the study of networks raise instructive problems and questions.
Wright’s reconstruction of the concepts of policy communities and policy
networks involves disaggregating the concepts to foster micro-level analysis.
Rhodes (1988: 370–1) also noted the need to disaggregate networks. Such a
research strategy is a ‘bottom-up’ approach; gives much weight to the role of
individuals and individual personality; and raises the (perennial) problem of
generalizability because each detailed case study is ‘unique’. There is a more
fundamental problem. In the National Health Service, it is possible to identify
several ‘sectoral’ policy networks (see Haywood and Hunter 1982; Hunter and
Wistow 1987). But if the health policy network can be disaggregated in this
manner, what holds it together? Does it make sense to talk of a professional-
ized policy network? Is a shared, mechanistic model of health the ideological
glue holding the network together? Or, inevitably, does the usefulness of the
concept evaporate as the focus shifts to the micro-level of the interactions in
personal networks? Organizational networks provide the context constraining
personal networks; they do not describe behaviour in those networks. If we
focus on individual behaviour, then necessarily the general concept will
become less useful. Equally, the theory will have little or no chance of being
parsimonious. To pursue micro-level analysis, to explore personal networks,
will provide a wealth of detail but make it increasingly difficult to generalize
about policy networks.
There are terminological problems with the modifications introduced by
the GIR initiative. Of course concepts should be modified by research findings,
but there seems little advantage in turning the concept of ‘policy community’
on its head. Almost every other author treats policy communities as tightly
integrated, commonly, but not exclusively, personal networks (cf. Grant et al.
1988: 11; Heclo and Wildavsky 1974: xv; Richardson and Jordan 1979). For
Wilks and Wright (1987) and Wright (1988a), however, policy communities
are not tightly integrated but, as already noted, cover ‘actor and potential
actors who share a common identity or interest’ (Wright 1988a: 606, emphasis
added). Without in any way disputing their argument for disaggregation and
subsectoral analysis, it can still be argued that their use of ‘policy community’
clouds rather than clarifies the issue and, indeed, is not followed by Grant et al.
(1988: 10, 11, and 55). Jordan and Richardson’s (1987b: 33–4) concept of
‘policy arena’, although it currently refers mainly to institutional arenas such
as parliament or the Cabinet, is a better candidate to describe potential actors
loosely linked by a shared identity. Policy communities can then continue to
be tightly integrated. Furthermore, and parenthetically, it would seem unhelp-
ful for Wilks and Wright (1987) to use the term ‘policy networks’ irrespective
of the degree of integration of subsectors. Obviously subsectoral policy net-
works vary in their degree of integration. Some adjectival recognition of this
fact would aid clarity.
Such terminological problems are not peculiar to the GIR initiative and
show every sign of growing. It is more productive to turn to matters of
substance. The comparison of the central–local and the GIR initiatives iden-
tifies four important lessons, which can be summarized as network character-
istics, comparison, disaggregation, and policy types.
First, Grant et al. (1988: ch. 3) do not attempt to provide a set of definitions
of the varieties of policy networks. They identify three characteristics of policy
communities: differentiation, specialization, and interaction. Rhodes (1988:
77–8) identifies interests, membership, interdependence (vertical and hori-
zontal), and resources. Marsh and Rhodes (1992a: 251) argues that policy
communities are characterized by a limited number of participants, frequent
interaction, continuity, value consensus, resource-dependence, bargaining, a
positive-sum power game, and regulation of members. These various charac-
teristics can be treated as a continuum (see Chapter 3). Alternatively, as
Saward (1990) suggests, some networks may be characterized by a dominant
interest, others by a high degree of interdependence: the relationship between
the dimensions needs to be explored. The degree to which any one or set of
characteristics is present, and the relationships between them, are not a matter
of definition. They are mainly matters for empirical investigation. Conse-
quently, developments do not hinge on the need for definitional agreement.
Nor are multiplying definitions a major problem, provided future research
looks for, and appraises, network characteristics and does not seek to pre-
empt, either by definition or by ‘ideal-type’ formulations, that which needs to
be investigated.
Second, the comparative study of IGR was a spin-off from the SSRC
central–local government relations initiative. It was not integral, and the
oversight was unfortunate (although see Page and Goldsmith 1987; Rhodes
and Wright 1987). The GIR initiative did not make the same mistake. Future
studies of policy networks should be cross-national.
Third, cross-national studies cannot assume that national variations are

significant. Sectoral variations may be more important and, in consequence, it
is also necessary to disaggregate and explore subsectoral variations.
Finally, it is clear there are marked variations between policy areas. The
cross-national, subsectoral comparisons must encompass several policy areas
and distinguish between grand and ordinary issues. The inclusion of an
economic and a welfare policy area would seem advisable. Alternatively, the
choice of policy area could be rooted in a typology of policy areas. The
emphasis would then fall on the conditions supporting the different policy
processes. Ripley and Franklin (1980: ch. 1) start with a typology of policy
areas, and only the relationships in their distributive policy arena would be
described as a policy network (Ripley and Franklin 1980: 22–3).
Add to this set of prescriptions a multi-theoretic approach and the research
agenda is daunting. It is also a measure of the distance travelled so far. The
concept of policy networks has added immeasurably not only to our under-
standing of IGR and GIR but also of policy-making in British government.
Inevitably, therefore, to develop the approach has to broaden its horizons. And
it is not a question of choosing between ‘the Rhodes model’ and the ‘GIR
modifications’: of determining which is ‘best’. Rather, it is important to ground
any appraisal of network characteristics, disaggregation, subsectoral and
cross-national variation, typologies of policy, and macro-level theory in stud-
ies of networks ‘in action’.
This chapter has categorized the literature on networks, appraised
the British contribution, and, finally, identified some problems and possible
lines of development. The research agenda may be daunting but the scale
remains modest. There are international and normative dimensions to
the study of networks. Thus, Keohane and Nye’s (1977: 19) analysis of
‘international regimes’ defines them as ‘networks of rules, norms and proced-
ures that regularize behaviour and control its effects’ (see also Krasner 1982
and, for a critical appraisal, see Strange 1982). Both Helen Wallace (1984: 141)
and William Wallace (1983: 409–10) have applied the concepts of regimes and
networks to the EU, exploring in the former case the emergence of trans-
national policy networks in, for example, food policy (and see Chapter 5,
this volume). The normative questions concern the closed, secretive nature
of this form of functional representation. In Rokkan’s (1966: 105) apposite
phrase, ‘votes count but resources decide’. Whose interests are favoured in
such decisions? The effects of networks on distributive policies place them at
the heart of any debate about equality.
The British contribution to the study of policy networks has demonstrated
that the concept has considerable utility for the analysis of policy-making in
Western democracies. Its heuristic value remains considerable because it
directly confronts, even mirrors, the administrative and political complexity
of advanced industrial societies.
A F T E R W O RD
I wrote this article to take stock of the work on policy networks during the
1980s. It is an apt starting point for this volume. By 1990, the concept of policy
networks was becoming ubiquitous. It was most commonly used as a meta-
phor. It was infrequently used with precision. It was rare for it to have any
explanatory value. There was no existing, comprehensive review of literature.
So, this chapter sought to draw attention to the variety of uses. From here my
interests broadened both theoretically and away from the study of local
government. And I had to start somewhere. Of course, it is a time-bound
piece. It does not need an afterword because I continue the story in Chapter 3,
this volume and reply to my critics in Chapter 12. However, revising this
chapter prompted four reflections, all of which I return to in later chapters.
First, I wrote in the modernist-empiricist tradition, which remains the
dominant idiom in the study of policy networks today (see, for example,
Klijn and Koppenjan 2015; Torfing et al. 2012). There is a marked contrast
with my later work in the interpretive tradition (see Chapters 6 and 7, this
volume). The difference is obvious in my style of writing. In this chapter,
I reveal myself to be a victim of American social science-ese. I cringed on
rereading the piece after I don’t know how many years. I decided to include
the chapter not only because it provides a baseline for the chapters that follow
but also because the change in language exemplifies the change in approach.
I have new neologisms to play with.
Second, looking back 30 years highlights the almost incestuous nature of
some of the debates. We were a small group of partisans and we argued over
definitions and typologies. The issues that divided us seemed important at the
time. They were not (see Chapter 3, this volume), although they can generate
heat even today (see, for example, the exchange between Jordan and Cairney
2013 and Marsh and McCaffrie 2015).
Third, there were many lazy criticisms. For example, Dowding (1994 and
1995) asserted that policy networks were but a metaphor. The comment was
accurate for Richardson and Jordan’s (1979) use of the phrase ‘policy com-
munities’ but is just plain wrong for the actor-centred, resource-dependency
model of networks (Rhodes 1988 and see Chapter 12). Yet Dowding’s criticism
is trotted out time and again (see Chapter 4, this volume for more examples).
There are problems with the policy network approach but they lie with
its modernist-empiricist epistemology and the reification of networks
(see Chapter 6, this volume).
Finally, from the vantage point of the 1980s, it was inconceivable that policy
networks would become standard fare in textbooks on British government.
They did. Research on networks blossomed. My suggestions that we disaggre-
gate and explore subsectoral variations, that we extend the analysis to the EU,
and that we needed more case studies of networks ‘in action’, anticipated the
directions of much future work. What I did not foresee was the practical use to
which research on policy networks would be put. The topic of how to manage
your network blossomed; it became the dominant concern (see Chapter 4, this
volume). More dramatically, there was the arrival of neo-liberal ideas, which
cast producer groups—networks by another name—as the bad guys and saw
strenuous efforts by Conservative governments to emasculate them. Policy
networks were no longer an idea for academics to play with. They became the
stuff of politics and management.
My story started in the 1980s when policy networks were an innovative
idea. The 1990s were their historical moment. I tell this story in the next
chapter. But obsolescence awaits all new ideas. There is no stopping, just
relentless pressure to reconstruct and not be swept away. My reconstruction
was the interpretive turn. But I get ahead of myself. The next step is to describe
the historical moment.
3
Policy Networks
The Historical Moment
INTRODUCTION: THE UBIQUITY OF NETWORK S
Network analysis comes in many guises. It is common to all the social science
disciplines. The vast literature ranges from social network analysis (Scott 2012)
to the network society created by the information revolution (Castells 2000),
from the actor-centred networks of technological diffusion (Callon, Law,
and Rip 1986) to cross-cultural analysis (Linn 1999). This chapter focuses on
that species of network analysis most common in political science—policy
network analysis.1
Few social science disciplines can ever agree on the meaning of an idea. So, a
policy network is one of a cluster of concepts focusing on government links
with, and dependence on, other state and societal actors. These notions
include issue networks (Heclo 1978), iron triangles (Ripley and Franklin
1980), policy subsystems or subgovernments (Freeman and Stevens 1987),
policy communities (Richardson and Jordan 1979), and epistemic communi-
ties (Haas 1992). I discuss these terms below. All are varieties of networks, so
I use ‘policy network’ as the generic term.
This buzzing, blooming confusion of terms has not detained us for long.
Defining policy networks will take no longer. Policy networks are sets of
formal institutional and informal linkages between governmental and other
actors structured around shared if endlessly negotiated beliefs and interests
in public policy-making and implementation. These actors are interdepend-
ent and policy emerges from the interactions between them. There could be
many qualifications to this definition, but it will do as a starting point for my
exploration.
1
An updated version of ‘Policy Network Analysis’. In M. Moran, M. Rein, and R. E. Goodin
(eds) (2006) The Oxford Handbook of Public Policy. Oxford: Oxford University Press,
pp. 423–45. Reprinted by permission of Oxford University Press.
THE L I TERA TU RE ON P OL I CY NE TWORKS
The term ‘policy network’ is used in three main ways in the literature: as a
description of governments at work, as a theory for analysing government
policy-making, and as a prescription for reforming public management.
Networks as Description
When describing government policy-making, the term ‘policy network’ refers

to interest intermediation, interorganizational analysis, and governance.
Networks as Interest Intermediation

The roots of the idea of a policy network lie, in part, in American pluralism
and the literature on subgovernments. For example, Ripley and Franklin
(1980: 8–9) define subgovernments as ‘clusters of individuals that effectively
make most of the routine decisions in a given substantive area of policy’. They
are composed of ‘members of the House and/or Senate, members of Congres-
sional staffs, a few bureaucrats and representatives of private groups and
organizations interested in the policy area’. The emphasis in this literature is
on a few privileged groups with close relations with governments; the resultant
subgovernment excludes other interests and makes policy. Some authors
developed more rigid metaphors to characterize this relationship. Lowi
(1964) stressed the triangular nature of the links, with the central government
agency, the Congressional Committee, and the interest group enjoying an
almost symbiotic interaction. This insight gave birth to the best-known label
within the subgovernments literature, the ‘iron triangle’ (see Freeman and
Stevens 1987: 12–13 and citations therein).
The literature on policy networks develops this American concern with
the oligopoly of the political marketplace. Governments confront a multi-
tude of groups all keen to influence a piece of legislation or policy imple-
mentation. Some groups are outsiders. They are deemed extreme in
behaviour and unrealistic in their demands, so are kept at arm’s length.
Others are insiders, acceptable to government, responsible in their expect-
ations and willing to work with and through government. Government
needs them to make sure it meets its policy objectives. The professions of
the welfare state are the most obvious example. Over the years, such interests
become institutionalized. They are consulted before documents are sent out
for consultation. They don’t lobby. They have lunch. These routine, stand-
ardized patterns of interaction between government and insider interests
become policy networks.
Policy Networks: The Historical Moment 39
Table 3.1 Types of policy networks.
Dimension Policy Community Issue Network
Membership: Very limited number, some Large

– No. of participants groups consciously excluded
– Type of interest Encompasses range of affected
Economic and/or professional interests
interests dominate.
Integration: Frequent, high-quality interaction Contacts fluctuate in frequency
– Frequency of of all groups on all matters related and intensity.
interaction to policy issues high quality
– Continuity Membership, values, and
– Consensus outcomes persistent over time Access fluctuates significantly.
All participants share basic values A measure of agreement exists,
and accept the legitimacy of the but conflict is ever present.
outcome.
Resources: All participants have resources; Some participants may have
– Distribution of basic relationship is an exchange resources, but they are limited,
resources within relationship. and basic relationship is
network Hierarchical; leaders can deliver consultative.
– Distribution of members. Varied and variable distribution
resources within and capacity to regulate
participating members
organizations
Power: There is a balance of power among Unequal powers, reflecting
members. Although one group unequal resources and unequal
may dominate, it must be a access. It is a zero-sum game.
positive-sum game if community
is to persist.
Source: Marsh and Rhodes 1992a: 251. The table was not in the original version of the chapter.
There are many examples of the use of policy networks to describe govern-
ment policy-making.2 Marsh and Rhodes (1992a) define policy networks as a
meso-level concept that links the micro-level of analysis, dealing with the
role of interests and government in particular policy decisions, and the macro-
level of analysis, which is concerned with broader questions about the distri-
bution of power in modern society. Networks can vary along a continuum
according to the closeness of the relationships in them (see Table 3.1).
Policy communities are at one end of the continuum and involve close
relationships; issue networks are at the other end and involve loose relation-
ships (and on the influence of this approach see Börzel 1998; Dowding 1995;
LeGalès and Thatcher 1995; Richardson 1999).
2
On Australia see Considine 1994; Davis et al. 1993; on Canada see Coleman and Skogstad
1990; Lindquist 1996; on the UK see Rhodes 1988, Richardson and Jordan 1979; on continental
Europe see LeGalès and Thatcher 1995; Marin and Mayntz 1991; on the USA see Mandell 2001;
O’Toole 1997.
A policy community has the following characteristics: a limited number of

participants with some groups consciously excluded; frequent and high-quality
interaction between all members of the community on all matters related to
the policy issues; consistency in values, membership, and policy outcomes that
persist over time; consensus, with the ideology, values, and broad policy
preferences shared by all participants; and exchange relationships based on
all members of the policy community controlling some resources. Thus, the
basic interaction is one involving bargaining between members with resources.
There is a balance of power, not necessarily one in which all members equally
benefit but one in which all members see themselves as in a positive-sum
game. The structures of the participating groups are hierarchical so leaders can
guarantee compliant members. This model is an ideal type; no policy area is
likely to conform exactly to it.
One can only fully understand the characteristics of a policy community if
we compare it with an issue network (see Table 3.1). McFarland (1987: 146),
following Heclo’s (1978) use, defines an issue network as ‘a communications
network of those interested in policy in some area, including government
authorities, legislators, businessmen, lobbyists, and even academics and jour-
nalists . . . [that] . . . constantly communicates criticisms of policy and generates
ideas for new policy initiatives’. So, issue networks are characterized by: many
participants; fluctuating interaction and access for the various members; the
absence of consensus and the presence of conflict; interaction based on
consultation rather than negotiation or bargaining; an unequal power rela-
tionship in which many participants may have few resources, little access, and
no alternative. The study of interest groups understood variously as issue
networks, policy subsystems, and advocacy coalitions is probably the largest
American contribution to the study of policy networks. They are seen as an
ever-present feature of American politics (and for surveys of the literature see
Baumgarten and Leech 1998; Berry 1997).
Obviously the implication of using a continuum is that any network can be
located at some point along it. Networks can vary along several dimensions and
any combination of these dimensions; for example, membership, integration,
resources. Various authors have constructed continua, typologies, and lists of the
characteristics of policy networks and policy communities (see, for example, Van
Waarden 1992). This lepidopteran approach to policy networks—collecting and
classifying the several species—has become a dead end.
Networks as Interorganizational Analysis

The European literature on networks focused less on subgovernments and
more on interorganizational analysis (see, for example, Rhodes (1999a [1981])).
It emphasizes the structural relationship between political institutions as the
crucial element in a policy network rather than the interpersonal relations
between individuals in those institutions. At its simplest, interorganizational

analysis suggests that a ‘focal organisation attempts to manage its dependen-
cies by employing one or more strategies, other organisations in the network
are similarly engaged’. A network is ‘complex and dynamic: there are multiple,
over-lapping relationships, each one of which is to a greater or lesser degree
dependent on the state of others’ (Elkin 1975: 175–6).3
The most impressive attempt to apply this variant of network analysis to
politics and policy-making is the several collaborations of David Knoke,
Edward Laumann, and Franz Pappi (see especially Knoke 1990; Knoke et al.
1996; Laumann and Knoke 1987). Their ‘organizational state’ approach argues
that ‘modern state–society relationships have increasingly become blurred,
merging into a mélange of inter-organizational influences and power rela-
tions’. These interorganizational networks ‘enable us to describe and analyse
interactions among all significant policy actors, from legislative parties and
government ministries to business associations, labour unions, professional
societies, and public interest groups’ (Knoke et al. 1996: 3). The key actors are
formal organizations, not individuals. In their analysis of national labour
policy in America, Germany, and Japan, Knoke et al. 1996 compiled the
list of key actors by, for example, searching public documents such as the
Congressional Information Service volumes for the number of times they
testified before the relevant Congressional or Senate Committee, including
only organizations with five or more appearances. The individuals in these
organizations responsible for governmental policy affairs were then inter-
viewed on such matters as the informant’s perception of the most influential
organization, the communication of policy information, and participation in
the policy area. Knoke et al. then use the techniques of network analysis to
map the links between organizations, employing classic network measures
such as centrality and density (for an introduction to such techniques, see
Scott 2012 and, for a compendium, see Wasserman and Faust 1994).
Knoke et al. argue that their data not only describes the power structure of
their chosen policy area but also explains the different policy outcomes. The
value of this species of network analysis lies in its use of the structural properties
of networks to explain behaviour and outcomes. Unfortunately, little work in
this idiom is explanatory. Instead, it describes power structures and network
characteristics. Moreover, ‘it has not yet produced a great deal that is novel’
(Dowding 2001: 89–90 and n. 2). It is hard to demur from this judgement when
Knoke et al. (1996: 210, 213) conclude that ‘the state clearly constitutes the
formal locus of collective decision-making that affects the larger civil society
within which it is embedded’, or that ‘the more central an organisation was in
3
See also Benson 1975; Crozier and Thoenig 1976; Hanf and Scharpf 1978; and Thompson
1967.
either the communication or the support network, the higher was its reputation
for being influential’ (see also Thatcher 1998: 398–404).
Networks as Governance
The roots of policy network analysis lie, finally, in the analysis of the sharing of
power between public and private actors, most commonly between business,
trade unions, and the government in economic policy-making (Atkinson and
Coleman 1989; Jordan 1981). Initially, the emphasis fell on corporatism, a
topic worthy of an article in its own right (see Cawson 1986a; Schmitter and
Lehmbruch 1979). There was also the long-standing and distinctive Scandi-
navian analysis of ‘corporate pluralism’ (Rokkan 1966; Heisler 1979), which
continues under such labels as ‘the segmented state’ (Olsen 1983: 118) and ‘the
negotiated economy’ (Nielsen and Pedersen 1988). Latterly, the main concern
has been with governance by (and through) networks, on trends in the
relationship between state and civil society government rather than policy-
making in specific arenas. Thus, ‘governance’ is a broader term than ‘govern-
ment’ with public resources and services provided by any permutation of
government and the private and voluntary sectors (and on the different
conceptions of governance see Kjær 2004; Pierre 2000).
There are several accounts of this trend for Britain, continental Europe, and
the USA. Thus, for Britain, there has been a shift from government by a
unitary state to governance by and through networks. In this period, the
boundary between state and civil society changed. It can be understood as a
shift from hierarchies, or the bureaucracies of the welfare state, through the
marketization reforms of the Conservative governments of Thatcher and
Major to networks to the emphasis on partnerships and joined-up government
by New Labour.4
There is also a large European literature on ‘guidance’, ‘steering’, and ‘indirect
coordination’, which predates both the British interest in network governance
and the American interest in reinventing government. For example, Kaufman et
al.’s (1986) edited volume on guidance, steering, and control is truly Germanic
in size, scope, and language. It focuses on the question of how a multiplicity of
interdependent actors can be coordinated in the long chains of actions typical of
complex societies (see also Bovens 1990; Luhmann 1982; Van Gunsteren 1976).5
For the USA, Osborne and Gaebler (1992: 20 and 34) distinguish between
policy decisions (steering) and service delivery (rowing), arguing bureaucracy
is a bankrupt tool for rowing. In its place they propose entrepreneurial
4
See, for example, Ansell 2000; Bevir and Rhodes 2003; Rhodes 1997a, 2000c; Stoker 2004;
and for a review of the literature and citations see Marinetto 2003.
5
Since I wrote this chapter there have been useful reviews of the literature by Börzel 2011;
Klijn 2008; and Klijn and Koppenjan 2015: ch. 2.
government, with its stress on working with the private sector and respon-
siveness to customers. This transformation of the public sector involves ‘less
government’ or less rowing but ‘more governance’ or more steering. In his
review of the American literature, Frederickson (1996: 84–5) concludes the
word ‘governance is probably the best and most generally accepted metaphor
for describing the patterns of interaction of multiple-organisational systems or
networks’ (see also Kettl 1993: 206–7; Salamon 2002). Peters (1996: ch. 1)
argues the traditional hierarchic model of government is everywhere under
challenge. He identifies four trends, or models of governance, challenging the
hierarchic model—market, participative, flexible, and deregulated governance.
Fragmentation, networks, flexibility, and responsiveness are characteristics of
flexible governance. In sum, talk of the governance transformation abounds
even if the scope, pace, direction, and reasons for that change are matters of
dispute (and for a survey see Pierre 2000).
I am not suggesting there is any convergence. There is no two-way street
(see Rhodes 2011b). American scholars brought their characteristic
modernist-empiricist skill set to bear on networks and governance. They
combined ‘large N’ studies of networks (Meier and O’Toole 2005) with an
instrumental or tool view that sought to make the study of networks relevant
to public managers (Agranoff 2007). Their European counterparts preferred
comparative case studies, although there was a shared focus on network
management and the allied subjects of partnerships and collaboration (see
Volume I, Chapter 4).
Policy Networks as Theory
There is a large theoretical literature on policy networks in Britain (see Rhodes

1988, 1997a, 1999a), the rest of Europe (see Börzel 1998; Kickert et al. 1997),
and the USA (see O’Toole 1997; Salamon 2002). There are two broad schools
of thought, depending on how they seek to explain network behaviour: power-
dependence or rational actor.6
Power-dependence
The power-dependence approach treats policy networks as sets of
resource-dependent organizations. Their relationships are characterized
6
Bob Goodin pointed out correctly that theories of complexity are also relevant to the study
of network (personal correspondence). See, for example, La Porte 1975; Luhmann 1982; Simon
1981[1969]. Such ideas exercised some influence on the ‘governance club’ research programme
at Erasmus University, Rotterdam (see, for example, Kickert et al. 1997). Complexity theories
have not been a major influence on the rest of the network literature.
by power-dependence; that is, ‘any organisation is dependent on other orga-

nisations for resources’, and ‘to achieve their goals, the organisations have to
exchange resources’. So, actors ‘employ strategies within known rules of the
game to regulate the process of exchange’. Relationships are a ‘game’ in which
organizations manoeuvre for advantage. Each deploys its resources, whether
constitutional-legal, organizational, financial, political, or informational, to
maximize influence over outcomes while trying to avoid becoming dependent
on the other ‘players’. So, behaviour in policy networks is game-like, rooted in
trust and regulated by rules of the game negotiated and agreed by network
participants. Variations in the distribution of resources and in the bargaining
skills of participants explain both differences in outcomes in a network and
variations between networks. Finally, the networks have a significant degree of
autonomy from government (Rhodes 1997a: ch. 2; 1999a [1981]: ch. 5).7
Rational Choice
The rational choice school explains how policy networks work by combining
rational choice and the new institutionalism to produce actor-centred institution-
alism. The best example is the Max-Planck-Institut’s notion of ‘actor-centred
institutionalism’. For Renate Mayntz, Fritz Scharpf and their colleagues at
the Max-Planck-Institut, policy networks represent a significant change in the
structure of government. They are specific ‘structural arrangements’ that deal
typically with ‘policy problems’. They are a ‘relatively stable set of mainly public
and private corporate actors’. The links between network actors serve as ‘com-
munication channels and for the exchange of information, expertise, trust and
other policy resources’. Policy networks have their own ‘integrative logic’ and the
dominant decision rules stress bargaining and sounding out. So, as with the
power-dependence approach, the Max Planck school stresses functional differen-
tiation, the linkages between organizations, and dependence on resources (Kenis
and Schneider, 1991: 41–3).
Scharpf (1997: chs 2 and 3) combines rational choice and the new institu-
tionalism to explain how policy networks work, an approach referred to as
actor-centred institutionalism. The basic argument is that institutions are
systems of rules that structure the opportunities for actors (individual and
corporate) to realize their preferences. So, ‘policy is the outcome of the
interactions of resourceful and boundedly-rational actors whose capabilities,
preferences, and perceptions are largely, but not completely, shaped by the
institutionalised norms within which they interact’ (Scharpf 1997: 195).
Networks are one institutional setting in which public and private
actors interact. They are informal institutions; that is, informally organized,
7
The analysis of ‘power-dependence’ is not limited to the study of networks. More generally
see Blau 1964; Emerson 1962; Keohane and Nye 1977, 1987; and Pfeffer and Salancik 1978.
permanent, rule-governed relationships. The agreed rules build trust and

foster communication while also reducing uncertainty; they are the basis of
non-hierarchic coordination. Scharpf uses game theory to analyse and explain
these rule-governed interactions.
In the UK, there have been vigorous exchanges between the two schools
(see, for example, Dowding 1995, 2001 versus Marsh 1998: 12–13, 67–70;
Marsh and Smith 2000). It is a case of ‘ne’er the twain shall meet’. The two
sides have irreconcilable differences of both theory and method. The disagree-
ments are as basic as the deductive, positivistic, quantitative approach of
economics versus the inductive, interpretive, qualitative approach of soci-
ology. For insiders, harmony is not threatening to break out any time soon.
To outsiders, the debate seems like a spat. The outsiders could well be right.
Policy Networks as Reform
The spread of networks and the recognition that they constrain government’s
ability to act has fuelled research on how to manage networks. The goal is now
‘joined-up government’ or a ‘whole-of-government’ approach. Networks are
no longer a metaphor or a site for arcane theoretical disputes but a live issue
for reforming public sector management. Here I concentrate on the public
sector literature.8
Kickert et al. (1997: 46) identify three approaches to network management
in the public sector: the instrumental, interactive, and institutional (insert
Table 3.2).
The instrumental approach focuses on how governments seek to exercise
legitimate authority by altering dependency relationships. The key problem
with the instrumental approach is the cost of steering. A central command
operating code, no matter how well disguised, runs the ever-present risks of
recalcitrance from key actors, a loss of flexibility in dealing with localized
problems, and control deficits.
The interaction approach stresses management by negotiation instead of
hierarchy. The trick is to sit where the other person is sitting to understand
their objectives and to build and keep trust between actors. So, chief executive
officers in the public sector must have ‘strong interpersonal, communication
and listening skills; an ability to persuade; a readiness to trade and to engage in
reciprocal rather than manipulative behaviour; an ability to construct long-
term relationships’ (Ferlie and Pettigrew 1996: 88–9). The key problem of
the interactive approach is the costs of cooperation. Network management is
8
On the private sector see Child and Faulkner 1998: ch. 6; Ford et al. 2003; and Pfeffer and
Salancik 1978.
Table 3.2 Approaches to network management.
Instrumental Interactive approach Institutional approach
approach
Focus Improving steering Cooperation Network arrangements

conditions and their impacts
Level of Focal organization Interactions of actors Network structure
analysis and its set (individuals and
organizations)
View of policy Closed and multi- Horizontal Product and context of
networks form object of interaction interaction and
steering governance
Characteristics Strategic steering Game playing to Incremental adaptation of
of network develop cooperation incentive structures, rules
management and prevent and culture of networks
blockages
Criteria of Effective problem- Satisficing policy, Institutionalized key
evaluation solving consensus interests and relationships
Source: Modified from Kickert, Klijn, and Koppenjan 1997: 186. See also Klijn and Koppenjan 2015. The table
was not in the original version of the chapter.
time-consuming, objectives can be blurred, and outcomes can be indefinite.

Decision-making is satisficing, not maximizing.
The institutional approach focuses on the institutional backcloth, the
rules and structures, against which the interactions take place. The aim is
incremental changes in incentives, rules, and culture to promote joint
problem-solving. The institutional approach has one major, even insur-
mountable, problem; incentives, rules, and culture are notoriously resistant
to change because networks privilege a few actors, who equate their sec-
tional interest with the public interest. They are well placed to protect their
sectional interests.
The literature specifically on managing networks grows apace in both
America and Europe. Salamon (2002) provides a comprehensive review of
the tools available for America’s new governance, covering the ‘classic’ instru-
ments such as grants, regulation, and bureaucracy, but laying great emphasis
on the collaborative nature of modern governing and the need to switch from
hierarchy and control to enabling and the indirect management of networks.
Collaborative governance and leadership are emerging areas of research
(Ansell and Gash 2008; Huxham and Vangen 2005).
What do you do if you have to run a network? Painter et al. (1997: 238)
provide specific advice on game management. They conclude that local author-
ities should: conduct an audit of other relevant agencies; draw a strategic map of
key relationships; identify which of their resources will help them to influence
these other agencies; and identify the constraints on that influence. As with all
new trends, there is an upsurge of advice from both academics and consultants.
So, the ten commandments of networking include: be representative of your
agency and network, take a share of the administrative burden, accommodate
and adjust while maintaining purpose, be as creative as possible, be patient and
use interpersonal skills and emphasize incentives (Agranoff 2003: 29). It is
certainly not ‘rocket science’ (Perri 6 et al. 2002: 130) and this list of lessons
gives credence to that claim. Wettenhall (2003: 80) reviews the literature on
partnerships, joined-up government, and the new governance. He concludes
that these terms have ‘become the dominant slogan in the turn-of-the-century
discourse about government’ (see, for example, Cabinet Office 2000; Cm 4310
1999; MAC 2004). So, any disapproving reader dismissing this literature should
pause to note that it is well on the way to becoming the new conventional
wisdom in public sector reform. Those of more caustic disposition, having
paused, might move on by noting that network management is an ephemeral
mix of proverbs and injunctions.9 Others hedge their bets and provide their mix
of injunctions (see Chapter 4, this volume)!
DEBATES AND CHALLENGES
Paralleling the earlier discussion, this section looks at the debates and chal-
lenges that confront policy network analysis. In turn, I examine some descrip-
tive, theoretical, and prescriptive pitfalls (see Chapter 12, this volume).
Describing and Comparing Networks
The notion of a policy network can be dismissed as mere metaphor. It is not a

metaphor because there is no analogy. It is commonplace to describe policy-
making as a set of interconnected events and communicating people. It is no
more a metaphorical term than bureaucracy. The term’s resonance and
longevity stem from the simple fact that for many it represents an enduring
characteristic of much policy-making in advanced industrial democracies.
In his review of British studies of pressure groups and parties, Richardson
(1999: 199) claimed that Dowding’s (1995) critique of policy networks marked
the ‘intellectual fatigue’ of the approach. The sheer number and variety of
articles published since this ‘watershed’, including Richardson’s (2000) own
prize-winning paper on networks and policy change, testifies to the continuing
9
The literature may be preoccupied with adducing lessons for would-be managers but it
also analyses such network management ‘tools’ as, for example, brokerage. See Bardach 1998;
Carpenter et al. 2004; Fernandez and Gould 1994; and Taylor 1997.
utility of the term. Not only are there innumerable case studies of British
policy networks but, casting the net wider, beyond the confines of political
science policy networks are staples in, for example, criminology (Loader 2000;
Ryan et al. 2001). The international relations literature on networks expanded,
with Haas’s (1992) notion of epistemic communities influential. They are
transnational networks of knowledge-based experts with an authoritative
claim to policy relevant knowledge within their domain of expertise. The
distinguishing features of these networks are their shared beliefs and profes-
sional judgements. Directly analogous to Haas’s network of experts are Keck
and Sikkink’s (1998) transnational advocacy networks of activist. For example,
the UN, domestic non-governmental organizations (NGOs), international
NGOs, and private foundations form an international issue network to coun-
ter the ‘forgetfulness’ of governments. The network is an alternative channel of
communication that argues, persuades, lobbies, and complains to inject new
ideas and information into the international debate on human rights (see also
Risse et al. 1999; Sikkink 1993).
Transnational networks are also a feature of policy-making in the EU. For
Peterson (2009: 119 and 129), ‘policy network analysis is never more powerful
as an analytical tool than when it is deployed at the EU level’ and ‘few . . . would
deny that governance by networks is an essential feature of the EU’.10 Policy
network analysis has also colonized intergovernmental relations in and
between states, most notably federal–state relations (Galligan 1995; Rhodes
1988; Wright 1978; and Chapter 4, this volume).
Finally, there is governance in a globalizing world. It comes in several
varieties. Keohane’s (2002: 204, 210–12, 214) version of global governance is
one of ‘networked minimalism’. In other words, there is no hierarchy but a
network of nation states, private firms, NGOs, and subunits of government,
which pursues ‘minimal rather than ambitious objectives’. The nation state
will remain the ‘primary instrument of domestic and global governance’ but ‘it
is not the only important actor’ (see also Slaughter 2003). Rosenau (2000:
172–3) provides a more dramatic vision of a ‘multi-centric’ world composed of
diverse transnational collectivities that both compete and cooperate and do
not lend themselves to hierarchic control or hegemonic coordination. The
world is a network and networks are the world.
In short, I doubt there could be a clearer example of ‘have theory will travel’
and, therefore, there is a problem. There is no synthesis of the findings of this
diverse literature. Indeed, a synthesis may not be possible. The key question
would be: ‘what type of network emerges in what conditions with what policy
outcomes?’ There have been many willing to tell us how to answer this
question (Dowding 1995; Thatcher 1998). Only a few brave souls have tried
to give an answer, and even then they confine their analysis to comparing
10
See also Ansell 2000, Anderson 1990; Josselin, 1997; Kassim, 1993; Mazey and Richardson,
1993; and Rhodes et al., 1996.
either several policy sectors in a single country or a single policy sector in

several countries (see, for example, Considine 2002; Marsh 1998).
When seeking to compare policy networks across countries, the problems
are probably insurmountable. Policy networks are but political science writ
small. The problems that bedevil comparative government also plague policy
networks. They were devastatingly summarized by MacIntyre (1972: 8):
There was once a man who aspired to be the author of the general theory of holes.
When asked ‘What kind of hole—holes dug by children in the sand for amuse-
ment, holes dug by gardeners to plant lettuce seedlings, tank traps, holes made by
road makers’ he would reply indignantly that he wished for a general theory that
would explain all of these. He rejected an initio the—as he saw it—pathetically
common-sense view that of the digging of different kinds of holes there are quite
different kinds of explanations to be given.
Such ‘modernist–empiricism’ (Bevir 2001: 478) treats policy networks as
discrete objects to be measured, classified, and compared. It may not be one
of ‘the more dangerous kinds of practical joke’ (MacIntyre 1972: 26) but it is
only one way of studying networks.
The story about the rise and rise of governance raises a second issue. This
‘new orthodoxy’ does not carry all before it. Marinetto (2003) disputes the
‘Anglo-Governance School’s’ claim that there has been a loss of central
control. He suggests that it exaggerates the ruptures in history, arguing there
has been a long-standing tension between centralization (government) and
fragmentation (governance) in Britain. In a similar vein, Holliday (2000)
insists Britain still has a strong core executive, the centre has not been
hollowed-out, networks have not spread, and the centre can and does exercise
effective control. Whether the Anglo-Governance School has ‘to undergo an
intellectual crisis wrought by the growing weight of criticism’ and the extent to
which this ‘critical response is underway, albeit gradually’ will become clear
over the next few years (Marinetto 2003: 605). I too expect to see ‘alternative
ways of conceptualising the institutions, actors and processes of change in
government’, to listen to a new generation of stories about governance, and to
ponder another round of debate about whether changes are epiphenomena of
present-day government policy or more deep-seated ruptures (see Chapters 7
and 12, this volume). Stick around long enough and the aphorism ‘what goes
around comes around’ sounds like a balanced summary of fads and fashions in
the social sciences rather than irony or even cynicism.
Explaining Change
The most common and recurrent criticism of policy network analysis is that
it does not, and cannot, explain change (for a summary of the argument and
citations, see Richardson 2000). So, policy network analysis stresses how
networks limit participation in the policy process; decide which issues will be
included and excluded from the policy agenda; shape the behaviour of actors
through the rules of the game; privilege certain interests; and substitute private
government for public accountability. Policy network analysis is about stabil-
ity, privilege, and continuity.
There have been several attempts to analyse change and networks but
I must make two preliminary points. First, it is no mean feat to describe and
explain continuity and stability in policy-making. Second, the analysis of
change may be a recurring problem but, and this point is crucial, it is not
specific to the study of networks. Just as there are many theories of bureau-
cracy, so there are many theories of policy networks. There is no consensus in
the political science community about how to explain, for example, political
change, only competing epistemological positions and a multitude of theories.
Students of policy networks can no more produce an accepted explanatory
theory of change than (say) students of bureaucracy, democracy or economic
development. Debates in the policy network literature mirror the larger
epistemological and ontological debates in the social sciences.
Of the several efforts to build the analysis of change into policy networks,
three have attracted attention: advocacy coalitions, the dialectical model, and
decentred analysis.
The advocacy coalition framework (ACF) has four basic premises. First, it
assumes that ‘understanding the process of policy change . . . requires a time
perspective of a decade or more’. Second, ‘the most useful way to think about
policy change . . . is through a focus on “policy subsystems”.’ Third, ‘those
subsystems must include an intergovernmental dimension’. Finally, ‘public
policies . . . can be conceptualised in the same manner as belief systems, that is,
sets of value priorities and causal assumptions about how to realise them’
(Sabatier and Jenkins-Smith 1993: 16). Sabatier argues that coalitions try to
translate their beliefs into public policy. Their belief systems determine the
direction of policy. Their resources determine their capacity to change govern-
ment programmes. Resources change over time, most commonly in response to
changes external to the subsystem. Most distinctively, Sabatier distinguishes
between core and secondary beliefs and argues that coalitions have a consensus
on their policy core that is resistant to change. In sharp contrast, secondary
aspects of the belief system can change rapidly (paraphrased from Sabatier and
Jenkins-Smith 1993: 25–34). Moreover, these beliefs are central to understand-
ing the actions of policy-makers who are not necessarily motivated by rational
self-interest. However, as Parsons (1995: 201) succinctly points out, the model
works well for the federal and fragmented government of America but there is
little evidence that it travels well.
The dialectical model proposed by Marsh and Smith (2000) suggests that
change is a function of the interaction between the structure of the network
and the agents operating in it, the network and the context in which it
operates, and the network and policy outcomes. They see networks as struc-
tures that can constrain or facilitate action, but they do not determine actions
because actors interpret and negotiate constraints. Exogenous factors may
prompt network change but actors mediate that change. So, we must examine
not only the context of change but also structure, rules, and interpersonal
relationship in the network. Finally, not only do networks affect policy
outcomes but policy outcomes feedback and affect networks. This dialectical
model provoked heated debate and lectures on how to do political science, but
little convergence and a mere tad of insight (compare Marsh and Smith 2000,
2001 with Dowding 2001).
Hay and Richards (2000) grapple with such issues as the formation, evolu-
tion, transformation, and termination of policy networks using a ‘strategic
relational theory of networks’; it is a sophisticated variation on the dialectical
theme. To begin with, they avoid the ambiguities of, and controversies sur-
rounding, the term ‘dialectical’. They argue that individuals seeking to realize
certain objectives and outcomes make a strategic assessment of the context in
which they find themselves. However, that context is not neutral. It too is
strategically selective in the sense that it privileges certain strategies over
others. Individuals learn from their actions and adjust their strategies. The
context is changed by their actions, so individuals have to adjust to a different
context. So, a networking is ‘a practice—an accomplishment on the part
of strategic actors . . . which takes place within a strategic (and strategically
selective context) which is itself constantly evolving through the consequences
(both intended and unintended) of strategic action’ (Hay and Richards 2000:
14; see also Hay 2002).
A different challenge comes from those who advocate an interpretive turn
and argue that policy network analysis could make greater use of such
ethnographic tools as: studying individual behaviour in everyday contexts;
gathering data from many sources; adopting an ‘unstructured’ approach;
focusing on one group or locale; and, in analysing the data, stressing the
‘interpretation of the meanings and functions of human action’ (paraphrased
from Hammersley 1990: 1–2). The task would be to write thick descriptions or
our ‘constructions of other people’s constructions of what they are up to’
(Geertz 1973: 9, 20–1; and for a similar recognition that the political ethnog-
raphy of networks is an instructive approach see: Heclo and Wildavsky 1974;
McPherson and Raab 1988; and Rhodes 2017, Volume II, Chapter 3).
Bevir and Rhodes (2003; and Chapter 6, this volume) argue for the
decentred study of networks, for a shift of topos from institution to individual,
and a focus on the social construction of policy networks through the ability of
individuals to create meaning. Bang and Sørensen’s (1999) story of the
‘Everyday Maker’ provides an instructive example of a decentred account of
networks. They interviewed 25 active citizens in the Nørrebro district of
Copenhagen to see how they engaged with government. They identify the
‘Everyday Maker’, who focuses on immediate and concrete policy problems at

the lowest possible level. Thus, Grethe (a grass-roots activist) reflects that she
has acquired the competence to act out various roles: contractor, board
member, and leader. There has been an explosion of ‘issue networks, policy
communities, ad hoc policy projects, and user boards, including actors from
“within”, “without”, “above”, and “below” traditional institutions of demo-
cratic government’. So, the task of the ‘Everyday Maker’ is ‘to produce concrete
outcomes’ (Bang and Sørensen 1999: 332). Political activity has shifted from
‘formal organising to more informal networking’ (Bang and Sørensen 1999:
334). Politics is no longer about left and right but ‘dealing with concrete
problems in the institutions around which . . . everyday life . . . is organised’
(Bang and Sørensen 1999: 336). In short, they draw a picture of Nørrebro’s
networks through the eyes of its political activists, constructing the networks
from the bottom up (and see Chapter 6 for a more detailed summary and
assessment).
This discussion highlights two points. First, the trend in the study of policy
networks to ethnographic methods mirrors general trends in political science.
Fenno (1990: 128) observed ‘not enough political scientists are presently
engaged in observation’. That was then. Now there is a growing interest in
the interpretive turn and ethnography in political science (see Rhodes 2017,
Volume II, Chapters 2 and 3). However, it is worth noting that the origins of
network analysis lie in social anthropology, which examines who talks to
whom about what in (say) a Norwegian village. So, this point is perhaps best
expressed as an overdue return to roots.
Second, all three approaches to network change are part of a broader trend
in political science to exploring the impact of ideas on policy-making. Again, it
would take us too far afield to cover this topic but Sabatier’s (1993) work on
advocacy coalitions stands alongside that of, for example, Kingdon (1984) on
policy ideas and policy agendas. The link between changing policy networks,
new ideas, and setting policy agendas is exploited to great effect in Richardson
(2000).
Managing the Institutional Void
If we live in a world of ‘polycentric networks of governance’, then the task

facing politicians, managers, and citizens is to manage ‘the institutional void’;
that is, to make and implement policy when there are no generally accepted
rules and norms for conducting policy-making (Hajer 2003: 175). Hajer’s vivid
metaphor may overstate the extent of change but it does dramatize the
problems of managing the network state. Four such problems recur: the mix
of governing structures, the diffusion of accountability, enhancing coordin-
ation, and devising new tools (and see Chapter 4, this volume).
It’s the Mix that Matters

In a world of policy networks where every service is a mix of bureaucracy,
markets, and networks, we need to understand when these governing struc-
tures for allocating resources work. We need to be clear about what we mean
when we call for effective service delivery because the criteria of effectiveness
vary. For example, the competition that characterizes markets conflicts with
the cooperation so characteristic of networks. Flynn et al. (1996: 136–7) argue
that trust became important in the British National Health Service because of
the difficulties in specifying contracts and participants’ experience of assertive
purchasers whose style ‘engenders or exacerbates suspicious attitudes and
feelings of mutual distrust’. So, market relations had ‘corrosive effects’ on
‘professional networks which depend on cooperation, reciprocity and inter-
dependence’. I would belabour the obvious if I gave examples of bureaucratic
failures. The apt conclusion is not that contracts or bureaucracies or networks
fail, but that they all do (Jessop 2000). Not every day or every week or for every
policy. The key is to understand the conditions under which each works and a
core lesson of that analysis is, ‘it is the mix that matters’ (see Chapter 11, this
volume). We not only need to know how to manage each governing structure
but also the relationship between them.11
The Problem of Many Hands

Conventional notions of accountability do not fit when authority for service
delivery is dispersed among several agencies. Bovens (1998: 46) identifies the
‘problem of many hands’ where responsibility for policy in complex organ-
izations is shared and it is correspondingly difficult to find out who is
responsible (see also van Gunsteren 1976: 3). For example, Hogwood, Judge,
and McVicar (2000) show that agencies and special purpose bodies have
multiple constituencies, each of which seeks to hold them to account. There
is no system, just disparate, overlapping demands. In a network, the constitu-
ent organizations may hold the relevant officials and politicians to account but
to whom is the set of organizations accountable? As Mulgan (2003: 211–14)
argues, buck-passing is much more likely in networks because responsibility is
divided and the reach of political leaders is much reduced. However, all is not
doom and gloom. Following Braithwaite (2003: 312) policy networks can be
seen as an example of ‘many unclear separation of powers’ in that the several
interests in a network can act as checks and balances on one another. How-
ever, it is more common for networks to be closed to public scrutiny, a species
11
See, for example, Considine and Lewis 1999; Thompson et al. 1991; Powell 1991; Rhodes
1997b; and Simon 2000.
of private government. The brute fact is that multiple accountabilities weaken

central control (Mulgan 2003: 225).12
The Holy Grail of Coordination

Weakened accountability is not the only consequence of networks. The spread
of networks also undermines coordination. Despite strong pressures for more
coordination, the practice is ‘modest’. It is ‘largely negative, based on persist-
ent compartmentalisation, mutual avoidance and friction reduction between
powerful bureaus or ministries’; ‘anchored at the lower levels of the state
machine and organised by specific established networks’; ‘rarely strategic, so
almost all attempts to create proactive strategic capacity for long-term plan-
ning . . . have failed’; and intermittent and selective in any one sector, impro-
vised late in the policy process, politicized, issue-oriented, and reactive
(Wright and Hayward 2000: 33). And that is before we introduce networks
into the equation. Networks make the goal ever more elusive. As Peters
(1998: 302) argues, ‘strong vertical linkages between social groups and public
organisations make effective coordination and horizontal linkages within
government more difficult’. Once agreement is reached in the network, ‘the
latitude for negotiation by public organisations at the top of the network is
limited’. However, these remarks presume hierarchy is the most important
or appropriate mechanism for coordination. Lindblom (1965) persuasively
argued many years ago that indirect coordination or mutual adjustment was
messy but effective. The San Francisco Bay Area public transit system is a
multi-organizational system (or network) and Chisholm (1989: 195) shows
that only some coordination can take place by central direction and so
‘personal trust developed through informal relationships acts as a lubricant
for mutual adjustment’. In sum, coordination is the holy grail of modern
government, ever sought, but always just beyond reach and networks bring
central coordination no nearer. However, they do provide their own messy,
informal, decentralized version.
Steering Not Rowing

The mainstream literature (for example, Salamon 2002) encourages a tool box
view of how to manage networks. The popular aphorism is steering not rowing
and refers to the skills of indirect management. Hands-off steering refers to
working with and through networks or webs of organizations to achieve shared
policy objectives. It involves continuously negotiating beliefs and exchanging
resources within agreed rules of the game (see also Torfing et al. 2012: 14;
12
On the need to rethink accountability in the nation-state see Behn 2001; on accountability
in a globalizing world see Keohane 2002: 219–44 and 2003.
Koliba et al. 2011: 60). The specific skills are said to include integrating agendas;
developing clear roles, expectations, and responsibilities for all players; agreeing
the criteria of success; and sharing the administrative burden.
Mastering these new tools is not the only challenge confronting would-be
network managers. The epistemological debate extends to the question of how
to manage networks. An interpretive approach encourages us to replace the
toolbox approach with storytelling. Although the label varies—the argumen-
tative turn, narratives—there is now a growing literature on storytelling as a
way of managing the public sector.13 It is not an example of academic whimsy.
It is an integral part of the everyday practices of public servants, indeed all
managers. Managers use stories not only to gain and pass on information and
to inspire involvement but also as the repository of the organization’s institu-
tional memory. Management is just as much about narratives and interpret-
ation as rational calculation (see pp. 217–20; and Gabriel 2000; Hummel 1991;
Rhodes 2011c).
CONCLUSIONS
In the 1970s, debate raged about the future of public policy-making and policy
analysis. Was it a distinctive field of study or just good old public administra-
tion under a new and fashionable label? It staked a claim to be a distinct
field of study. Now we no longer discuss the question. Policy analysis is
established. In this sense, there is no longer a debate about the future of policy
networks. The story of policy networks follows the same trajectory as public
policy-making. The subject is here to stay—a standard topic in any public policy-
making textbook (Parsons 1995) or textbooks on British government (Richards
and Smith 2002).
What was all the excitement about? It is not just the story of the rise of an idea.
It is about a new generation of political scientists. ‘Young—well youngish—
Turks’ carved out a reputation for themselves by challenging their elders and
betters. Sound and fury are essential to such uprisings. In Britain, added edge
came from the challenge to the Westminster model, which had run out of steam
as a way of understanding the changes in British government. The debate was
not only about networks but also about how to study British government. It
should be no surprise, therefore, that the recurrent problems of the policy
network literature, for example in explaining change, mirror issues in broader
political science. The rise of governance was our story of how British govern-
ment had changed. It was not the story in the graduate and postgraduate texts on
13
See Bevir 2011; Gabriel 2000; Hummel 1991; Rein 1976; Van Eeten et al. 1996; and Weick
1995.
which we were raised. We abandoned the eternal verities of the British consti-
tution. In sharp contrast to the fuddy-duddies, we could explain both continuity
and change. Of course, we were wrong but we weren’t about to admit it. Anyway
the spats were fun!
The story of policy networks is a story of a success. The ‘Young Turks’
won their elevation to the professorial peerage, and moved on. A flood of
doctorates and case studies followed. It is no longer an innovative idea but a
commonplace notion in almost every nook and cranny of both political
science texts, and British government textbooks in particular (see, for example,
Diamond 2014; Dorey 2005; Richards and Smith 2002). Controversies in
policy network analysis now parallel controversies in political science, whether
they are about how to explain political change or the uses of ethnographic
methods.14 Policy network analysis has become one more locus for the endless
debates about how we know what we know in the social sciences (and I return
to these debates in Chapter 12, this volume). I doubt the founders could have
hoped for more. I am sure their expectations were less because:
All people, things, institutions and environments that are innovative and avant-
garde at one historical moment will become backward and obsolescent in the
next. . . . All individuals, groups and communities are under relentless pressure
to reconstruct themselves; if they stop to rest, to be what they are, they will be
swept away (Berman 1983: 78).
Below, I return to several of the themes in this chapter where I discuss: man-
aging networks (Chapter 5); ethnographic studies of networks (Chapter 6); and
the interpretive turn (Chapter 7).
14
Of course, we also respond to debates and problems in the ‘real’ world. Much of the
literature reviewed in this chapter sees networks as an effective way of managing complex
problems in health and education. However, Al Qaeda and the war on terror have focused
attention on ‘dark networks’ (Raab and Milward 2003), a term that also encompasses drug
smuggling, the arms trade and failed states. Fieldwork may not be an option but the problems of
policing dark networks cannot be ignored.
4
Policy Networks and Policy-making

in the European Union
I N T R O D U C TI O N
It is commonplace to treat policy-making in the European Union (EU) as

distinctive. The EU is neither a federation nor an intergovernmental body,
although it has features of both. Policy-making varies and does not conform to
a single pattern or model (Nugent 1994: 297; 2010: 289). However, this
chapter adopts a different approach because it does not start with the differ-
ences between the EU and the member states.1 It focuses on similarities and
asks whether models of policy-making known to be useful in analysing
member states are also useful for analysing EU policy-making.2 It assumes
that all governmental units confront similar imperatives and respond in
similar ways. In particular, it examines critically the literature on policy
networks and asks whether this idea is useful for understanding EU policy-
making; does it travel well?
The first section of the chapter briefly reviews the relevant literature.3 The
second section looks at the ways in which the idea helps to explain EU policy-
making. The third section examines criticisms of the network approach.
Finally, I present a summary evaluation of the policy network approach,
identifying the conditions under which it is useful.
1
An updated version of R. A. W. Rhodes, I. Bache, and S. George (1996) ‘Policy Networks
and Policy Making in the European Union: A Critical Appraisal’. In L. Hooghe (ed.), Cohesion
Policy and European Integration. Oxford: Clarendon Press, pp. 367–87. I have removed the
introduction to the concept of policy networks (see Chapter 3, this volume). Reprinted by
permission of Oxford University Press.
2
As this question applies to all member states, it matters not whether Britain is a member of
the EU. Britain’s exit means it will lose its right to membership of all networks by right. Now
Britain will have to negotiate access sector by sector.
3
In the original chapter, this section summarized mainly the other chapters in Hooghe
(1996). The summary was necessary for that book but it is less relevant here so I have abbreviated
it and updated the review of the literature in the Afterword.
POLICY NETWORKS AND E U P OLICY-MAKING

IN THE 1 990 S
In the 1980s and early 1990s, several authors used ‘policy networks’ to explore
EU policy-making, often tentatively.4 Helen Wallace (1984: 141) talks of
‘the emergence of horizontal policy networks which cut across national
boundaries’. Simon Bulmer (1994: 14) claims that subsystem government is
‘well established in integration studies’ ever since the publication of Wallace,
Wallace, and Webb (1977). Dudley Coates’s (1984: 158) account of the foods
standards regulatory policy similarly concludes that implementation is ‘heav-
ily dependent on a rather small policy community’ (see also Lewis and Wallace
1984). He notes that ‘the complexity of the institutional arrangements does
not seem to inhibit the process’. Rhodes (1986c) examines the relationship
between the EU and local governments, points to the triadic links between
central, local, and supranational governments, and suggests there are ‘emer-
gent’ policy networks in some policy areas. These early studies have four
features in common. First, they focus on policy implementation and not policy
initiation. Second, they stress the large differences between policy areas. Third,
they talk of incipient or emergent networks, not settled policy communities.
Finally, they all recognize the need—in a complex, intergovernmental, policy-
making system—to aggregate and coordinate the many affected public and
private interests. Networks do not recognize institutional boundaries. Policy
emerges from the struggle between government and non-governmental
organizations (see Atkinson and Coleman 1992; Rhodes 1995b).5
The 1990s saw more interest in ‘policy networks’, mainly to describe EU
policy-making. Thus, Peters (1992: 77) provides a broad account of EU policy-
making, arguing that it is ‘best understood as bureaucratic politics’ and such
decision-making takes place in ‘policy communities’. The EU executive is
fragmented and ministers of functional departments are ‘involved in games
over particular policy interests’ (Peters 1992: 79). EU policy-making is both
differentiated and specialized and ‘many policy communities or networks
appear to exert great influence, if not control, over public policy, more than
in most national governments in Europe’ (Peters 1992: 81). Three interlocking
games recur in EU policy-making: the national game in which each member
state tries to maximize its return from EU membership; the institutional game
in which the institutions ‘seek to gain more power relative to others’; and the
4
The interest in networks is not confined to students of public administration and public
policy. Haas’s (1992: 3) notion of ‘epistemic communities’ or ‘network of professionals’ is
equivalent to Rhodes (1988: 78) on professionalized networks. See also Keohane and Hoffman
(1991).
5
There were still relatively few accounts of policy networks in the EU by 1996. In addition to
the work cited in the text, see Grant, Paterson, and Whitson 1988; Josselin 1994; Preston 1984;
Scharpf 1988; Smith 1990.
Policy Networks and Policy-making in the EU 59
bureaucratic game in which the Directorate-Generals of the Commission

(DGs) ‘compete for policy space’ (Peters 1992: 106–7). For Peters, differenti-
ation, specialization, interdependence, and bureaucratic politics characterize
EU policy-making. The problem is that his account is as brief as it is broad; for
example, he does not explain the links between the policy communities and his
three games.
Mazey and Richardson (1993: 4) are both more focused and more cautious.
They explore interest groups in EU policy-making and suggest that, although
‘some interests have managed to become part of a cohesive policy community
at the EU level’, the majority ‘are involved in less integrated types of policy
networks’. There is ‘no dominant model or style of EU-group relations’ and
the ‘procedural ambition’ of the Commission to set up ‘stable and regularised
relationship with the affected interests’ (Mazey and Richardson 1993: 9)
remains just that, an ambition.6 They conclude:
at the EU level there are indeed quite significant variations in the nature of policy
networks (and . . . in some specific policy areas no network may exist) but there is
at least a case to be made that the network concept is quite useful
(Mazey and Richardson 1993: 253).
Marks (1992, 1993, 1996) focuses on EU structural policy, arguing that since
1988 the European Regional Development Fund (ERDF) has been ‘creating
policy networks that encompass sub-national governments and private interests
in individual regions’ (Marks 1992: 192). In effect, the reforms are an exercise
in ‘institution building that strengthens the Commission’ and attempts to
‘technocratize—and in a narrow sense depoliticize—a key growing area’.
Also, the reforms create direct links between the funds and regional political
institutions, and so challenge ‘centralised decision making within member
states’ (Marks 1992: 212). The several layers of government are ‘enmeshed in
territorially overarching policy networks’ (Marks 1993: 392 and 402). Marks
(1993: 407) generalizes his conclusions about structural policy to cover inter-
governmental relations in Europe.7 He claims:
we are witnessing the emergence of multi-level governance in the European
Community, characterised by co-decision-making across several nested tiers
of government, ill-defined and shifting spheres of competence (creating a
6
The Commission’s ambitions became even more difficult to realize because of the increasing
number of new interest groups in several policy areas; for example, ERDF policy in the UK in
the 1980s.
7
The phrase intergovernmental relations (IGR) has several uses; for example, it can refer to
the links between the national governments of member states of the EU. This use is too
restrictive. In the study of federalism, it refers to interactions between governmental units of
all types and levels (see Rhodes 1981: 76). It is so used here and includes EU governmental units.
I see no difference between IGR so defined and MLG. The quotation from Marks (1993: 407) in
the text applies equally to American federalism and the EU.
consequential potential for conflicts about competencies), and an ongoing search
for principles of decisional distribution that might be applied to this emerging
polity.
Even in this initial formulation, there were two problems with Marks’s analysis
(and see the Afterword at the end of this chapter). First, although the links
between levels of government multiply, they are not necessarily an effective
challenge to centralized decision-making. Marks’s evidence most commonly
refers to the participation of local and regional governments in decision-
making and not to their effect on the outcomes of decision-making. Yet,
there is great variation between member states in the effectiveness of subna-
tional participation.
Second, Marks deliberately avoids the theory-laden notion of networks. The
phrase ‘multi-level governance’—from now on MLG—describes the changing
structure of EU government. It refers to the policy-making system as a whole
and is used to draw attention to the common feature that many functional
policy networks involve the participation of several levels of government. It is
not used to explain either variations in that structure or why it has changed.
Marks (1996: 399 and n. 10) eschews ‘a theory-impregnated conceptualisation’
of ‘network’, using the term to refer to ‘a more or less stable set of political
relationships among actors’. Networks are central to his analysis of MLG
because he compares the ‘diverse array of networks across individual member
states’. Instead of looking to resource-dependence to explain this diversity he
looks to variations in the national policy style and the different functional
characteristics of each stage of decision-making to explain differences in local
and regional participation.8
In the 1990s, there was an extensive literature on the links between local
and regional government and the EU.9 Peters (1992: 112) suggested the EU
was developing ‘“picket-fence federalism” and intergovernmental relations
similar to those in the United States’. The top-down conception of intergov-
ernmental relations, in which central government proposes and local govern-
ment disposes, gave way to multi-level negotiations. The intergovernmental
game was one of several games in EU policy-making.10 Such games are not
8
This summary draws on Marks (1992, 1993, 1996). On his more recent work see the
Afterword at the end of this chapter.
9
See, for example, Batley and Stoker 1991; Keating and Jones 1985; Leonardi 1992; Mitchell
1994; Rhodes 1986c; Sharpe 1993.
10
Putnam’s (1988) influential article on domestic–international linkages characterizes them
as a two-person game in which domestic groups lobby the national government to adopt
favourable policies and the government tries to satisfy these groups in international negotiations
without incurring adverse foreign policy developments. However, this model assumes the
primacy of the national governments. It ignores the role of such supranational bureaucracies
as the Commission. It ignores linkages between interests and subnational governments which
bypass the national government. It ignores transnational interest groups. In short, it is a two-
level model which does not deal with multi-level policy-making.
limited to implementation, but can be found at all stages of policy-making.

Bargaining and negotiation lie at the heart of policy networks that seek to
integrate not only interest groups and government but also the several types
and levels of governmental units that are the focus of MLG.
As a final example of policy networks and the intergovernmental game,
Hooghe’s (1996) collection of essays on cohesion policy uses the network
metaphor to describe variously the links between levels of government, cross-
border regional cooperation, and transnational regional lobbies. Thus, Nanetti
(1996: 86) concludes the Community Support Framework (CSF) planning
process meant ‘the networks of territorial and institutional relations became
more complex . . . more articulated and differentiated’; and saw the ‘steady
emergence of the regional level as the new institutional partner of the Com-
mission’. Rynck’s (1996: 158) account of Flemish regional policy-making talks
of ‘multi-level governance processes’ and ‘EU cohesion policy networks’.
Balme and Jouve (1996: 251) argue that policy networks are ‘crucial’ for
understanding the linkages between state institutions and between the state
and its environment They argue that the EU affects domestic policy networks
by changing the distribution of resources. They conclude that, although the
central state ‘clearly dominated all networks’ and ‘European public policy has
helped to consolidate existing networks rather than to replace them or super-
impose new networks on them’, nonetheless, the EU ‘appears as a driving
force for the regionalisation of the state’ or ‘a regional polity grounded in the
expansion of inter-organisational policy making’. Laffan (1996: 337 and 340)
describes how EU regional policy brought local actors into the CSF policy
network and enhanced their role in policy-making.
Such descriptions of EU policy-making are essential but it is more import-
ant to explain variations in that policy process. ‘Policy networks’ are used to
analyse EU policy-making in specific policy areas, most notably by Jeffrey
Anderson and John Peterson.
Anderson (1990) explicitly uses Rhodes’s (1988) notion of networks as
clusters of public and private agencies exchanging resources. He focuses on
domestic policy networks arguing:
The EU, which ‘commands resources, distributes benefits, allocates markets, and
adjudicates between conflicting interests’ (Wallace 1982: 61), can alter domestic
networks by affecting the inter-member distribution of resources, and thus their
level of interdependence. Community initiatives can increase, decrease or leave
unchanged the resource dependencies of network members (Anderson 1990: 422).
He compares British and German responses to the reforms of the ERDF since
1979, concluding that the ERDF reforms ‘exposed resource dependencies
previously of little consequence’, especially ‘the lack of administrative capacity
to develop joint projects’ (Anderson 1990: 442). So, British subnational actors
became more dependent on central government because the shift of emphasis
to programmes instead of projects encouraged the regional offices of the

Department of the Environment to play a proactive or ‘mothership’ role in
mobilizing local authorities. The reforms reinforced the civil service’s position
as a ‘gatekeeper’.
The ERDF reforms had a disruptive effect on German federal government,
‘exacerbating existing conflicts within the national joint policy-making frame-
work’ (Anderson 1990: 443). Anderson argues there was a decrease in resource-
dependency because the Bund could use the Commission as a scapegoat for
checks on Länder regional programmes. The reforms of the ERDF since 1979
did not uniformly strengthen the regions at the expense of the centre because
domestic policy networks have ‘different levels of resource interdependence’
that ‘vary across otherwise similar policy areas’ and so the EU has varying effects
(Anderson 1990: 445). Anderson (1996: 188–9) updates this analysis of Ger-
many and the structural fund. He concludes there was a bifurcation of
the territorial network. For the West, the EU did ‘influence the content and
structure’—the resource-dependencies—of its territorial policy network,
although the Länder ‘remained tied to the federal policy network’ and ‘were not
especially successful’ in fending off the Commission’s interventions. By contrast,
the East benefited from EU resources, which gave ‘the new Länder . . . the means
to carve out a sphere of independence, albeit constrained, from the Federal
government’.
Peterson (1992: 244) uses the policy network approach to explore European
technology policy and concludes there was ‘a tightly integrated policy com-
munity’. Peterson (1994) employs the Rhodes model of policy networks
because it ‘captures’ three key features of EU decision-making: ‘bargaining
between multiple public institutions and private interests; sectoral variations
in the policy process; and the resource dependencies which fundamentally
shape relationships between different actors’ (Peterson 1994: 21). His emphasis
on ‘meso-level decision making’ is an important qualification to the scope of the
model’s application. He argues the concept is of limited use in analysing either
the history-making decisions made at intergovernmental conferences or the
major policy-setting decisions taken by the Council of Ministers. It is most
useful in analysing ‘“second-order” decisions that address the question: how do
we do it?’ (for example, technology policy). Again, the focus on coordination
and implementation is paramount.
The distinctions between history-making, policy-setting, and policy-
shaping decisions pose problems, as does the emphasis on implementation.
Rather like Hoffman’s distinction between high and low politics (Hoffman
1966), Peterson’s distinction has a common-sense appeal but is difficult to
apply consistently because his criteria are broad. For example, history-making
decisions cover revisions to the treaties, strategic decisions about agenda,
priorities, and finance and legal decisions handed down by the European
Court of Justice. The distinction between strategic decisions and policy-setting
decisions is difficult, resurrecting the means and ends distinction with all its
known ambiguities and problems. Also, although policy networks are central
to implementation, they can play an important role in policy initiation. They
can shape the policy agenda by excluding options of which they disapprove.
The distinction between policy-making and its implementation is a useful
analytical tool but separating the two can be difficult. It is a commonplace of
policy analysis that the details of policy can decide outcomes and the ‘bottom-
up’ approach to implementation has shown that street-level bureaucrats can
reshape policy (see, for example, Sabatier 1986). We can recognize that net-
works are important in implementation without excluding their contribution at
other stages in policy-making.
In sum, the idea of policy networks has several virtues as a tool for
describing and analysing EU policy-making. It has the vocabulary and tech-
niques for describing complex organizational linkages. The distribution of
resources in a network explains the relative power of its members. Differences
in the pattern and distribution of resources explain the differences between
networks. Finally, the idea of networks and their management underpin the
Commission’s strategy for managing MLG. However, although the policy
network approach has some distinct advantages, it is not without its critics.
A CRITICAL APPRAISAL
There are several general critiques of the ‘theory’ of ‘policy networks’, which
I discuss in Chapter 12 (this volume).11 Here, I focus on criticisms of its
usefulness in analysing EU policy-making. I consider these criticisms under
five headings: explanation, level of analysis, institutions, boundaries, and policy.
Explanation
Bennington and Harvey (1994: 954; see also Dowding 1994: 62) claim that
policy networks do not offer any explanatory insights. They make this inaccur-
ate assessment because they focus on the typology of networks and ignore
the power-dependence model that underpins that typology. Networks differ
because they have different patterns of resource-dependence. The section on
‘Policy Networks and EU Policy-making in the 1990s’ (pp. 58–63) provides
many examples of this point.
11
See, for example, Dowding 1994; Marsh and Rhodes 1992a; Mills and Saward 1994; Rhodes
1986b; Rhodes and Marsh 1994; Smith 1993. Many of these general criticisms recur in the
specific context of EU policy-making.
Level of Analysis
Bennington and Harvey (1994: 957) claim the Rhodes model has ‘an inad-
equate conception of the state’ and there is ‘an under-theorisation of the inter-
relationship between different levels’ (see also Mills and Saward 1994). In a
similar vein, Kassim (1993: 22) argues the usefulness of the policy networks
approach depends on ‘the availability of a macro theory’ and ‘where such a
theory is absent, the approach is of limited value’. There is no ‘authoritative or
fully articulated’ macro-theory of the EU, so the policy network approach is of
limited value.
Marsh and Rhodes (1992a: 266–8) also argue that networks must be located
within a broader theory of the state. Rhodes (1988: 48–77) provides that
context for Britain. Peterson (1995a: 14–15) identifies intergovernmentalism
and neo-functionalism as ‘plausible macro-theories of EU politics’ and
Peterson (1994) sketches his own approach to linking networks and a broader
analysis of the EU context. Alternatively, Peters’s bureaucratic politics model
could provide the basic building blocks for such a theory because it identifies
the conditions under which policy networks emerge in the EU. Thus, the
games prevent the Commission from exerting unified leadership. The result-
ing bureaucratic differentiation, with each DG a distinct organization with its
own goals and links to the national bureaucracies, encourages the develop-
ment of vertical policy networks. I do not try to develop a theory of EU policy-
making here. I simply point out there is no shortage of contenders.
The key tasks are not only to develop a macro-theory of EU policy-making
but, as important, to delimit the circumstances under which the policy net-
work approach applies. Little is achieved by dismissing the approach with
the tautological assertion that middle-range theory is not comprehensive
(for fuller discussion, see Rhodes and Marsh 1994: 14–17).
Institutions
Kassim (1993: 8 and 11) claims the policy network approach can deal with
neither the fluid and fragmented nature of EU institutions nor its institutional
complexity and density. He insists that institutions are especially important in
the EU. Similarly, Bulmer (1994) emphasizes the institutional distinctiveness
of the EU. In reply, Peterson (1995a: 9) accepts the criticism that ‘EU institu-
tions often pursue their own agendas within policy networks’. However, the
policy network approach is a modern variant of the institutional approach to
politics (Rhodes 1995a: 26) that focuses on ‘behaviour within institutional
contexts’ (Gamble 1990: 417). Rhodes (1986a: 20 and 409) clearly shows
how central departments use policy networks to advance their own interests.
The relationship between central and local government is described repeatedly

as ‘asymmetric’. In short, institutional complexity and competing institutional
agendas are the stuff of policy networks. The EU is different only in degree
from complex, domestic institutional environments. Peterson (1995b: 18)
makes the point well:
Kassim reveals his misreading of the Rhodes model when he complains that the
‘analogue of the government department or agency at the national level’ is
difficult to locate at the EU level. The explanatory power of the policy networks
model does not depend on the existence of a single public agency which mono-
polises the policy agenda. As long as an institution has an interest in a policy
sector, the resources to effect outcomes, and a need for other resources (which it
does not possess) to pursue its policy objectives, it occupies a place within a policy
network [emphasis in original].
Boundaries
Kassim (1993: 24) claims ‘the possibility of delimiting networks’ is ‘highly

problematic’ (see also Bennington and Harvey 1994: 956). Marsh and Rhodes
(1992a) recognize the problem (see Stones 1992: 224) as do Rhodes et al.
(1982: 32 and 49). It is a long-standing debate in network analysis in all its
forms (see, for example, Whitten and Wolfe 1972). Critics overstate the
problem. Networks are sets of resource-dependent organizations. The extent
of the resource exchanges between those organizations will fix the boundaries.
There are proven techniques for measuring such exchanges (see, for example,
Evan 1976b, Part 4; Josselin 1994). Defining the boundaries of organizational
networks is arbitrary but the problem is not disabling (see, for example,
Laumann and Pappi 1976).
Policies
Kassim (1993: 20) describes the EU policy process as fluid and claims that
‘differences between policy issues may be more significant than any similar-
ities at the sectoral or sub-sectoral level’. Bennington and Harvey (1994: 955)
argue the model leads to:
a preoccupation with the institutional level of analysis and prescription, where,
for example, a concern with procedural arrangements is elevated above substan-
tive issues and outcomes.
However, the policy network approach can identify likenesses and differences
between policy areas. Several studies show that networks exist at the subsec-
toral levels in EU policy-making. For example, Grant et al. (1988) show that
the European chemical industry’s policy community can be broken down into
several subsectoral networks, such as pharmaceuticals, fertilizers, agro-chemicals,
the paint industry, and the soap detergent and toilet preparations industry.
The policy networks approach identifies both the common bonds of the policy
community and the subsectoral differences within that community. Whether
these likenesses and differences are more or less significant than differences
between policy issues is an empirical question, which will not be resolved by
assertion.
It is inaccurate to say the policy networks approach downplays substantive
issues and outcomes. Marsh and Rhodes (1992a: 262–4) discuss policy out-
comes and directly address the issue of which interests are dominant and
benefit. There is nothing in the policy network approach that precludes the
analysis of policy outcomes (and, for examples, see Rhodes 1988: 387–406; and
Marsh and Rhodes 1992b). Finally, the importance of policy networks varies
with the stage of the policy-making. Thus, Marsh and Rhodes (1992b: 185–6)
stress its relevance for analysing policy implementation. Even in the states
with a long tradition of centralized decision-making such as Britain, France,
Greece, and Ireland, local and regional authorities play an important role in
implementing cohesion policies.
Although the policy network approach stands accused of stretching a good
idea too far, this fault lies with the critics, and not the proponents, of the
approach. Too many critics assume the policy networks approach seeks to
explain national or EU policy-making. Its scope is more restricted. ‘Policy
network’ is a useful tool for analysing the links between types of governmental
units, levels of government, and between governments and interest groups
(especially the professions). It aids understanding of policy-making but it is
only one variable in that process. In other words, the critics fail to recognize
that their criticisms are in fact statements of the conditions under which the
policy network approach does not work. I discuss the conditions under which
networks do not work below.
THE L IMITS TO P OLICY NETWORKS
The term ‘policy network’ is used in at least four ways:

• As a metaphor, covering any policy that emerges from the interactions of
several actors or institutions.
• To refer to personal links between decision-makers.
• To refer to the links between public organizations and between public and
private organizations needed to implement policy.
• To refer to a set of resource-dependent organizations.
I avoid the first use because it is too imprecise. I avoid the second use because it is
too narrow. I focus on networks as a set of resource-dependent organizations.
Implementation structures are simply a specific case of resource-dependence.
This ‘theory impregnated’ concept of policy networks has several advantages.
It fits the analysis of MLG where there is a high degree of resource-
dependence between the several affected public and private institutions. It
links the analysis of domestic and supranational politics (Bulmer 1983). Most
important, the distribution of resources explains the relative power potential
of actors within a network and the differences between networks. It is a
middle-range or meso-level theory that helps us to understand policy-making
at the national or supranational levels by comparing variations between policy
sectors. It focuses on who gets what, when, where, and how. It is less successful
in explaining the constitutive or history-making decisions of the EU (Peterson
1994: 7; Bulmer 1994: 6). However, even here, it can help to explain, for
example, the ways in which the policy agenda is shaped. Also, it is only one
governing structure, directly equivalent to the notion of ‘subsystem govern-
ance’ based on policy programmes (Bulmer 1994: 14). There are several other
types of such structures; for example, bureaucratic regulation, market compe-
tition, solidarity, association (Rhodes 1995b; Wright 1991). Finally, it is
possible to identify the factors that sustain policy networks. I discussed several
of these factors above and, therefore, present them as a list with only brief
explanatory comments. I also identify several policy areas that seem poten-
tially fruitful for testing the concept of policy networks.
Factors Sustaining EU Policy Networks
National Style of Policy-making

As Marks (1996: 408) points out, ‘knowing where (i.e. in which country) policy
is made’ is more important than ‘knowing the stage at which policy is made’
(emphasis in original). European policy networks are more likely to occur
when policy networks are an established feature of national systems. However,
policy networks are not useful tools for analysing all political systems. The
approach assumes a degree of pluralism, the relative separation of public and
private actors, and complex policies needing many resources that are not
concentrated in the state. For example, if resources are concentrated on a
strong national gatekeeper, policy networks are less likely to emerge and,
where they do exist, will be less important for explaining policy outcomes.
Degree of Resource-dependence
Resource-dependence will be high when the policy sector is characterized
by the institutional fragmentation of MLG, when the policy is complex, and
when the Commission needs information and expertise. The Commission’s

dependence on other actors can be high because it is small with limited
expertise; it is an ‘adolescent bureaucracy’ (Mazey and Richardson 1993: 10).
However, it is important to stress that networks are characterized by inter-
dependence. So, the Commission controls resources—most notably, authority
and money—that interest groups, local and regional governments, and
national governments need.
Characteristics of the Policy Area

The important characteristics are the extent to which: the Commission and its
DGs play the lead role in constructing the networks; the policy is ‘low politics’;
policy-making is routinized; and the policies are not politicized. Routinization
is more likely to occur where there is agreement on both core policy values (for
example, the distribution of authority between levels of government);
and secondary values (for example, instrumental decisions about rules and
resources), although agreement on core values is not a precondition of an
effective network.
Stage of Policy-making
Policy networks are more likely to occur when the Commission depends on
other actors for implementing its policy, although policy networks can play an
important role at any stage in the policy process including policy initiation (for
a more detailed discussion, see Marks 1996).
Aggregation
Policy networks always involve aggregating interests. The Commission needs
to aggregate interests as a strategy to counter institutional fragmentation and
coordinate policies. The Commission’s transparency package announced in
February 1994 seeks to develop a code of conduct for pressure groups, to limit
the number of consultative exercises, and to restrict the number of groups
consulted (for a more detailed summary and discussion, see Peterson 1995b).
In short, it seeks to aggregate interests and simplify a complex lobbying
process. The Committee of the Regions is another example of the same
ambition ‘to recreate 1970s-style corporatism’ (Peterson 1995b).
Functional Representation
Representing economic, professional, and subnational interests through net-
works is a key source not just of information and advice but also of legitim-
ation for the Commission. Where the Commission needs to legitimate its
activities in, for example, a new policy area, it has an incentive to create a
policy network to provide support in dealings with the Council of Ministers
and members states.
A F T E R W O RD
The original version of this chapter concluded that the concept of policy
networks should be part of the toolkit of any political scientist.12 It identified
the conditions under which networks emerged. It is a truism to describe
policy-making in the EU as varied, but there are stable patterns of policy-
making. There are identifiable governance structures. The literature shows
that policy networks are one form of sectoral governance in the EU. The idea
combines accurate description with an explanation of the differences between
networks. It is, therefore, an indispensable part of any political scientist’s
toolkit. How has this broad assessment stood up over the years? In what
ways can the study of networks continue to contribute to the study of EU
policy-making?
Assessment
At first, I felt certain that a chapter published in 1996 would be well past its
sell-by date.13 Indeed, I did not include it in my original selection until one
of my referees urged me to do so. To my great surprise, and pleasure, I was
wrong. The concept of policy networks is not outmoded. To the contrary,
there is the danger of belabouring the obvious in suggesting that policy
networks have become part of the conceptual vocabulary of studies of EU
policy-making. They are one of the several forms of interest representation in
the EU and explain ‘slices’ of EU policy-making. Policy networks are a fixed
part of EU studies.14
Peterson (2009) provides the authoritative survey of policy networks and
the study of EU policy-making. He opines that ‘the Rhodes model of policy
12
I have deleted the original section on future research and replaced it with an overall
assessment of the usefulness of policy networks in understanding EU policy-making; and a
discussion of new directions in the 2010s.
13
For the obvious reason that they were the co-authors of the original chapter or former
colleagues, I draw on the work of Ian Bache, Stephen George, and John Peterson in defending
this position.
14
See, for example, Bache 2008: 31–3; Bache et al. 2014: 29–34; Kenealy et al. 2015: 16 and
passim; Peterson and Bomberg 1999: 27–8 and passim; Piattoni 2010: 20; Rosamond 2000: 126;
Wallace et al. 2014: 36–7 and passim.
networks has probably been employed more often than any other in the study
of EU governance’. He suggests that ‘policy network analysis is never more
powerful an analytical tool than when it is deployed at the EU level’. It provides
‘a language to describe and perhaps sometimes to explain’ EU governance. It
captures the variety of EU policy-making. He concludes that ‘the EU governs
largely by policy networks’ (Peterson 2009: 108, 109, 113, and 120).
The EU lends itself to policy network analysis because it is an ‘extraordinary
differentiated polity’ characterized by ‘discrete distinctive and largely discon-
nected’ policy networks underpinned by ‘an extraordinary complex labyrinth
of committees that shape policy options’ (Peterson 2009: 106; see also
Jachtenfuchs 2001: 253–5; Kohler-Koch and Eising 1999). It explains subsys-
temic policy-making in the EU. It is also compatible with both intergovern-
mental and neo-functional accounts of the EU, neither of which are theories of
policy-making. Also, it is compatible with the several approaches to EU
governance such as epistemic communities (Haas 1992), network governance
(Jordan and Schout 2006) and MLG (Hooghe and Marks 2003). The analysis
of policy networks is central to answering the question of ‘how do things work
around here?’
New Directions
Policy network analysis is also a fertile source of new avenues of exploration.

Briefly, I discuss three such avenues: MLG, network management, and
constructivism.
Multi-level Governance
MLG has become a settled feature of the EU literature (see, for example, Bache
2008; Bache and Flinders 2004, 2015; Jeffery 2015; Piattoni 2010). The core
‘territory’ for MLG is the EU’s cohesion policy and its links between state
actors at local, regional, national, and EU levels; that is, multi-level govern-
ment, not governance (Hooghe 1996; Hooghe and Marks 1996; Marks 1993).
As the approach spread beyond its initial boundaries, Type I MLG was
supplemented with Type II MLG in which jurisdictions and membership
overlap and non-state actors play a role; that is, governance and horizontal
coordination (Hooghe and Marks 2003). Bache and Flinders (2004: 197)
identify four areas of agreement across the multiplying contributions to the
topic: the growing role of non-state actors; emerging complex networks
rather than nested territorial levels of government; the changing role of the
state in steering and coordinating networks; and the challenges to democratic
accountability posed by the foregoing trends. These trends bring the analysis
of MLG close to the analysis of network governance. As Bache (2008: 32) notes,
the analytical potential of linking policy networks, network governance, and

MLG ‘remains undeveloped’:
Not only are there similarities in terms of language and metaphor but also both
approaches share a concern with detailed empirical investigations of multiple
interactions within policy process, particularly sector specific, from policy initi-
ation through to and including policy implementation.
So, building on Peterson and Bomberg (1999), Rhodes (1988, 1997a), and
Bevir and Rhodes (2003), Bache (2008: 31–6) builds a ‘conceptual bridge’
between the two approaches. The bridge leads to a focus on: the exchange of
resources between actors and organizations; disaggregating the state, especially
its core executive; and the beliefs, preferences, and practices of actors. It is an
agenda remarkably close to that discussed in Chapters 6 and 7, this volume.
Network Management
I noted earlier that the Commission is seen as an ‘adolescent bureaucracy’.
Such a description might capture the differences from classic Weberian
bureaucracy but it sheds little light on the Commission’s role in networked
governance. As Kassim et al. (2013: ch. 3) describe, the Commission must rely
on policy networks or hands-off steering rather than its own bureaucracy or
hands-on steering to succeed. Traditionally, its characteristics numbered
strong centrifugal forces, Cabinets that are national enclaves, and baronial
DGs at war with one another (Kassim et al. 2013: 181). Latterly, there is more
effective interdepartmental coordination and a more interventionist centre,
but problems remain. As one of Kassim et al.’s (2013: 179) interviewees noted,
‘Power [in the Commission] is never with somebody for a very long time. It
shifts all the time, and as soon as you see that it is somewhere visible too long,
it’s rebalanced somewhere else.’ So, it fluctuates.
The various DGs continue to have a relatively free hand in building
networks. Each has a nodal position spanning the multitude of networks in
its sector. Each has the opportunity to coordinate and steer, if not direct, its
networks. In other words, the hierarchic, ‘command and control’ view of
bureaucracy is relevant to only some policy sectors. In other sectors, the
Commission has moved from steering to rowing; that is, to indirect or
hands-off measures. This version of the Commission’s statecraft is elaborated
by Torfing et al. (2012: 156–9; and ch. 7). They suggest the central agency—in
this case, the Commission’s DGs—must ‘balance autonomy of networks with
hands-on intervention’. To do so, they can ‘campaign for a policy, deploy
policy narratives, act as boundary spanners, and form alliance with politi-
cians’. They become ‘metagovernors’ managing the mix of bureaucracy, markets,
and networks (and see Chapters 5 and 11, this volume).
Belying the multiplying neologisms, this description is not just academic

theory, but also Commission practice. For example, Katsaris (2014) docu-
ments how the Commission created policy networks with the Moroccan
government to build a technical consensus over climate change mitigation
and the use of renewable energy. More sceptical, Jordan and Schout (2006:
272) document the struggles of the Commission to build networks that foster
environmental protection integration (EPI) and identify the many factors that
inhibit network management. To the (growing) list noted already (see pp. 66–9;
and Chapter 5, this volume), they add:
• The absence of a network manager
• The speed of the EU policy-making
• The distinct, separate loyalties of national and EU public servants
• The large number of actors and their distinct preferences
• Reactive coordination
• Lack of skills in managing cross-sectoral challenges.
Many of these conditions refer to effective coordination generally rather than
network management specifically. Nonetheless, they conclude the Commis-
sion lacks capacity for network management (see also Kassim et al. 2013: 100)
and that EPI network is a ‘policy mess’ (citing Rhodes 1985b: 11). However,
the Commission’s lack of capacity to manage networks does not alter its
conviction that it:
can better achieve its objectives with networks than without them and therefore
the Commission deliberately and strategically creates them
(Sbragia 2000: 235; see also Cram 2009).
And, for my purposes, this discussion shows that the study of policy networks
is of continuing relevance in the study of EU policy-making. The common
ground is obvious; network management and metagovernance are central foci
in the analysis of network governance (see Chapter 11, this volume).
Constructivism
The study of the EU turned to interpretive approaches in the 2000s under the
label of constructivism. The main foci are the process of Europeanization and
the discursive construction of an EU polity (see, for example, Checkel 2007;
Christiansen et al. 2001; Diez 2015; Risse 2009; Schimmelfenning 2012;
Wodak 2011). They follow the lead of anthropology (see, for example,
Abélès 2004; Shore 2000). The study of policy networks trod the same path
under the label of interpretivism. As Peterson (2009: 110) observes, there are
‘affinities’ between policy network analysis and constructivism. Indeed, there
are many connections. However, in this chapter, I need do no more than point
out that this is a promising new direction because I develop the interpretive
approach to policy networks in both Chapters 6 and 7 in this volume. Also,

Rhodes (2017b, Volume II) is devoted to interpretivism in political science.
For now, I am content to claim that the notion of policy networks is a
permanent fixture in the conceptual vocabulary of studies of EU policy-
making, and it continues to open new directions of study.
5
How to Manage Your Policy Network
INTRODUCTION
Chapter 3 surveyed the massive literature on policy networks, and its continuing
debates. In passing, I noted there was a growing literature on how to manage
your network, although much is not accessible to a general reader (see, for
example, Agranoff 2007; Goldsmith and Eggers 2004; Goldsmith and Kettl
2009; Kickert, Klijn, and Koppenjan 1997; Klijn and Koppenjan 2015). This
chapter returns to the topic of managing networks and seeks to answer four
questions. What are the key characteristics of policy networks? When do policy
networks succeed? When do policy networks fail? How do central agencies
manage networks?
I have been delivering some version of this chapter as a lecture for practi-
tioners for over a decade without ever getting round to publishing it.1 Indeed,
on the one occasion when I bothered to submit it to a practitioner-oriented
journal, it was rejected because I was recycling old ideas—I was, deliberately—
and it provided no new theory or empirical data—I was not, deliberately. So, to
stress what should be obvious, this chapter does not add to the already over-
lengthy list of theoretical articles, nor is it yet another case study of a network
‘in action’. Rather, it seeks to translate and distil the existing literature into
useful lessons for practitioners.
WHAT ARE THE KEY CHARACTERISTICS

OF POLICY NETWORKS?
Policy networks are the sets of formal institutional and informal linkages
between governmental and other societal actors structured around shared, if
1
This version of the paper was presented to a workshop for the Commonwealth Secretariat,
15 February 2013 and was published in Mandarin as ‘罗茨著 R. A. W, ‘王宇颖译. 如何管理政
策网络？’[J]. 中国行政管理, (11`) 2015: 139–44 (Chinese Public Administration) after a short
lecture tour of China.
How to Manage Your Policy Networks 75
Table 5.1 Characteristics of networks.
Bureaucracy Markets Networks
Basis of relationships Employment relationship Contract Resource exchange

Degree of dependence Dependent Independent Interdependent
Medium of exchange Authority Prices Trust
Means of conflict resolution Rules and commands Haggling and Diplomacy
and coordination the courts
Culture Subordination Competition Reciprocity
Source: Adapted from Powell 1991. See Rhodes 1998.
endlessly negotiated, beliefs and interests in public policy-making and imple-

mentation (see Chapter 3, this volume). Such networks can and do develop from
the bottom up. They are also initiated by central agencies. Both operate in the
shadow of hierarchy because they depend on central agencies for resources. Yet
networks have distinctive characteristics of their own and Table 5.1 provides a
brief comparison of the three main ways of delivering services.
Networks are a distinctive coordinating mechanism and, therefore, a sep-
arate governing structure from markets and hierarchies, because trust is their
central coordinating mechanism in the same way that commands and price
competition are the key mechanisms for bureaucracies and markets respect-
ively. Trust is essential for cooperative behaviour and, therefore, the existence
of the network. It is the defining attribute. Networks are also characterized by
diplomacy and reciprocity. These characteristics shape the task of managing
networks. I discuss each in turn.
Trust
At the heart of networks and management by diplomacy is the notion of trust;

it is ‘the most important attribute of network operations’, the central coord-
inating mechanism (see Kramer and Tyler 1996). Shared values and norms are
the glue that holds the complex set of relationships together; trust is essential
for cooperative behaviour and, therefore, the existence of the network. More-
over, trust is non-calculative. As Powell (1996: 63) points out, trust is ‘neither
chosen nor embedded but is instead learned and reinforced, hence a product
of on-going interaction and discussion’. Preserving trust is, therefore, a recip-
rocal and endless task. Fox’s (1974: 362) conclusions about trust in industrial
relations are equally apt for networks. Thus, in networks with high trust
relationships, participants share values, recognize their long-term obligation
to one another, offer support, communicate openly, and give one another the
benefit of the doubt.
Reciprocity
Networks involve friendship, loyalty, even altruism (Thompson 1993: 54–8)

but above all network culture is characterized by reciprocity. As Powell (1991:
272–3) comments, reciprocity is rooted in ‘the normative standards that
sustain exchange’, especially indebtedness, obligation, and a long-term per-
spective. So, a lack of equivalence creates a moral sanction, bonds that keep the
parties in touch with one another; the books are balanced only in the long
term. However, as Thompson (1993: 58) points out, reciprocity is also a
symbolic relationship and ‘in the constant ritual of exchange, deep obligations
and duties are established, symbolic statuses confirmed, metaphorical social
references invoked’. In this way, network coordination becomes stabilized.
Diplomacy
Diplomacy refers to management by negotiation. The diplomat must persuade

‘another government to accept and perhaps actually help to promote the
policies which it is the ambassador’s function to advocate’. The main technique
is ‘the maintenance by continual persuasion of order in the midst of change’
(Watson 1982: 125 and 223). Nicholson (1950: 15) defines diplomacy as ‘the
management of international affairs by negotiation’. For all its slightly quaint
air Nicholson signals an important shift in style to a language that stresses
sitting where the other person sits and helping other people to realize their
objectives. We relearn old lessons (de Callierès 1963 [1716]). The emphasis lies
not in imposing one’s objectives on another but on finding out about the other.
Diplomacy may be an old-fashioned word but the arts of negotiation and
persuasion are not confined to it. As Sir Douglas Wass, former Head of the
British Civil Service, said, ‘finesse and diplomacy are an essential ingredient in
public service’ (Hennessy 1989: 150). Such skills lie at the heart of steering
networks. Public servants have been managing networks for years but either
chose not to talk about it or, like Molière’s Monsieur Jourdin, did not know
they had been speaking prose all their life. The idea is not new, although it can
seem novel; it was just temporarily misplaced. Words like diplomacy, trust,
and reciprocity are central to managing networks.
WHEN DO NETWORKS SUCCEED?
Networks thrive where markets and hierarchies fail, where trust and reci-
procity characterize the relationships between organizations, and where
management is by negotiation, not command. This much is obvious from
TEN LESSONS
Conditions favour network building when:

Multi-agency cooperation spanning the public, private, and voluntary sectors is required.
Professional discretion and expertise are core values.
Quality cannot be specified or is difficult to define and measure.
Actors need reliable, ‘thicker’ information, or local knowledge.
Commodities are difficult to price.
The policy arena is insulated from party politics.
Service delivery is localized.
Central monitoring and evaluation incur high political and administrative costs.
Implementation involves chains of organization and there is the prospect
of many ownership disputes.
There is a shared narrative or overlapping narratives of what we are doing and why.
Figure 5.1 When to build a network: ten lessons.

Source: Updated from Rhodes 1998.
Figure 5.1. Also, as with any other form of public sector management, success
depends on the relevant information, skills, and resources. When actors hoard
information and resources, when in effect they refuse to share, then the
cooperation that defines networks is unlikely to be forthcoming. The existing
literature also identifies several other more specific conditions under which
networks will thrive (see Figure 5.1).
In other words, networks succeed when service delivery is cooperative,
depoliticized, professionalized, localized, and customized.
WH EN DO NETWO RKS FAIL?
Networks, like all other resource allocation mechanisms, are not cost free.
Managing the institutional void is difficult. Networks are hard to steer—it has
been likened to herding cats, or pulling rubber levers. It is a time-consuming
and slow process.
THE S OUR L AWS OF NETWORK GOVERNANCE
There are at least four recurring dilemmas: managing the mix, the problem of
many hands, the holy grail of coordination, and local ownership. I introduced
these sour laws in Chapter 3, this volume. I expand on them here.
‘Why Can’t They Be More Like Us?’
Networks are difficult to combine with other means for delivering services. It
is too difficult to manage the mix of cooperation and the competition of (say)
contracting. One clear effect of marketization is that it undermines the
effectiveness of the networks it spreads. Contracts undermine trust, reci-
procity, informality, and cooperation. I am not arguing that networks are
unworkable. All governing structures fail. Governments have to find the right
mix because the several mechanisms can mix like oil and water. Competition
and cooperation are uneasy bedfellows. When government promoted both
competition and contracting out in the National Health Service, the result was
to ‘corrode . . . common values and commitments’ and ‘to create an atmos-
phere of mistrust’. Market relations had ‘corrosive effects’ on ‘professional
networks which depend on cooperation, reciprocity and interdependence’
(Flynn et al. 1996: 115 and 136–7). In short, contracts undermine trust,
reciprocity, informality, and cooperation.
Individuals cope with the clash between cooperative and competitive styles
by drawing on folk theories; on their inherited stories of how things work
around here (see pp. 84–5). They can live in this mixed world. They
calculate which service delivery mechanism will work in which context. But
the dilemma is pervasive. They miscalculate which message to send to other
actors or which setting is appropriate for their message. For example, joint
action between the police, the health authorities, and social care on a problem
housing estate foundered on such difference. The draft action plan was under
consideration. The police representative noted that only his colleagues had
operational performance indicators. The indicators for the other participants
were aspirational and vague. He knew that this difference would mean that the
police would attract criticism; it would be obvious when they ‘stuffed up’. The
other parties pleaded that their work did not lend itself to precise targets. As he
left the meeting, the frustrations of the police representative could be heard in
an exasperated yet heartfelt plea for other agencies ‘to be more like us’.
‘Not Me, Guv’
I noted in Chapter 3 the ‘problem of many hands’, where responsibility for

policy in complex organizations is shared and it is correspondingly difficult to
find out who is responsible. Bovens (1998: 46) also notes that fragmentation,
marketization, and the resulting networks create ‘new forms of the problem of
many hands’ (Bovens 1998: 229). In a network, the constituent organizations
may hold the relevant officials and politicians to account but no one holds
the set of organizations accountable. Performance indicators may enhance
efficiency but, in a multi-agency context, buck-passing is much more likely

because responsibility is divided and the reach of central agencies is much
reduced (Mulgan 2003: 211–14). All too often, networks are closed to public
scrutiny; they are a species of private government. The brute fact is that
multiple accountabilities fuel fears of duplication, overlap, and inefficiency
as they weaken central control.
King and Crewe’s (2013: ch. 14) description of the refurbishment of various
underground lines in London graphically illustrates the point. The bidding for
contracts was slow and protracted, and involved many lawyers, officials,
government departments, and private contractors. The Public–Private Part-
nership (PPP) involved three consortia, and one consortium, known as
Metronet, involved eight individual firms. The fragmentation of ownership
and investment made no sense to any outside observer:
For the PPP to work, every bit of it had to work, and there were far too many bits
of it, and the relationships among them were far too complicated. Everything had
to go right, and there were far too many things that could go wrong
(King and Crewe 2013: 219).
And things went wrong. It was a fiasco that ‘wasted years and cost billions’:
‘You couldn’t make it up. You simply couldn’t make it up’. Even worse, ‘no
one—except the British taxpayer—paid any kind of price for the Metronet
fiasco’ (King and Crewe 2013: 221 and 351). The main culprits among
politicians and political appointees were Gordon Brown, John Prescott, Geoffrey
Robinson, and Shriti Vadera but the lawyers, management consultants, and
individual firms were also culpable. No one was held to account and amidst the
complexity everyone could plead it was ‘not me, guv’.
‘You Can’t Shake Hands with a Clenched Fist’
The search for coordination has a centralizing thrust. Its advocates seek to
coordinate departments and other agencies—whether central agencies, the
states, or local governments, whether public or private—by imposing a new
style of management on other agencies. A command operating code, no
matter how well disguised, runs the ever-present risk of recalcitrance from
key actors and a loss of flexibility in dealing with localized problems. Gentle
pressure relentlessly applied is still a command operating code in a velvet
glove. When you are sitting at the top of a pyramid and you cannot see the
bottom, control deficits are an ever-present unintended consequence.
What we see here is an age-old problem dressed up in fashionable phrases.
That problem is coordination. For example, in Australia, the ‘whole-of-
government approach’ sought to ensure that public service agencies worked
across departmental and jurisdictional boundaries to achieve shared goals
and integrate government action (Management Advisory Committee (MAC)

(2004: 1). A Cabinet Implementation Unit (CIU) was set up in the Depart-
ment of Prime Minister and Cabinet (PM&C) to support the initiative. As with
all such top-down coordination initiatives, there were problems.
First, how do you get ministers to buy into interdepartmental coordination?
The short answer is reluctantly. They want to make a name for themselves, not
their colleagues. Second, departments are competing silos. The rewards of
departmentalism are known and obvious. For interdepartmental coordin-
ation, it is the costs that are known and obvious! Coordination costs time,
money, and staff. Indeed, as Wanna (2005: 9) argues, whole-of-government is
a side show for most managers. Third, coordination is for central departments!
It serves their priorities, not those necessarily of the departments. Fourth,
there is a tension between managerialism, which seeks to decentralize
decision-making, and the call for better coordination, which seeks to centralize
it. Fifth, in Australia, federalism is a major check of Commonwealth aims. It is
brutally simple. Coordination is for the Commonwealth, not state govern-
ments and other agencies. The Commonwealth does not control service
delivery. It has limited reach, so it has to negotiate. As in the UK, central
coordination presupposes agreement with the priorities of central agencies
when it is the lack of such agreement that created many of the problems.2 The
participants complain that ‘you can’t shake hands with a clenched fist’. In
short, coordination is the ‘philosopher’s stone’ or ‘holy grail’ of modern
government, ever sought, but always just beyond reach, all too often because
it assumes both agreement on goals and a coordinator (Seidman 1975: 190).
There are alternatives. Bottom-up coordination based on informal coord-
ination is ever present. Decision-makers adjust to one another of their own
accord to serve their own goals. They simply adapt to decisions around them
or seek to induce changes in other decision-makers. There is no central
authoritative decision-maker. Rather, each decision-maker seeks to induce
others to adjust using such methods as bargaining, reciprocity, manipulation,
and compensation. Local decision-makers are often comfortable being left to
their own devices. Central decision-makers dislike such informal coordination
because it is ‘messy’ and difficult for them to control.
‘They Talk the Talk but They Can’t Walk the Walk’
This conflict comes in various forms, notably: individual versus organizational

commitments; local versus national public expectations; local flexibility versus
2
The UK’s version of the whole-of-government approach, known as joined-up government,
also experienced most of these problems; see Bogdanor 2005; Ling 2002; Pollitt 2003; Rhodes
2000c.
national rules; and network goals versus national regulators. Local networks
cease to be local networks when centrally manipulated or directed. In effect,
when networks are managed centrally, horizontal relationships are trans-
formed into vertical relationships. Central agencies have to calculate whether
the costs of agreement are greater than the costs of imposition and all too often
it finds for the latter. Such relationships are exercises in official consultation; at
least this phrase does not imply any local discretion. The dilemma is between
hands-on versus hands-off styles of intervention; between strategic guidance
and local flexibility. Central actors can adopt a decentralized negotiating style
that trades a measure of control for agreement. This style of hands-off
management involves setting the framework in which networks work but
then keeping an arm’s-length relationship. Central actors find self-denial
even harder to keep than New Year’s resolutions.
For example, a children’s playground had fallen into disuse and was used as
a hang-out for unruly youths and drug users. The local neighbourhood
decided to take back the playground. They repaired the lighting, removed
the rubbish, whether broken bottles or needles, repaired the swings, and called
in the police to move the troublemakers on. The local elected councillor found
out about their initiative, met the parents, and suggested seeking the support
of the local authority. So, they applied for a small grant from the local
authority to install rubber safety flooring. They got the grant and one of the
parents volunteered to get the safety flooring cheap and install it himself. It
was not allowed. A requisition had to be submitted and approved through
central purchasing, and so on. The parents had to deal with local bureaucracy,
its forms, and reports. They resented such intrusions. It was their initiative;
they owned it. In their eyes, the local authority was supposed to support local
people, not tie them up in red tape. As one parent caustically observed, ‘they
can talk the talk but they can’t walk the walk’.
HOW D O CENTRAL AGENCIES MANAGE NETWORKS?
This question assumes a top-down view of network management; that central

agencies will have a hands-on approach to network management. Central
agencies confront two broad tasks; managing its portfolio of networks, and
managing individual networks.
Central agencies are the nodal points of networks. Commonly, the
term refers to one or other of central government in a unitary state, a
federal government, state governments, and local authorities. Each of these
‘central agencies’ belongs to, and seeks to manage, a group of networks;
its ‘multi-network portfolio’ (Ysa and Esteve 2013; see Wassmer 2010 for
a review of the literature). Managing the network portfolio has its own distinct
TEN COMMANDMENTS
Build a shared narrative. Share, adjust continuously, and locally negotiate

the network’s objectives.
Be a diplomat; represent your agency and the network.
Constantly nuruture; be patient and facilitate interactions by working with and
through people.
Integrate agendas and understandings of what the agenda is for.
Set ‘rules of the game’ for the broad operating context but leave action on
local problems to network members.
Know your limitations, develop clear roles, expectations and responsibilities
for all players and stay within the boundaries of your network.
Recognize each other’s expertise and shared working culture.
Share agreed and delegated decision-making and service delivery arrangements.
Provide incentives for agreeing criteria on what constitutes success.
Take your share of the joint administrative burden.
Figure 5.2 How to manage your network: ten commandments.

Synthesized and modified from Agranoff 2007; Ansell and Gash 2008; Denhardt and Denhardt 2000;
Goldsmith and Eggers 2004; Goldsmith and Kettl 2009; Goss 2001; Klijn and Koppenjan 2015; McGuire
2002; and Rhodes 1997b, 2006a.
challenges. The most obvious challenge is to find out which networks the
agency is trying to manage. All too often, an agency has no map of its own
networks let alone the networks of other central agencies. There will be no
mechanisms for coordinating the responses of a central agency to either the
portfolio or individual networks (see Heimeriks et al. 2009 for a review of the
tools for managing portfolios).
If central agencies treat networks as tools, then the means for managing
individual networks fall into three broad groups. Instrumental tools include
rewards, sanctions, and micro-management of the network. Interaction tools
encompass negotiation and diplomacy. The phrase ‘institutional tools’ refers
to the central agency changing the rules of the game; that is, resetting the
boundaries to network behaviour (see Kickert et al. 1997; and Chapter 3, this
volume). My list of Ten Commandments combines these different types of
tools (see Figure 5.2).
So, a public servant has to master some specific skills for managing net-
works. They include: integrating agendas, representing both your agency and
the network; setting broad rules of the game that leave local action to network
members; developing clear roles, expectations, and responsibilities for all
players; agreeing the criteria of success; and sharing the administrative burden.
As important as the specific lessons and commandments of Figures 5.1 and
5.2 are the ideas embedded in them: top-down versus bottom-up, organiza-
tional glue, rules of the game, collaborative leadership, and storytelling.
Top-down versus Bottom-up
We know from studies of implementation and of street-level bureaucrats that

top-down initiatives are often confounded. Top-down models of implementa-
tion take the government’s stated policy objectives as the starting point and
examine why they do not succeed (see, for example, Pressman and Wildavsky
1984). Failure is attributed to a diffusion of intent as multiple agencies become
involved, and a lack of compliance or understanding or both. Oversight, moni-
toring, and stronger direction from central agencies are common palliatives,
often with little effect. The view from the bottom up is different; it highlights that
implementation is mediated through the actions of front-line workers whose
perspectives reflect local conditions, local knowledge, and professional expertise.
For example, Lipsky (1980: xii) argued that ‘the decisions of street level bureau-
crats, the routines they establish, and the devices they invent to cope with the
uncertainties and work pressures, effectively become the public policies they
carry out’. My account of networks takes it as self-evident that they are messy and
amorphous and that central agencies may intervene but cannot control.
Organizational Glue
The notion of organizational glue overlaps with many other notions; for
example, occupational culture, institutional memory, and departmental
philosophy. All have in common the idea that the inherited beliefs and
practices of members of an organization are the social glue that binds an
organization together. Networks are no exception, with the obvious qualifica-
tion that an incipient network will have little or no inheritance and so little in
the way of organizational glue. There are at least two ways to develop such
glue. First, the network leaders can construct a narrative that frames the
experience of other network members (see pp. 84–5). Second, the workings of
the network can create shared experience. In rising order of difficulty, network
members can share information, agree to limited working together, undertake
strategic planning, provide integrated service delivery, and pool resources. What
is feasible is best because ‘quick wins’—for example, successful joint ventures—
will meet members’ expectations, build trust, and reinforce network behaviour.
Networks provide the glue that holds together contending interests.
Rules of the Game
Many networks work in the shadow of hierarchy; that is, they are dependent
on central agencies for legal authority and financial resources but are at arm’s
length for implementation. In turn, because they include the private and
voluntary sectors, the networks provide more resources for central agencies.
Networks are a bridge to civil society. The role of central agencies is to set the
boundaries to networks’ actions by, for example, strategic planning. The
problem with strategic planning is that it can become a millstone around
the network, daunting in its length, detailed prescriptions, and wealth of
performance indicators—and that is before it is overtaken by events. It need
not be so; for example, one senior British government minister brought in his
department’s strategic plan scribbled on one sheet of A4 paper. His strategic
plan was a signpost for the department, not an A–Z street map.
Collaborative Leadership
Ansell and Gash (2008: 544) define collaborative governance as a collective

decision-making process ‘where one or more public agencies directly engages
non-state stakeholders’ in the ‘formal, consensus oriented, and deliberative’
implementation of public policy or management of public programmes. The
key question is whether opposing stakeholders can work together in a collab-
orative way. The answer is a ‘cautious yes’ and a key part of that answer is
leadership, which is ‘crucial for setting and maintaining clear ground rules,
building trust, facilitating dialogue, and exploring mutual gains’ (Ansell and
Gash 2008: 12–13). Such leadership is variously described as: hands-off, soft,
integrative, facilitative, or diplomatic. This leader as local entrepreneur is about
building networks. For Stoker (2004: 139), the task is ‘to facilitate voice in
diverse communities, and reconcile differences, develop shared visions and
build partnerships to ensure their achievement’. Leadership is not about control
but about supporting people as they find their own solutions.
Storytelling
The ‘art’ of collaborative leadership lies in storytelling, which is increasingly

recognized as a tool for managers (and see Chapter 12, this volume). Thus, the
business studies literature sees leadership as being about ‘the management of
meaning’ and a way of leaders ‘exerting their influence on followers’; they
‘educate, inspire, indoctrinate and convince’ (Shamir et al. 2005: 14 and 15). It
is ‘socially constructed through interaction’ and effective leadership ‘rests
heavily on the framing of the experience of others’ in which ‘language, ritual,
drama, stories, myths and symbolic construction . . . play an important role’
(Smircich and Morgan 1982: 258 and 262). Such insights are translated into
practical lessons throughout the organization in the stories told by senior
management (see, for example, Denning 2004; Rhodes 2011a). So, the key
facilitative role of the collaborative leader in building and managing a network

is framing the shared story about what we are doing and why we are doing it.
CONCLUSIONS: SOME P LAUSIBLE CONJECTURES
Political scientists have a poor track record of prediction but we can aspire to
‘plausible conjectures’; that is, to making general statements that are plausible
because they rest on good reasons and the reasons are good because they are
inferred from relevant information (paraphrased from Boudon 1993). The
relevant information underpinning my conjectures about the sour laws are
summarized in Table 5.2, which summarizes the preceding sections of this
chapter.
Table 5.2 The sour laws of network governance.

Dilemmas Unintended Consequences Reaction of Network Members
Cooperation versus It’s the mix that matters ‘Why can’t they be more like us’
competition
Accountability versus The problem of many ‘Not me guv’
efficiency hands
Control versus mutual The holy grail of ‘You can’t shake hands with a
adjustment coordination clenched fist’
Strategic direction versus Disputed ownership ‘They talk the talk but they can’t
flexibility walk the walk’
What specific conjectures follow from the analysis in this chapter? Five
seem obvious.
1. Networks are not tools of central agencies to be managed to achieve
central objectives. Their strength lies in independence from central
agencies.
2. Central agencies can help to build and manage networks by limiting
their interventions to managing the network portfolio, training collab-
orative leaders, and providing the technical assistance.
3. The central agency can provide strategic signposting that sets the broad
boundaries for local network action to ensure networks work in the
shadow of hierarchy.
4. Collaborative leadership needs diplomatic skills and the art of
storytelling.
5. The goal is to develop organizational glue from information sharing,
working together, integrated service delivery, and pooled resources, thus
ensuring the network’s future.
Strategic Storyteller
A call came via our help desk on a

Thurs afternoon at 5:30pm. It
was Mrs. Glover, a client
of ours for 15 years. What
she wanted was a . . .
Figure 5.3 Strategic storyteller.
But networks are messy. There are no guarantees of successful results; only
the relentless pressure from the sour laws of network governance and the
imperatives of constant nurturing. To update François de Callières, network
leaders, whether drawn from the central agency or from network members, need
to stand in the other person’s shoes or (if you prefer) sit where the other person
sits. According to Clifford Geertz (1973: 9), cultural anthropology is about telling
‘our own constructions of other people’s constructions of what they and their
compatriots are up to’. The would-be network manager is a budding anthro-
pologist. This strategic storyteller must understand and share other people’s
stories to create the glue that holds the network together (see Figure 5.3).
6
Putting the People Back into Networks
Where you sit does not determine what you see. You cannot read off beliefs
and preferences from institutional position. This chapter looks at the experi-
ences of consumers, managers, and permanent secretaries of living and work-
ing in networks to illustrate the proposition that we must put people back into
networks, and to recommend an ethnographic approach.1
The chapter illustrates an argument. In Chapter 3, I criticized the existing
literature’s concern with typologies and abstruse theoretical arguments. In
place of such modernist-empiricism, I argue there is no essentialist account of
networks that can be used either to produce law-like generalizations or to
legitimate advice to policy-makers (see Chapter 7, this volume). The road to
understanding lies in decentred accounts focusing on the political ethnog-
raphy of networks and on narratives that recognize the creative individual.
Individual actors construct networks. They are not created by governments or
imposed by the researcher. So, the key question is ‘whose network?’ As
researchers, we write constructions about how other people construct the
world; we produce ‘thick descriptions’ of networks (Geertz 1973: ch. 1).2
CUSTOMERS ’ ‘FACTIONS ’: ‘ SEEMS L IKE

A RAILWAY STATION’
These cases are based on the files of a local authority in northern England.3
The social workers involved wrote them. After discussing the draft with the
1
This chapter is an edited version of R. A. W. Rhodes (2002) ‘Putting the People back into
Networks’, Australian Journal of Political Science, 37 (3): 399–415. © 2002 Routledge. Reprinted with
permission.
2
For general introductions to ethnography, elite interviews, and related qualitative
approaches, which are relevant to political science, see Rhodes 2017, Volume II, Chapter 3.
3
In this chapter, the methods involved are elite interviewing and accounts written by partici-
pants but my argument covers the range of ethnographic tools, such as participant observation.
social workers, I edited them. They then agreed the version produced below.
All names have been changed to protect the anonymity of both customers and
local authority employees.4
Case 1
Mr and Mrs R live in a two-bedroom house in the suburbs of a town with a
population of some 200,000. Mr R is 83 years old, and wheelchair-bound
following a stroke six months ago. Mrs R is 79 years old, still active mentally
and physically but not strong enough to help with her husband’s personal
care without help from one other person.
For the past six months they have had a care assistant from a private
agency to help Mr R with getting up, toileting, washing, and dressing every
morning. A local authority Home Help calls at lunch to help with toileting,
and personal care tasks if necessary. The Home Help also calls twice weekly
to do shopping, as Mrs R can’t leave Mr R, because he gets distressed when
left on his own. Mr R has a catheter that is managed by his wife and checked
by a Community Nurse twice weekly. Three nights a week (Friday, Satur-
day, and Sunday) a private agency care assistant calls to help Mr R to go to
bed. The council’s Home Help service help on the four remaining evenings
a week. The evening call can take place any time from 7.00 p.m. to 9.00 p.m.
depending on daily demand on staff. The local authority care manager
arranged and purchased the private agency.
Mr and Mrs R moved their double bed into the lounge because the
bathroom is downstairs at the back of the kitchen and Mr R cannot get
upstairs. They live and entertain in their small kitchen. Mr R cannot get out
without being lifted because there are three steep steps at the front and at
the back of the house that make it difficult to install a ramp.
To make themselves more comfortable their care manager suggested
moving to a new comfortable sheltered housing complex in the centre of
town. They have an offer of a one-bedroom flat with a kitchen and living
room on the first floor. There are lifts. There is a communal room with
regular activities.
Mr R would be able to move freely around the flat and use the kitchen, as
the units are wheelchair height. He would be able to use the lift and attend
the activities at the communal room. He would need assistance at home for
personal care. Mrs R would be able to get out to do some shopping while her
husband is joining in the communal activities. She would not be as isolated
as she would be able to join in with her husband.
4
The local authority care workers were told to refer to the users of their services as
‘customers’. I follow local practice. I use the term ‘faction’ to make it clear that the fieldwork
has been anonymized and disguised to protect the people involved.
Putting the People Back into Networks 89
Mr R will not consider looking at the flat until he knows he can have the same
carer from a private agency who calls every morning. This will not be possible
because his care arrangements will be provided by different locally based staff.
His wife needs help to explain this. The Home Care Manager responsible for the
new area visits the couple to reassure Mr R that he and his wife will get all the
help that they need. The couple visit the new flat and accept the offer.
Case 2
Mrs T is 80 years old and arthritic. Her local GP has referred her, asking for
Home Help. She lives on her own in a bungalow. She uses a walking frame to
help with walking. She can no longer manage pans and cooking for herself.
She was coping well until she fell five days previously fracturing her wrist. She
visited the hospital casualty department for treatment on the wrist and was
discharged home. A friend has been helping but she is elderly and finding the
constant help that Mrs T needs too much of a struggle for her.
The Home Care Manager visits and assesses Mrs T. She is slow and finds
holding the frame difficult because of the arthritis in her hands and frac-
tured wrist. She has difficulty with washing, dressing, toileting, bathing,
preparing food, cooking, and shopping.
The friend who calls in has been cooking and shopping and helping with
personal care. She would still like to visit her friend twice a week and will do the
small amounts of shopping and get Mrs T’s pension when she gets her own.
Mrs T’s three children all live away from their hometown, have their own
families, and work. The eldest expects to retire in the next year. The family
arranged to take it in turns to visit on Sundays, keep the house and garden tidy
and in good repair.
The Home Care Manager asks for an urgent visit from the Occupational
Therapy Services to assess Mrs T for equipment for daily living. While
waiting for this assessment a home help will call at mealtimes and help
with dressing in the morning. The friend will call about 7.00 p.m. to help
with undressing.
Two days later an Occupational Therapy Assistant (OTA) calls to assess
for equipment to help with daily living. Mrs T can eat with special cutlery
and a plate guard. She can manage a cup of tea with a kettle tipper if it is laid
out for her. Mrs T can manage toast or cereal for breakfast if put out before.
Tea is manageable with bread, butter, cheese, or cold meats. Mrs T can
manage her gas cooker with the help of replacement dials. The kitchen is
well organized. With a perching stool she can sit at the work surface next to
the cooker to eat her meals. She can wash and dress herself with equipment
but needs help with doing up buttons, laces and zips, and putting on
stockings. With carefully selected clothing from the wardrobe, Mrs T will
need minimum help to dress and undress. She needs a raised toilet seat and
frame in the bathroom and a bath board on the bath with a grab rail on the
wall. The equipment, except the grab rail, is provided later that day. An
emergency warden call system will be installed by the end of the week by the
council’s housing services. The Gas Board will call within 48 hours to
replace the dials on the cooker.
The Home Care Manager rearranges the home help. She provides a
morning call from her own services Monday to Friday and arranges a private
agency on Saturdays. The home help will help with buttons and to collect
shopping and pension or to do some basic cleaning. They will do the laundry
and ironing. One hour a day is allowed. The home carer helps Mrs T to use
her bath board to have a bath one morning a week. Breakfast and tea are laid
out and the kettle is filled for the day. A twilight service will call any time
between 7.00 and 9.00 p.m. Monday to Saturday to help with undressing;
fifteen minutes are allowed. These services are arranged and purchased by the
Home Care Manager.
The Women’s Royal Voluntary Service (WRVS) delivers Meals-on-
Wheels Mondays and Fridays. Frozen meals are cooked at a local primary
school and delivered by the home help. On Saturdays Mrs T will treat
herself to a meal cooked and delivered by a local hotel.
Mrs T does not get out at all and with increasing disability does not
feel that she can consider going out. She is isolated. Various local
centres have activities for the elderly either run by the council or
voluntary agencies such as Age Concern, which runs a post-hospital
discharge support service. Used to her own company, Mrs T is nervous
about mixing with others. She is so grateful for all the help she gets; she
does not want to be a nuisance and does not like to ask for information
and more help. She is also hard up, getting only her pension. The Home
Care Manager is busy and now that all the arrangements are in place
will make only a quick visit to check every six months.
Mrs T will pay the second-tier home care charge of £5.00 a week because
she does not receive Income Support or Council Tax benefit. Her meals will
cost £1.40 each Monday to Friday and her meal on Saturdays will cost £3.00.
She pays for all her service charges, Council Tax (£520 p.a.) and other
outgoings from her pension of £61.15 a week. She has savings of £7,000.
The social worker concerned wonders if Mrs T’s quality of life would be
improved if a care manager or social worker assessed her. Is the quality of
Mrs T’s life improved by the range of services provided by the Home Care
Manager and OTA? Social workers may have other resources at their
fingertips. Could someone take her to a local lunch club, a day centre?
She may like playing bingo or whist. What about a stay in a residential home
or a holiday with her friend perhaps? Is Mrs T entitled to more money?
What about Income Support, Attendance Allowance, and Council Tax
Benefit? Would Mrs T have more choice about the services she would like
if she had more income? Would she have to pay more for some of
the services?
Case 3
Mrs K was admitted to the local psychiatric hospital having become aggres-
sive towards her daughter. Mrs K was 78 years old and had been suffering
from dementia for two years. She lived alone in a large, detached Victorian
house in a village some ten miles from the local town.
Before admission, Mrs K’s family, particularly her daughter and two chil-
dren, visited daily and Mrs K had been coping well. The GP and Consultant
Psycho-Geriatrician kept in regular contact with the family and a community
psychiatric nurse visited once a week. Mrs K attended the specialist day hospital
one day a week and the local elderly persons’ day care centre one day a week.
The social services Home Help service called in a morning (Monday–Friday) to
see that Mrs K was dressed, breakfasted, and ready for any transport that would
be calling. On the days Mrs K was at home the home help called at lunchtime to
make sure that she had eaten the hot meal delivered by WRVS or the home
help (prepared at the local elderly persons home). The family called in at
teatime and again later in the evening to check that Mrs K was all right. Mrs
K tended not to sleep at night and would telephone neighbours in the middle of
the night for help. She had been known to wander out in the middle of the
night, disturbing the neighbours.
To give the family a break Mrs K was admitted to the local elderly
persons’ home for a two-week short stay. After two days Mrs K became
confused and disorientated, demanding to go home. The home manager
became concerned when Mrs K began to wander outdoors, something she
had not done while attending day care. Mrs K hit a member of staff who had
tried to escort her indoors. A decision was made to allow Mrs K to go to the
day hospital as usual and arrangements were made for her to return home.
On returning home Mrs K became even more disorientated, wanting to return
to her home to be with mummy and daddy. She hit her daughter when her
daughter tried to take her around the house to convince her that she was at
home. The daughter broke down, saying that the family could no longer cope
and that she could not allow her children to visit with their children if Mrs
K was going to be violent.
The Consultant visited and decided that Mrs K needed reassessment and
a review of her medication. She was admitted to the psychiatric hospital.
The assessment showed that Mrs K’s condition had deteriorated rapidly
into a delusional and challenging phase. No long-term beds were available
from the health trust but, with the correct medication, Mrs K’s condition
and behaviour could be reasonably well controlled. Mrs K was referred for
permanent care to the social services care manager attached to the hospital.
The care manager decided that Mrs K needed 24-hour supervision and
assistance with all aspects of daily living and personal care. The care
manager discussed this with the family who are clearly distressed at the
idea of nursing home care for Mrs K. With the new medication, Mrs K was
less aggressive and they believed that they could manage with a full package
of care for Mrs K. They insisted that Mrs K returned home.
The Care Plan was complicated and expensive, costing an average £360 a
week. Cover for each hour of 24 hours was worked out for two-week blocks.
The Care Plan involved the health trust’s psychiatric day services, private
nursing home day care, private agency home care, voluntary home care
sitting services, social services home care, and several transport services
supporting the daughter to provide 24-hour care at home. The social
services felt that it would be more cost-effective for Mrs K to go into a
nursing home and assessed the cost to the local authority of her care to be
£186 a week. Following recent case law the local authority was within its
rights to offer nursing home care to meet the need. They could have asked
the family to supplement the cost of the care.
Because there were continuing health care issues, the care manager was
able to secure funding from the joint finance continuing care fund for the
first four weeks. The care manager was unhappy about the arrangements
because it is widely known, and particularly in the case of Mrs K, that
constant changes in carers and location add to disorientation and confusion
for people with dementia.
The daughter was concerned that she would have to give up work and
asked if a private agency could be bought in for the equivalent amount of
money. This would not help; there would continue to be many carers visiting.
The house would seem like a railway station with so many people calling.
The arrangements were set up on discharge but quickly became difficult
to manage as Mrs K would not attend day care some days and more care
had to be bought in from the private home care agency. There were frequent
problems with maintaining regular home care support with home carers
being ill, late or not turning up and replacements being difficult to find in a
rural area. Mrs K’s son and daughter-in-law did what they could to help but
they did not live locally and were not as committed as their sister.
After one month the care manager decided to approach Community
Service Volunteers (CSV), a national volunteer agency that provides young
volunteers as full-time live-in carers. Four weeks later two young volunteers
began providing 24-hour care and supervision for Mrs K. The care plan
covered a three-week block with supplementary help for the CSVs. The cost
of the care was similar. Two months later the young volunteers were feeling
the stresses and strains of caring full-time for an elderly person with demen-
tia. They were fighting among themselves and one young person began taking
the medication prescribed for Mrs K and drinking alcohol. She was asked to
leave. One week later Mrs K died.
MANAGEMENT ‘ F A C T I O N’ : ‘ WE ARE
IN THE S HIT IF W E DON ’ T’
This example is from an interview with the chair of a primary care trust. He
has been in post for less than a year and the Primary Care Trust (PCT) is a new
organization. He is soberly dressed and the interview is businesslike. It takes
place in his study in his modern house in an exclusive housing development
aimed at the executive and professional classes. We sit at opposite sides of the
table. If I can be forgiven a shorthand cliché, he represents the ‘new manage-
ment’ in the UK public sector. His job is to provide strategic leadership for
the primary care sector of the National Health Service (NHS). Primary care
comprises the services provided in a specific geographic area by family or
general practitioners and the community health services, such as midwifery,
which look after people in their own homes. The following words are his own.
I have edited them into a continuous text. The passages in italics are my
comments or questions.
START OF TRANSCRIPT
Our major partner is the local authority as a whole, not just social services.
One of the first people I met was the chief executive of a local authority. I rang
him up and said I’d like to see him and he and the leader of the council came
round to see me. The chief executive talks about the town and, I mean, he is
strongly committed to the notion of the local authority as community leader
but is very, very keen to have as many meetings as possible outside the council
offices. One of the first things I talked to them about was developing a
common planning capacity to support the community plan. We are a small
organization and I’m keen for us to play a part right across the community
because the potential impact of what they do on us is great and vice versa.
Within a few weeks they asked me if I would be the vice-chair of the
community safety partnership and that, it turned out, was really to chair it
because the person who chairs it, who is the former leader of the council, was
never there. Whenever I’ve been there, which is three times now, I’ve chaired
it. So, yeah, they were very keen to get us into things like that. I knew the
former director of social services very well. He moved on to a job in the
Department of Health at Christmas. Yeah, I’ve known him for years and
years, and I guess he knows a lot about what I’ve done in the past and I’m sure
he’s passed that on to people in the local authority.
I certainly see the director of the community services, who is the chief
officer responsible for community safety, regularly. I see the guy who—he
has chief officer status—is responsible for the local strategic partnership in
the community plan. There’s a lot of issues about mental health and crime
and only this week I discovered a whole set of issues around prison health.
We primarily meet in a partnership group. It is one of five task groups. It
reports to the overall local strategic partnership. The local strategic part-
nership is the over-arching liaison, strategic, planning mechanism that
brings together all the elements of the community plan, but if we need to
have a one-to-one, yes. For example, I wanted to see the local commander
and again he came around to the office for about an hour just to chat.
Apart from the local authority and the community plan, the other key
actors are the provider trusts and in our case there is an acute trust, a mental
health and learning disabilities trust. Then, in addition, there’s the whole
primary care sector. Obviously in some respects they are major providers
but in the main they are still independent contractors and they are not on a
contract with us, but they are our partners, they are part of the trust.
There are separate meetings of all the chief executives, and there are
separate meetings of all the chairs, though as a result of a proposal I made at
the last chairs’ meeting, we’re gonna have some joint meetings. But most of
the business is done through bilaterals. There are some exceptions that sort
of prove the rule, like there was a review of acute services. There is the
financial agreement each year, what’s called the Service and Financial
Framework (SAFF), which is certainly the centre of the financial frame-
works. Essentially that’s where each purchaser agrees with the local pro-
viders what the cash envelope is in the coming year, and what targets they
will meet in terms of delivering their services, and that is a politicized
process with a small ‘p’, which has been brokered by the health authority.
This year because there are lots of deals that have to be done around the two
big acute trusts—which have implications for the rest of the services, as they
tend to swallow up a lot of the growth—those deals have to be brokered on
an area-wide basis.
The health authority is also a major actor. We have to sign an annual
accountability agreement with it. The essential element of it is that we will
meet the targets laid down in the national NHS plan. We meet them on a
quarterly basis. There’s the regional office of the NHS executive, to which
we are accountable via the health authority. Our provider trusts are directly
accountable to the regional office. The regional office has to broker the
SAFFs if the health authority can’t do it, and the regional offices in our
region put a lot of pressure on the health authorities to get everything signed
up by the end of March. The regional office monitors us through the
quarterly returns that we make to the health authority and then the regional
office monitors the health authority.
I mentioned the meetings of the chairs of the trust and the health
authority. The first meeting I went to, it was absolutely clear that we’ve got
a major problem with the East acute trust, which has been built up to provide
regional specialties. It was saying that it couldn’t meet its cost reduction
targets and if it can’t do that, then there is no growth money available to use
to develop some other areas of service in which I’m strongly interested. So,
I’ve used a combination of informal lobbying of some of the other chairs,

plus the meetings to say, ‘Actually, I don’t think we can let this happen.’ We
can’t have the chair of another trust coming along and saying I’m sorry but
we’re not going to meet these targets. What I’ve done so far is I have been
careful to build up a relationship with the acute provider on the West bank
who provides most of our acute services and who also suffers if East gets all
this cash; and with the community mental health trust who will lose a lot.
You know we’ve got poor mental health services which I’d like to see
improved and there’s no way they’ll get improved if we can’t get this
money out of the trust. It hasn’t been necessary to take it outside the area
so far. It may be, it may be.
Within the town, all the major players work in offices within ten minutes’
walk of one another. Domestically and socially, everybody knows where you
live and where you went. At the senior level, a number of people meet for
lunch and have drinks during the day and things like that. I’ve met the
director of social services to have a pub meal. There is, undoubtedly, a local
network which is beginning to self-consciously think about organizing
itself, rationalizing a lot of the activities. I think it was Wednesday I was
chairing the community safety thing, and we talked about rationalizing the
way that youth offending, drugs, and community safety are handled, so now
it’s all gonna be within a single framework.
The voluntary or the private sectors are not immensely significant for
decision-making in the arenas that I operate in. It is not as significant in this
area as in some places because of the long tradition of municipal Labourism.
If not alive and well, it’s still there (laughs). The voluntary sector is, however,
a major player in service delivery, although it is not a major player in terms
of strategic development.
We have a responsibility to consult with and involve the public in setting
priorities and getting feedback about health services and we’ve got what we
call the communications forum, which meets about every two months. We
can get as many as 60 people into a meeting. We’ve invested a lot of time in
developing good relationships with a wide range of community groups.
There really is a local area network that has to be worked, managed, and
learnt.
Absolutely. Yep. Every week I will be involved in a meeting with some
part of it, which will be a decision-making meeting in the main.
We haven’t talked also about the inter-relationship between my other
roles because I’m part of national and regional networks by virtue of my job,
which plays directly into my PCT role as well.
You are actually plugging the local network into a bigger network.
Oh yeah. A very large part of my role is networking, ambassadorial. I was
reflecting on this over the last couple of weeks, perhaps partly because we
were gonna talk and you know partly because I was reflecting on the job
because I’ve been doing it for a while. It’s almost entirely self-managed.
There’s no requirement on me to make a lot of all my links.
What are the benefits and costs of this kind of management?
The time issue. We have a central government that is behaving proactive-
ly in relation to a whole range of sort of issues. So, people on the ground
are suffering from initiative-itis. The benefits are (long pause). The way
I conceive of health, and the role of health organizations, means that it’s
impossible to achieve any goals without working with and through other
organizations and other key actors regionally, and especially locally, and to
some extent nationally. It would not be possible to do the work that I do, it
wouldn’t fulfil the goals I have, unless I was approaching it in that way.
I guess I am trying to turn this into more of a managed network. I’m hoping
to talk to the chief executive of the local authority in the next week or so
about how we can rationalize some of our activities and how we can get this
common planning support capacity.
The cost of it? You know it obviously is time-consuming. I guess if you
didn’t naturally enjoy this kind of work then it could be difficult—if you were a
sort of shrinking violet as it were, did not have the personal qualities that go
with this. (Long pause.) I guess there have to be costs in terms of juggling so
many things at the same time. It would be easy to burn out yourself as well as
the organization. There’s no question about that. So, you have to keep it within
limits. I can’t . . . I mean . . . I suppose, to be honest, the honest answer is that it
is so new and I’m enjoying it so much that it doesn’t have any obvious costs at
the moment but over time they may become much more evident.
Networks presuppose some agreement on values and that agreement on
values is very elusive.
That’s absolutely right. There is an agreement on some basic values on the part
of most of the people I meet and the major one goes back to the old thing about
city pride. You know it’s the really old thing; we are working on behalf of the
people of the area. There’s a public service ethos, there’s no doubt about that.
Unless we work together, then we can’t actually manage ourselves out of some of
the difficulties we’ve got because if you want to do what you want to do, you have
to work with them. There’s a strong recognition that we are in the shit if we don’t.
END OF TRANSCRIPT
PERMANENT SECRETARY ‘ FACTION’ : ‘K E E P

T H E SH O W ON TH E R O A D’
The permanent secretaries of government departments sit at the top of a

hierarchy where three main tasks come together: political advice (to ministers),
management (of their departments), and diplomacy (or managing external

relations). It is a singular combination. The job’s ingredients would be instantly
recognized by earlier generations (and, for a more detailed account, see Rhodes
2011a). In these extracts from an interview transcript, a current permanent
secretary talks about his role in managing links with the rest of the civil service
and with the outside world. The interview takes place in a well-apportioned,
large office in Whitehall in the early evening. The private office serves tea. I am
sat in an easy chair. Initially he sits on the settee facing me across a coffee table.
He takes his jacket off. Before long he is stretched out along the settee. He is
relaxed. The conversation is punctuated with laughter and innumerable con-
versational asides. Again, if I may be forgiven a shorthand expression, he exuded
‘effortless superiority’ or, to avoid the pejorative connotations associated with
this phrase, the self-confidence of proven ability. His remarks are edited from six
hours of taped interviews.
START OF TRANSCRIPT
PS You knew people, you knew all of your own type, you know, the whole
generation of them, all the people who are assistant principals, you knew all
of them. You knew some bits of the department quite well, but you had
quite a narrow focus, which is unusual, because there are only one or two
really big departments like that. Then there was the network thing. You
were internally focused really, so you were looking up and across in our case
for the X division. So, you had that network. Then you go to the private
office. You were now on a different network. You’ve got both the network
down inside your department, across the whole of it, so you’ve got a
coverage which is huge, which is quite tricky, and you’ve got the network
across Whitehall and to an extent you’ve also got an international network
as well. So, you have to establish those relationships, keep them lubricated,
keep the show on the road.
Obviously what appeals about it is you’re dealing with people in Number
10, and in our case, we had well established ways of working, which
generally worked very successfully because we had the same little group of
people we’re dealing with. So, we knew Number 10, the Foreign Office, to an
extent the Treasury, to an extent the DTI, and defined bits of the Cabinet
Office, two defined bits of the Cabinet Office. If you worked those systems a
lot, the people you were working with were on secondment and you knew
them. Or if you didn’t know them the day you got there, they took you into
their circle. People in the civil service are basically open and welcoming and
want people to succeed. So, they took you into their network and you then
worked their network and then by the time you came out of the other end,
since this was basically a development thing, you’d developed an idea
about how the government as a whole worked; you knew how to work
that machine and you had a top-of-the-pyramid view of your own organ-
ization. And it works you know. It’s good training. I didn’t know this at the
time ’cause I didn’t really think about networks and how it all works, but the
people above you are getting to know who you are.
RAWR I see. So, presumably, necessarily, you begin to build networks
around you when you move to these kinds of posts?
PS You build it, yes.
RAWR You have to build it consciously?
PS Yes, and you build it differently. I’ve taken over posts from different
personalities and you do the job differently. I don’t know whether you do it
better or worse, but I’ve taken over from very capable people, where I’ve thought
‘well I’m doing this job much worse than them’, but others obviously were quite
happy at how I was doing it. But you do it differently. So, yes, you stamp your
personal style. Some people you know are great producers of paper and great
writers to people and so on and some are very assertive. I always thought I was
more sort of consensual and they’d probably say I was more manipulative. You
can have different styles, but you’re delivering and people see whether you’re
delivering or not. Now what are they measuring? They are measuring whether
you’re achieving the goals of that bit of the organization relative to other bits of
the organization, whether you’re delivering what ministers want, whether
you’re turning the paperwork over, whether the PQs are being answered,
whether there’s trouble, whether you can dodge it when it comes inevitably.
Most of what I was doing wasn’t the very glamorous stuff. It was
underpinning the work of the official committees and so on. So, you build
a network across departments at official level. At the end of doing that, you
know how the central government machine works.
I then moved to be the private secretary to the permanent secretary and
spent about two and a bit years as his private secretary. This again was a sort
of classic career move for a civil servant, a principal level civil servant. You
were sitting next door to the great man. You saw how he worked. You
underpinned what he did. The notional job description was fairly menial.
You were organizing the flow of paper and taking the records of meetings.
But you were on the inside and you could see how top people worked. So,
you saw another network, in a sense the permanent secretary’s network.
RAWR Was there a social network supporting the work network?
PS There was and there still is. There was a sort of network of private
secretaries. There was a network of permanent secretaries as there still is—a
lot of mutual loyalty amongst them. There was a sort of parallel network of
private secretaries. So, you were all in the same boat together, you worked
together, you got to know each other, and you all went up the machine
together. So, again I suppose you were creating these cross-boundary
networks. You spent a lot of your time working to make the system succeed
and you had that network, but you didn’t meet socially very often. I mean it
was quite funny. You might meet once a year at a party or something, and
you’d finally put the faces to the voices.
RAWR How has your job changed since becoming a permanent secretary?
PS Well, if you are in any big department and you are the permanent
secretary you are trying to give a sense of leadership to the whole depart-
ment under ministers. You don’t tend to get involved in those bits that are
going well. I used to leave them to get on with it. They could do it. They
knew what they were doing. They took the glory. If it went wrong, or
I thought it was going wrong, I would get involved. Also you spend a lot
of your time on sort of broader civil service management, corporate issues
across the whole service. If I worked out how I spend my time, quite a lot of
my time actually is spent on corporate issues across the whole civil service
and things in support of the government as a whole.
RAWR Could you give me an example of this work?
PS There is a civil service management committee, there’s honours work,
there’s discussions about where the civil service goes more informally and
so on, all those sorts of things
RAWR I wondered if this work linked into the current interest in joined-
up government?
PS There’s now a stronger sense of the need for permanent secretaries to
get together and talk about some of these joined-up issues. I find I have a
slightly different view on this to some others because in my previous
department we had long since discovered joined-up government. We
had worked on the basis that we joined up everything. We did with the
Foreign Office and with the Cabinet Office and with 10 Downing Street.
This was just deep in our culture; this was the way you worked and we got
this off to the finest art possible so that although you could get the
ministries saying something different about something to people, it was
only by design. You know everything we did we worked out with them. So,
joined-up-ness isn’t a great revolutionary idea for me. But I think it is
difficult on the civilian side.
There is a traditional role where permanent secretaries meet up and deal
with things, the paraphernalia of state in various ways, the senior civil
service appointments, the group that does honours. But I suppose under
the present government there’s been a feeling that the corporate manage-
ment of the civil service, as opposed to the management of the process of
selecting senior people, was insufficiently strong and we are still developing
that and that, as you say, policy-making was insufficiently joined-up and
that permanent secretaries needed to be involved in that and to give a lead.
END OF TRANSCRIPT
WHAT DO THE F ACTIONS TEACH US?
About Networks
The basic claim made for ethnographic method in general is that ‘It captures the
meaning of everyday human activities’ (Hammersley and Atkinson 1983: 2).
In a similar vein Fenno (1990: 2) argues, ‘The aim is to see the world as they see
it, to adopt their vantage point on politics.’ It encourages the researcher to
get out there and see what actors other than the elite are thinking and doing.
It generates descriptive accounts valuable in their own right (Hammersley
and Atkinson 1983: 237). Also it aids the development of theory because
extensive contact with people challenges the preconceptions of social scientists
(Hammersley and Atkinson 1983: 23). It is exploratory—‘unstructured soaking’
(Fenno 1990: 57)—and encourages fresh lines of thought. Research strategies
and ideas can be adapted quickly. And for those who are so inclined, it
can be used to test theory; by, for example, the detailed study of key cases
(Hammersley and Atkinson 1983: 24). Although my fieldwork extracts are brief,
nonetheless they do illustrate the potential of an ethnographic approach; of
‘thick descriptions’.
The most obvious point is that ‘network’ is an everyday term used by
consumers and managers alike to describe the web of relationships in which
they are embedded. But there are significant differences in what the term
means to each of them. Consumers experience networks as given, as complex,
confusing structures.5 Thus, Mrs K’s daughter sees her mother’s house as a
railway station because of its endless stream of visitors. British government
recognizes that this complexity is a problem for citizens. The clear, central
message of the consumer factions is the dependence of consumers.
The contrast with the view of the chair of the PCT is sharp. He sees himself
as constructing—that is, designing, building, and managing—a network not
just for his organization but for the local area. Indeed, I would argue that,
during our conversations, the chair worked out he was constructing an area
network. It was not an explicit part of his strategy at the start of our conver-
sation. He recognized that he is not required to build the linkages but believes
he will be in the shit if he doesn’t.
The permanent secretary also sees himself as constructing a network to get
the job done. In his case, however, it is a lifelong network built up as he is
socialized into the workings of the civil service. He is groomed to inhabit
5
Brian Hardy asked if I used ‘given’ in both senses of the word. When he asked, I had not.
I meant that the service delivery structures seemed fixed and immutable to customers. But he
makes an important point. Dependent customers also see the services as ‘given’ in the sense of
gifts, and they are grateful. Gratitude sustains dependence, thereby sustaining the immutable
quality of inherited beliefs and practices.
networks, trained to develop them. It is also an inward-looking network. ‘Other

organizations’ for the PCT chair refer to a disparate set of bodies but for the
civil service it refers substantially if not exclusively to other central depart-
ments. Central government has been described as a federation for many years.
Networking provides the glue that holds the parts together. The permanent
secretary understands networking as essential to producing corporate glue.
My claim for these short extracts is a simple one. The term ‘network’ has
different meanings for each respondent and the study of networks must
capture these meanings, not read them off from the beliefs and practices we
assume adhere to a specific position.
There is one important limitation to the analysis in this chapter. As the
observer, I report the interviews as if I am neutral and as if the data is given to
me in a pure or unmediated form. I am not that naive. All observers construct
their material drawing on their prior theories. We understand Mrs K, the chair
of the PCT and the permanent secretary, by understanding their beliefs and
practices relative to a specific context. We understand their context by looking
at the traditions in which they were socialized.6 I accept that my ‘factions’ are
my constructions of how my interviewees see their world, and that it is crucial
to locate people’s beliefs against a background of traditions.
About Methods
Fenno (1990: ch. 3) provides perhaps the most insightful account I have come
across of the opportunities and pratfalls of ethnography, especially participant
observation and elite interviewing.7 I draw on it extensively in this volume but
expand my discussion in Rhodes (2017, Volume II, Chapter 5).
First, the aim is to see the world through the eyes of the consumer, manager,
top civil servant; to make our construction of their construction of the world.
So, the key question is ‘Whose network?’ The network is not given to us. It
is built up through the accounts of its members. Don’t assume—ask, and
listen to the reply. As academics we are used to, even love, the sound of our
6
I develop these ideas in Rhodes 2017, Volume II, Chapter 2; and in Bevir and Rhodes
2003, 2006a.
7
My thanks to Nelson Polsby (University of California, Berkeley) for giving me a copy of this
slim but insightful monograph. I must demur from his view that it is easier to practise non-
participant observation in Britain than America because the academic, administrative, and
political elites mingle in Oxford senior common rooms and equivalent places. It is not easier.
It is different. It is more difficult to maintain distance; to be the outsider. I am not looking for
companions on high table but respondents. The more familiar, the more friendly the relation-
ship, the greater the constraints on what can be said and done. I should also note that
I conducted repeat interviews. For some interviewees I have more than six hours of conversation
on tape. These encounters are best described as conversations because the conventions of a
formal interview cannot be sustained over six hours.
own voice. In fieldwork, our voice can be the equivalent of static or white
noise—it interferes with reception.
Second, trust is essential—‘being nice to people and trying to see the world
as they see it. You need to be patient, come on slow, and feel your way along.
Two handy hints: Go where you are driven; take what you are given; and,
when in doubt, be quiet.’ I would add: be patient and stick around. Gradually
you become part of the furniture.
Third, both insufficient and too much rapport are problems. A professional
relationship can slip into a personal friendship. ‘I did not want them as
friends—only respondents’ (Fenno 1990: 75). If they invite you home, you
may not be able to refuse, but don’t take notes! Switch off as a researcher and
forget what you hear. Or, as one permanent secretary remarked to me about
an invitation from his minister, ‘It was right of him to ask, and right of me to
refuse.’ Both understood local custom and practice. To keep your distance,
Fenno suggests some rules of neutrality. ‘I have not registered with a party;
I have not engaged in partisan activity; I sign no political petitions; I join no
political organizations or interest groups; I engage in no radio, TV or news-
paper commentary. I do not allow my name to be used for political purposes’
(Fenno 1990: 67).
Fourth, be critical of yourself. It is all too easy to contaminate the relation-
ship between observed and observer and cause respondents to behave differ-
ently. The aim may be to remain the outsider but for lengthy on-site visits and
extensive repeat interview, you have to have a conversation. You cannot just
nod. Observing has its costs. You get tired, you forget quickly, and interviews
produce anxiety. Your notes are selective, a reconstruction. ‘The data is not
better then quantitative data. It is just different’ (Fenno 1990: 90).
Finally, there are sound criteria for judging the work. Judgement is made by
the researched—do they recognize themselves and their world? Judgement is
made by the academic community—does it ring true, does it say anything new
and insightful? (Rhodes 1997a: 190–2).
There is no one way to do research in the social sciences. I am all too aware
of the limits of ethnography (see Rhodes 2017, Volume II, Chapter 5). Such
methods do not work well in analysing such aggregates as the nation state.
Interviewees can be self-serving and misleading. The validity and reliability of
the data can always be disputed. But, despite such problems, I am prepared to
defend vigorously the proposition that our understanding of political life,
whether in the guise of political parties or policy networks, must be grounded
in observation.
7
Analysing Networks as Narratives

of Beliefs and Practices
Interpretive approaches begin from the insight that to understand actions,

practices, and institutions, we need to grasp the relevant meanings, the beliefs,
and preferences of the people involved.
By Bentham . . . men have been led to ask themselves, in regard to any ancient or
received opinion, Is it true? And by Coleridge, What is the meaning of it? The one
took his stand outside the received opinion, and surveyed it as an entire stranger
to it: the other looked at it from within, and endeavoured to see it with the eyes of
a believer in it . . . Bentham judged a proposition true or false as it accorded or not
with the result of his own inquiries . . . With Coleridge . . . the very fact that any
doctrine had been believed by thoughtful men, and received by whole nations or
generations of mankind, was part of the problem to be solved, was one of the
phenomena to be accounted for (Mill 1969 [1840]: 119–20).
In this chapter I ask, after Coleridge, ‘what is the meaning of it?’, where ‘it’
refers to policy networks.1
DECENTRING NETWORKS
In my summary of networks in Chapter 3, I sought to present a balanced

summary of a continuing debate. In this chapter, I confront the issue of: ‘How
do we know what we know in the study of networks?’ I do not summarize
other people’s criticisms of networks but present my own critique and outline
an anti-foundational way of studying networks as an alternative to the social
1
An abbreviated version of an article originally published in Greek as R. A. W. Rhodes (2003)
‘Analysing Networks: From Typologies of Institutions to Narratives of Beliefs’, Science and
Society, 10 Spring: 21–56.
science approach. I ask what networks will look like from an anti-foundational
perspective.
Anti-foundationalism, like social constructivism, supports four positions
that differ markedly from positivist social science (adapted from: Berger and
Luckman 1971; Gergen 1986; Rosenau, P. 1992: ch. 7).
• ‘External reality’: positivist social science tries to discover external reality;
anti-foundationalists ‘hold that there are no adequate means for repre-
senting it’ (and see Gergen 1986 for an extended discussion).
• A ‘constructivist theory of reality’: ‘To the extent that the mind furnishes
the categories of understanding, there are no real world objects of study
other than those inherent within the mental makeup of persons’ (Gergen
1986: 141).
• A ‘contextualist theory of reality’: ‘all knowledge claims (all facts, truths,
and validity) are “intelligible and debatable” only within their context,
paradigm, or “interpretative community”’ (citing Fish 1989: 141) . . . ‘Real-
ity is the result of social processes accepted as normal in a specific context’.
• Reality is a ‘linguistic convention’: ‘There are no independently identifi-
able, real world referents to which the language of social description is
cemented’ (Gergen 1986: 143).
‘Constructivist’ theories of the human sciences often suggest narratives are the
stuff of all the human sciences where narratives are ‘as much invented as
found’ so there is an ‘irreducible and inexpungable element of interpretation’
(White 1978: 51 and 82). For example, Collingwood (1939, 1993) argues that
historians ask questions and answer them with stories to make sense out of
‘facts’ that in their raw form make no sense at all. He summarizes his position
as follows:
history should be (a) . . . an answering of questions; (b) concerned with human
action in the past; (c) pursued by interpretation of evidence; and (d) for the sake
of human self-knowledge (1993: 10–11).
And Collingwood means knowledge is ‘Created, not discovered, because evi-
dence is not evidence until it makes something evident’ (Collingwood 1965:
99, italics in original). This approach does not mean there are no ‘facts’, only
that facts are constructed by the historian. The human sciences are construct-
ed and shaped by language, context, and the theories used. The resulting
interpretation is always incomplete, always open to challenge.2
2
On the constructivist theory of history, see, for example, Collingwood 1939, 1965, 1993;
Oakeshott 1983, 2004; White 1973, 1978, 1987. For a good introduction, see Jenkins 1995. For a
boisterous debate, see the exchange between Marwick 1995 and White 1995. For a vigorous
critique of the British empiricist view of history, see Skinner 2002 [1997].
Networks as Narratives of Beliefs and Practices 105
Such a conception of the human sciences contrasts markedly with the views
commonly found in political science where the influence of the natural science
models is great (Kavanagh 1991). The contrast between constructivist human
science and positivist social science is sharp. The latter strives after simplifi-
cation and successive approximations to a given truth. The former rejects all
such truth claims, accepting there are multiple realities and no given founda-
tions for asserting the superiority of one interpretation over another.
In this chapter, I adopt the anti-foundational position and argue for a
constructivist approach to the human sciences rooted in the notions of
tradition and narrative, an approach that keeps an ideal of objectivity (Bevir
and Rhodes 2003, 2006a). I discuss the interpretive approach in more detail in
Rhodes (2017, Volume II, Chapter 2). For present purposes, I simply need to
define my key terms: tradition, narrative, and dilemma.
A tradition is a web of inherited beliefs and practices forming the back-
ground against which people construct the world. Traditions are contingent,
constantly evolving, and necessarily located in a historical context. They are
handed on from generation to generation, whether from parent to child in
families or from elder to apprentice in organizations and networks. Tradition
is a starting point, not a destination. Traditions do not determine the beliefs
that people go on to adopt or the actions they go on to perform. They are
diverse. In any society there is a multiplicity of traditions. I adopt a pragmatic
notion of tradition. Investigators choose a particular tradition to explain
whatever set of beliefs or practices happen to be of interest to them. Traditions
are essentially artefacts. The justification for any choice of traditions lies in the
claim that they best answer the research question, in this case the changing
beliefs and practices in policy networks.
Narratives are the form theories take in the human sciences; they are to the
human sciences what theories are to the natural sciences. The point I want to
make by evoking narratives is that the human sciences do not offer us causal
explanations that evoke physically necessary relationships between phenom-
ena. Rather, they offer us stories about the past, present, and possible futures;
stories that relate beliefs, actions, and institutions to one another by bringing
the appropriate conditional and volitional connections to our attention.
Although narratives may follow a chronological order and contain such
elements as setting, character, actions, and happenings, their defining charac-
teristic is that they explain actions by reference to beliefs, desires, and other
pro-attitudes. The human sciences rely, therefore, on narrative structures akin
to those found in works of fiction. However, the stories told by the human
sciences are not fiction. The difference between the two lies not in the use of
narrative, but in the relationship of the narrative structures to our objective
knowledge of the world.
Crucially, this anti-foundational approach to the human sciences allows
for the possibility of judging competing narratives by agreed standards of
comparison. Objectivity arises from criticizing and comparing rival webs of

interpretation about agreed facts with established rules of intellectual honesty.
The key rules are accuracy and openness. Accuracy means using established
standards of evidence and reason; so we will prefer one narrative over another
if it is more accurate, comprehensive, and consistent. Openness means taking
criticism seriously and preferring positive speculative theories that open new
avenues of research and make new predictions supported by the facts. These
rules provide the criteria for comparing webs of interpretation or narratives.
The clear difference between this approach and conventional approaches to
studying government is that all interpretations are provisional. We cannot
appeal to a logic of vindication or refutation. Objectivity rests on criteria of
comparison. The web of interpretation we select will not be a web that reveals
itself as a given truth. Rather, we will select the ‘best’ interpretation by a
process of gradual comparison.
MODERNIST-EMPIRICISM AND POLICY NETWORKS
The interpretive and modernist-empiricist approaches to the study of net-

works differ markedly in four main ways. First, the social science approach
adopts a positivist epistemology, treating networks as social structures from
which we can read off the beliefs, interests, and actions of individuals. The
network to which an individual belongs allegedly establishes the content of his
or her beliefs and interests. In contrast an anti-foundational approach regards
networks as enacted by individuals. The beliefs and actions of individuals are
not determined by their ‘objective’ position in an organization or network.
Rather, their beliefs and actions construct the nature of the organization or
network. An anti-foundational approach, therefore, encourages us to decentre
networks. It encourages us to explore the ways in which they are made and
remade through the activities of particular individuals using, for example, the
tools of political ethnography.
Second, current explanations of network changes rely on exogenous, not
endogenous, causes. Thus, Marsh and Rhodes (1992a: 261) argue that net-
works create routines for policy-making and change is incremental. They
identify four broad categories of change: economic, ideological, knowledge,
and institutional, and all are external to the network. A decentred account
avoids such exogenous causes and starts with individual dilemmas. A dilemma
arises for an individual or an institution when a new idea stands in opposition to
an existing idea and so forces reconsideration. By focusing on the individual’s
responses to dilemmas, exogenous change is built into the heart of networks,
with change taking the form of confronting new beliefs and responding to the
actions of others.
Third, the network literature is characterized by typologies (see Chapter 3,

this volume). An anti-foundational approach does not treat network dimen-
sions and characteristics as given. It is a commonplace observation that even
simple objects are not given to us in pure perceptions but are constructed in
part by the theories we hold true of the world. Turning our attention to
complex political objects, the notion they are given to us as brute facts verges
on absurd. The ‘facts’ about networks are not ‘given’ to us but are constructed
by individuals in the stories they hand down to one another. The study of
networks, therefore, is inextricably bound up with interpreting the narratives
on which they are based.
The final characteristic of the network literature is that it is practical,
seeking to improve network management. Chapter 5 covers this topic in detail.
The social science model of networks treats them as given facts; as if they are
cars and the researcher is the car mechanic, finding the right tool to effect
repairs. An anti-foundational approach posits that networks cannot be
understood apart from traditions. The individuals whose beliefs, interests,
and actions constitute a network necessarily acquire the relevant interests
and beliefs against the background of traditions. In other words, there is no
essentialist account of a network but only the several stories of the participants
and observers. So, there can be no single toolkit for managing them. An anti-
foundational approach claims that practitioners learn by telling, listening to,
and comparing stories.
In short, an anti-foundational approach turns the current approaches to
networks on their head by insisting that networks are enacted by individuals
through the stories they tell one another and cannot be treated as given facts.
So, ‘Where do we go?’ How do we develop an anti-foundational approach to
an understanding of networks?
RECONSTRUCTING NETWORKS
I have summarized the current state of the policy networks debate. I have
offered my own criticisms of that literature. I now offer a way forward in the
analysis of networks and, therefore, of British government. I do so through the
notions of decentring, traditions, and dilemmas.3
3
The original framework was clearly rooted solidly in a positivistic social science epistemol-
ogy. However, it genuflected in the direction of the human sciences approach with its reference
to Sir Geoffrey Vickers and the idea of appreciative systems (Vickers 1968; ch. 4). The term refers
to that combination of factual and value judgements which describe the ‘state of the world or
reality’ (Rhodes 1981: 104). It is the individual decision-maker’s map of the world. Constructing
maps of how decision-makers make sense of the world is a defining characteristic of a decentred
approach to networks.
Decentring and the ‘Everyday Maker’
A decentred study of a network explores the way it is created, sustained, or

modified through the beliefs and actions of individuals. It decentres the
subject of research, representing a shift of topos from institution to individual.
Such decentred studies are essential because we cannot read off the beliefs
and actions of individuals from knowledge of objective social facts about
them. Although social discourses undoubtedly inform individual utterances,
individuals can still exercise their particular reason in given social contexts
(Bevir 1999).
Current approaches to policy networks focus on the oligopoly of the
political marketplace. They stress how networks limit participation in
the policy process; decide which issues will be included and excluded
from the policy agenda; shape the behaviour of actors through the rules of
the game; privilege certain interests; and substitute private government for
public accountability.
A decentred account of networks makes no such assumptions. It would
focus on the social construction of policy networks through the ability of
individuals to create meaning. Bang and Sørensen’s (1998 and 1999) story of
the ‘Everyday Maker’ provides an instructive example of a decentred account
of governance as networks focused on the beliefs and actions of individuals.
They interviewed 25 active citizens in the Nørrebro district of Copenhagen to
see how they engaged with government. They argue there is a long tradition of
networking in Denmark. They argue that Denmark has recently experienced
the conflicting trends of political decentralization through governmental
fragmentation, which has further blurred the boundaries between public,
private, and voluntary sectors; and political internationalization, which has
moved decision-making to the EU (Bang and Sørensen 1998: 11). They
describe this shift from Government to ‘governance networks’ as ideal typical
and suggest the governance of Denmark is a paradoxical mixture of Govern-
ment (hierarchy) and governance (networks).
In a system of governance, the ‘Everyday Maker’ focuses on immediate and
concrete policy problems at the lowest possible level. Civic engagement
is about:
balancing relations of autonomy and dependence between elites and lay-actors in

recursive, institutional networks of governance within or without the state or civil
society (Bang and Sørensen 1998: 3).
The ‘Everyday Maker’ has:

a strong self-relying and capable individuality; a perception of politics as the
concrete and direct handling of differences, diversity and dispute in everyday life;
a notion of commonality as relating to solving common concerns; an acceptance
of certain democratic values and procedures in the handling not only of high but
also of low politics (Bang and Sørensen 1998: 3).
Thus, Grethe (a grass-roots activist) reflects that she has acquired the compe-
tence to act out various roles: contractor, board member, leader. There has been
an ‘explosion’ of ‘issue networks, policy communities, ad hoc policy projects, and
user boards, including actors from “within”, “without”, “above”, and “below”
government’. So, the task of the ‘Everyday Maker’ is ‘to enter in and do work at
one point of entry or another’ (Bang and Sørensen 1998: 15). Political activity
has shifted from ‘formal organising to more informal networking’ (Bang
and Sørensen 1998: 20). And amidst these networks ‘You do in fact miss local
government—a visible local government. They become visible at once when
there are hullabaloos . . . in ordinary everyday life, they are conspicuous by their
absence’ (Bang and Sørensen 1998: 21). Politics is no longer about left and right
but engaging in what is going on in institutions (Bang and Sørensen 1998: 23).
In short, Bang and Sørensen draw a picture of Nørrebro’s networks through the
eyes of its political activists.
There are some instructive contrasts between Bang and Sørensen’s research
and a decentred, anti-foundational account. First, they employ an ideal-
typical research method, specifying not only the characteristics of the
‘Everyday Maker’ but also the maxims that guide their political behaviour.
Specific instances are then compared with these ideal-typical formulations.
A decentred account would not assume the ‘Everyday Maker’ had these
characteristics. Rather, for example, it might use such ethnographic tools as:
studying individual behaviour in everyday contexts; gathering data from many
sources; adopting an ‘unstructured’ approach (that is, collecting data in a raw
form not to a preconceived plan); focusing on one group or locale; and, in
analysing the data, stressing the ‘interpretation of the meanings and functions
of human action’ (paraphrased from Hammersley 1990: 1–2; see also Geertz
1973: 20–1).4
Second, the ‘Everyday Maker’ is a normative ideal. Her behaviour epitom-
izes civic engagement in Denmark. A note of caution is in order. The ‘Every-
day Maker’ may be an endangered species. Jensen (1998) shows how the
democratic experiment in Danish social housing is confounded by the fatalism
of tenants and the lack of suitable democratic skills. Normative ideals could
lead the researcher to ignore the fatalist for whom networks will have a
different meaning.
Third, Bang and Sørensen’s account of networks focuses on the beliefs and
actions of only one group of actors and does not provide a ‘thick description’
(Geertz 1973: ch. 1). A decentred account implies a micro-analysis but does
4
For a similar recognition that the political ethnography of networks is an instructive
approach, see Heclo and Wildavsky 1974; McPherson and Raab 1988; and Rhodes 1997a: ch. 9.
not imply necessarily a bottom-up approach. The analysis is not restricted

to any one category of actor. So, to the ‘Everyday Maker’, we need to add: the
street-level bureaucrats, who can make and remake policy; services users,
whose experiences can differ markedly from the expectations of the service
provider; and the beliefs and actions of the political and managerial elite who
seek to steer other actors in the network. Decentred studies of networks would
build a multifaceted picture of how the several actors understand them. There
is no expectation there will be the one ‘true’ account. The researcher con-
structs stories about how other people understand what they are doing in
networks. These stories will conflict but overlap. They will be built out of the
several organizational, network, and political traditions that actors have learnt
and constructed as they enact and remake networks in their everyday lives.
The researcher constructs and compares webs of meaning.
Finally, Bang and Sørensen note but do not explore the traditions of
governance as networks that shape the ‘Everyday Maker’, and indeed other
actors in the networks, a point I will now return to.5
Traditions and Narratives
One popular social science explanation for the growth of governance posits
that advanced industrial societies grow by a process of functional and institu-
tional specialization and the fragmentation of policies and politics (Rhodes
1988: 371–87). For some authors, differentiation is part of a larger context. For
example, regulation theory sees it as an outcome of the shift from Fordism to
post-Fordism (see also Jessop 1997: 308–15: Stoker 1998: 126–7 and 1999b). In
contrast an anti-foundational approach stresses how different governmental
traditions understand and respond to governance as networks. Networks are
understood through traditions. In addition, networks construct or reconstruct
their own traditions. Individuals learn about the network and its constituent
organizations through stories of famous events and characters. Traditions are
passed on from person to person. They are learnt. Much will be taken for
granted as common sense. Some will be challenged; for example, when beliefs
collide and have to be changed or reconciled. The several traditions will
produce different stories, which we will compare. We may prefer one story
5
See also Bang and Sørensen 2001 and Bang, 2005. Wagenaar (2012: 97) agrees with me that
ethnographic analysis should be located in its broader historical context—traditions. I agree with
Wagenaar that Bang and Sørensen are a good example of decentred analysis. However, I do not
agree with his suggestion that everyday makers ‘operate outside political belief systems’. The
everyday maker was formed against the backcloth of both decentralized local public institutions
and Noerrebro’s tradition of left-wing parties and grass-roots social movements. Their identity is
defined in relation to those two traditions as Bang (2005: 166–70) makes clear. They have distinct
and distinctive beliefs and practices but they are not ‘outside’ their inherited context.
to another because it is more accurate and open. But that story will still be
provisional.
One way of illustrating this approach would be to explore the traditions and
narratives that inspire political actors. In this way I could show how govern-
ance as networks arises out of the multiple narratives that legislators, bureau-
crats, and others have come to adopt through a process of modifying
traditions to meet specific dilemmas. However, because I do not know their
relevant stories, I will fall back on academic accounts of the rise and nature of
governance as networks, showing how these accounts reflect different govern-
mental traditions.
Governance as networks is a narrative interpreted through traditions and
in Britain it is possible to identify several traditions; for example, Tory, Liberal,
Whig, and Socialist (Bevir and Rhodes 2003). Here I illustrate the argument by
looking at the New Right and the New Labour traditions, both of which exercise
a powerful influence on how we currently understand British government.
Henney (1984: 380–1) writes in the liberal tradition. He sees governance as
networks as an example of the corporate state; ‘the institutionalised exercise of
political and economic power’ by the various types of local authority, govern-
ment, the unions, and to a lesser extent business. They ‘undertake deals when
it suits them; blame each other when it suits them; and cover up for each other
when it suits them’. These interactions are conducted ‘behind closed doors’
and each network builds a ‘cultural cocoon’ rationalizing their interests with
the public interest. They ‘institutionalise irresponsibility’. Producer interests
rule OK, only for Henney it isn’t, and he wants to cut local government down
to a manageable size by removing some functions and transferring others to
the social market. But the problem of networks as producer capture is not so
easily resolved. Marketization is the alleged solution but it fragments service
delivery structures, creates the motive for actors (individuals and organiza-
tions) to cooperate and, therefore, multiplies the networks and opportunities
for producer capture that Henney’s reforms seek to counter. Beliefs in the
virtues of markets have to confront the defects of quasi-markets and resilience
of networks.
The socialist tradition in the guise of New Labour sees governance as
networks as a problem of integration. For Perri 6 (1997) government confronts
‘wicked problems’ that do not fit in with functional government based on
central departments and their associated policy networks. Such functional
government is costly, centralized, short-term, focuses on cure not prevention,
lacks coordination, measures the wrong things, and is accountable to the wrong
people (Perri 6 1997: 26). The solution is holistic government, which will span
departmental cages. The twelve recommendations include: holistic budgets
designed around outcomes, not functions; cross-functional outcome measures;
integrated information systems (for example, one-stop shops); and culture,
value for money, and preventive audits (Perri 6 1997: 10–12 and chs 4–7).
This report epitomizes the long-standing Fabian tradition in the Labour

Party, which sees salvation in administrative engineering. But again the
problem of network integration is not so easily resolved. Perri 6’s proposed
reforms have a centralizing thrust. They aim to coordinate the departmental
cages, a centralizing measure, and to impose a new style of management on
other agencies, a central command operating code. But network structures
need a decentralized, diplomatic, negotiating style. Beliefs in ‘leaders know
best’ confront the belief that decentralized structures need indirect or hands-
off management.
Governance as networks has important implications for other state tradi-
tions. Loughlin and Peters (1997: 46) distinguish between the Anglo-Saxon
(no state) tradition; the Germanic (organic) tradition; the French (Jacobin)
tradition; and the Scandinavian tradition, which mixes the Anglo-Saxon and
Germanic. Thus, in the Germanic tradition state and civil society are part of
one organic whole; the state is a transcendent entity. Its defining characteristic
is that it is a rechtsstaat; that is, a legal state vested with exceptional authority
but constrained by its own laws. Civil servants are not just public employees
but personifications of state authority. The Anglo-Saxon tradition draws a
clearer boundary between state and civil society; there is no legal basis to the
state and civil servants have no constitutional position. The Jacobin tradition
sees the French state as the one and indivisible republic, exercising strong
central authority to contain the antagonistic relations between state and
civil society. The Scandinavian tradition is also ‘organic’, characterized by
rechtsstaat, but differs from the Germanic tradition in being a decentralized
unitary state with a strong participation ethic. (This paragraph is a précis of
Loughlin and Peters 1997: 46–55.) These traditions underpin different inter-
pretations of governance as networks.
To return to the Bang and Sørensen (1998) example, local networks with
high participation are a long-standing feature of the Danish governmental
tradition. The Nørrebro district of Copenhagen is famous even in this trad-
ition for its activism, which extended to (inconceivable) street riots over
Danish membership of the EU. So, the ‘Everyday Maker’ acts in national
and local traditions characterized by beliefs in political activism. So, for the
Danish tradition with its participation ethic governance as networks poses the
issue of how to keep the multiplying networks under democratic control.
In the Germanic tradition, the legal framework sets the boundaries to, and
guides, official action. The direct imposition of control is unnecessary. There is
a high degree of tolerance for the multi-level networks (politikverflechtung) so
common in federal systems. On the other hand, the Jacobin tradition with its
assumption of conflict between state and civil society sees networks as a
potential threat to state authority unless subject to state control, for example
through strong mayoral leadership. In other words, in seeking to interpret and
understand governance as networks, we have to ask whose interpretation, in
which tradition. Moreover, I have illustrated an argument. As Loughlin and

Peters (1997: 60) are the first to admit, this account of state traditions is broad
brush. Traditions do not exist as, for example, ideal types to which we
compare specific instances. A more thorough account must cover the variety
and nuances of traditions as learnt. Nonetheless, I have illustrated how
traditions shape our understanding of governance as networks, both nation-
ally and cross-nationally (see also Dyson 1980).
DILEMMAS AND THE ANALYSIS OF CHANGE
As noted earlier, a dilemma arises for an individual when a new idea stands in
opposition to an existing idea and so forces reconsideration. Because we
cannot read off the ideas and actions of individuals from objective social
facts about them, we can understand how their beliefs, actions, and social
practices change only by exploring the ways in which they think about, and
respond to, dilemmas. Thus, an analysis of change and developments in
government must take place through a study of relevant dilemmas. I build
change into the heart of my account of networks by exploring how individual
actors respond to dilemmas and reinterpret and reconstruct traditions.
Stoker’s (1999c) analysis of the new public management (NPM) in British
local government shows how dilemmas stemming from inflation and chan-
ging beliefs about public spending led to a new story, not about NPM, but
about local governance, illustrating people’s contingent responses to dilem-
mas. Ideally, of course, I should tell the story through the eyes of public
managers but their version of the story is not available. So, instead I use
Stoker’s accounts of how public managers responded to the dilemma of
inflation and reduced public spending; that is, academic ‘constructions of
other people’s constructions of what they are up to’ (Geertz 1973: 9).6
Inflation had become a major problem for the British economy by the end
of the 1970s and it was widely accepted that: the key monetary levers should
be interest rates rather than fiscal policy; the supply side of the economy
should be considered more significant than demand management; low infla-
tion should be as important a goal of economic policy as low unemployment;
6
Again, I simply illustrate the argument that the notion of dilemma helps us to understand
change. I do not provide a detailed exploration of change in networks. Any such account would
need to recognize that individuals have several antidotes to, and coping mechanisms for,
challenges to their belief systems. Such challenges can take the form of responding to different
beliefs or to the actions of others and any response will be affected by the salience of those beliefs
and actions for the several parties.
and government should develop monetary policy in accord with rule, not
discretion, to preserve credibility. These neo-liberal beliefs had direct and
immediate consequences for public spending; it was to be cut.
Local authorities are a major vehicle for delivering welfare state services and
account for much public spending. They are thus a prime target for any
government committed to low inflation and the attendant curbs on public
spending. Management reform was one part of the effort to contain public
spending. The new public management’s rhetoric told a story of economy,
efficiency, and effectiveness—the ‘3Es’—which contrasted sharply with the
story of the local government officer as professional with clients. In theory, the
‘3Es’ would deliver more public services for less money. There was a second
strand to NPM: marketization. This term refers to the use of market mechan-
isms in the delivery of public services, covering contracting out (for example,
compulsory competitive tendering of many local government services); quasi-
markets in the guise of the purchaser–provider split (for example, in the
National Health Service (NHS)); and experiments with voucher schemes (for
example, nursery education).
Neo-liberalism in the guises of the ‘3Es’ and marketization generated
unintended consequences. Thus, Stoker (1999a) identifies several, negative
unintended consequences, including fragmentation, loss of accountability, and
a decline in the public service ethic. More significant for the argument here, he
also identifies important unintended benefits. First, NPM disrupted the sys-
tem. Second, local authorities were increasingly forced to account for their
actions in public. Third, these twin pressures produced a sense of crisis, which
helped to create new policy ideas. The delicious irony is that the new ideas
were not those of NPM but of local governance. So, local authorities adopted a
wider role of concern for the well-being of the locality, worked in partnership
with many actors and agencies, and focused on the outcomes of services
delivered through the partnerships. As Stoker (1999a: 15) concludes:
It is in some respects ironic that the pressures unleashed by new management
have encouraged local authorities to rethink and redefine their role. The vision of
the new management reformers aimed at a more efficient and customer-oriented
service delivery by local authorities has been challenged by a broader vision of a
new community governance.
So, the number of networks multiplied. Their membership grew and it was
drawn from more sectors. By both intent and as an unintended consequence
of reform, the capacity of the centre to steer those networks declined (Rhodes
1997a: 12 and 45). The response to the dilemma posed by inflation and public
spending cuts can be seen in the evolving managerial story about central
government reform.
The neo-liberal story began with the new public management (NPM); a set
of inherited beliefs about how private sector management techniques would
increase the economy, efficiency, and effectiveness—the 3Es—of the public

sector. The emphasis fell on hands-on, professional management; explicit
standards and measures of performance; managing by results; and value for
money. Subsequently, the story evolved to embrace beliefs about competition
and markets. Initially, the privatization of public utilities lay at the heart of the
reforming agenda. Latterly, the emphasis switched to ideas about restructuring
the incentive structures of public service provision through contracting out,
quasi-markets, and consumer choice. New Labour gave the story a distinct and
distinctive twist with its focus on service delivery, consumer choice, and
joined-up government. The earlier stands of managerialism had their roots
in management theory and neo-classical economics. This strand drew on
different types of social science, mainly new institutionalism and communi-
tarianism (Bevir 2005: chs 2 and 3). Managerialism was now the ‘delivery
agenda’, which begat a new blueprint in response to the global financial crisis.
The austerity narrative is again about cutting public expenditure, but at its
heart is that age-old neo-liberal ambition for the minimal state. It involves a
concentration of power in the hands of the Minister and the central depart-
ment. Intermediate institutions are abolished or bypassed for various forms of
markets. Services are outsourced or privatized. Managers are empowered as
‘technicians of transformation’. The new quasi-market is subject to regulation
by a performance measurement regime that is also intended to foster choice by
giving parents more information.7
These changes in the management story illustrate the contingent nature of
the way people responded to the dilemma posed by the need to curb public
spending. There is no objective or rational or necessary reason for NPM to
evolve into the new blueprint for centralization and privatization. It is the
response of individuals to the dilemmas they encountered informed by the
ideas of modernist social science, notably neo-liberalism.
CONCLUSIONS
I have defended an anti-foundational approach to the study of networks and

the propositions that facts are not given, and beliefs and actions cannot be read
off from knowledge of so-called objective social facts. I have shown how
decentring, traditions, and dilemmas can be used to understand networks in
governance.
The social science literature on governance as networks identifies and
focuses on key changes in government and it poses distinctive, new questions
7
This new paragraph is an up-to-date illustration of the same argument in the original
version of the paper. It is taken from Bevir and Rhodes 2016.
about government, for example about reshaping the state. However, an anti-
foundational approach to studying governance as networks teaches important,
additional lessons for both academics and practitioners.
For academics, I argue there is no essentialist account of networks that can
be used either to produce law-like generalizations or to legitimate advice to
policy-makers. Second, the road to understanding lies in decentred accounts
focusing on the political ethnography of networks and on narratives that give
due recognition to the creative individual, not the techniques of positivist
social science. Networks are constructed by individual actors and not created
by governments or imposed by the researcher. As researchers, we write
constructions about how other people construct the world.
For practitioners, the key lesson of an anti-foundational approach is that
there is no single toolkit they can use to steer networks. However, they can
learn by listening to and telling stories. The social sciences offer only provi-
sional knowledge but an awareness of our limits does not render the human
sciences useless. If we cannot offer solutions, we can define and redefine
problems in novel ways. We can tell the policy-makers and administrators
distinctive stories about their world and how it is governed (see, for example,
Chapter 12, this volume). The language of networks challenges the language of
managerialism, markets, and contracts. The language of narratives challenges
the language of predictive social science.
This chapter provides a language for re-describing the world of networks;
for understanding how several actors construct and reconstruct the meaning
of networks when faced with government reforms. In particular, it challenges
the dominant, managerial discourse about networks. Too often the analysis of
networks is reduced to managerial skills. In no way do I wish to suggest that
learning how to steer networks is unimportant. I do want to suggest, however,
that steering networks is about understanding participants’ stories as much as
more technical means. The analysis of governance as networks needs a
decentred exploration of traditions and dilemmas. I continue this exploration
in Rhodes 2017, Volume II. Now, I turn from policy networks to the related
notion of governance and my exploration of the changing nature of British
governance.
Part II
Governance
8
The Hollowing Out of the State
Administrative reform breeds more cynicism than efficiency and effectiveness.

Anyone familiar with the reforms of the late 1960s and early 1970s will agree.
The fates of the Fulton Committee on the reform of the civil service, the Maud
Committee on the management of local government, the Redcliffe-Maud
report on the reorganization of local government, the ‘Grey Book’ on the
management arrangements for the reorganized NHS, and the reorganization
of the NHS structure show that aims and achievements often diverge mark-
edly. Any practitioner who lived through these changes will remember the
uncertainty and the long hours of work but will be hard-pressed to identify
tangible benefits. The simple fact that there has been a continuous stream of
reforms clearly shows the reforms ‘underachieved’. This chapter takes a
jaundiced view of trends in public sector reform in the 1980s and 1990s.1
I will spare the reader yet another potted history of the period. The chapter is
avowedly speculative. Inevitably, it will be wrong about some trends and some
outcomes. Nonetheless, it is instructive to survey the broader picture.
My aims are clear from the chapter’s title. The phrase ‘the hollowing out of
the state’ suggests the British state is being eroded or eaten away. I freely admit
the term is controversial and it may even be inaccurate. Please note, however,
I refer to processes that contribute to a hollowing out of the state and I do not
suggest the era of the hollow state has arrived. I will try to show it is a revealing
expression that teaches us something about what is happening to British
public administration. I use the phrase to cover four interrelated trends.
(1) Privatization and limiting the scope and forms of public intervention.
(2) The loss of functions by central government departments to alternative
service delivery systems (such as agencies).
1
This chapter originally appeared as R. A. W. Rhodes (1994) ‘The Hollowing Out of the State’,
Political Quarterly, 65: 138–51. Guy Peters (University of Pittsburgh) first drew my attention to the
phrase ‘the hollow state’ in his (then) unpublished, undated paper entitled ‘Managing the Hollow
State’ (see Peters 1994). Reprinted with permission of John Wiley and Sons.
(3) The loss of functions by British government to EU institutions.

(4) Limiting the discretion of public servants through the new public
management (NPM), with its emphasis on managerial accountability,
and clearer political control through a sharper distinction between
politics and administration.
‘Hollowing out’ is not restricted to central government, or even to Britain.
I draw examples from the UK public sector, not just the civil service. After
describing and illustrating each trend, I discuss possible problems under the
headings of fragmentation, accountability, catastrophe, and central capability.
Any social scientist with the temerity to predict has a death-wish but, hopeful
of longevity, I suggest the year 2000 may witness a return to bureaucracy to
restore the diminished central capability of British government!
PRIVATIZATION AND REDEFINING

PUBLIC INTERVENTION
The 1980s bred its own clichés of which ‘Thatcherism’ ending the ‘post-war
consensus’ is one of the more common. The terms of this debate are less than
clear. Few can agree on the contents of either Thatcherism or the post-war
consensus, let alone what changed and by how much. But clichés get this
status because, once, they did say something important. In this case, it draws
attention to important questions. Does this task need to be done? Does it need
to be done by government? If not, who should do it? In other words, the
Conservative government challenged conventional views about the scope of
public sector activity and it sought to reduce the size of the public sector.
There are several ways of measuring changes in the size of the public sector
and the extent of government intervention. Three indicators of particular
relevance to the study of public administration include: public spending
(both in real terms and as a proportion of GDP); the public ownership of
industry; and levels of government employment.
Public spending rose in real terms throughout the 1980s and 1990s.
Whether it accounted for a greater or lesser proportion of GDP depended,
therefore, on the economy growing more quickly than the growth in public
expenditure. In 1979, general government expenditure accounted for 43.3 per
cent of GDP. It remained higher until 1986 when it fell to 42.8 per cent,
reaching a low of 39.5 per cent in 1988, rising to 42 per cent in 1992. The
average for the 1980s was 43.5 per cent. There was no significant cut, therefore,
although the fortunes of individual services within this total varied.
The story of privatization, or the sale of the assets of government-owned
enterprises to the private sector, is well known. The scale of the programme is
The Hollowing Out of the State 121
impressive. Since 1979, over 50 per cent of the public sector, along with
650,000 employees, was transferred to the private sector. The nationalized
industries had accounted for 9 per cent of GDP. By 1991, the figure had fallen
to less than 5 per cent and continues to fall.
There was an equally dramatic cut in the civil service. The total number of
industrial and non-industrial civil servants fell by some 24 per cent between
1979 and 1992, from 732,000 to 554,000. There was also a 38-per-cent decrease
in the staff employed in non-departmental public bodies, although expenditure
rose. The fall in numbers in other parts of the public sector is less sharp.
Employment in local government remained constant. In the National Health
Service, employment rose by 7 per cent between 1979 and 1983, and fell by 4 per
cent between 1984 and 1991. In sum, the 1980s saw a notable reduction in
public employment for a significant proportion of the public sector.
As important was the changed attitude towards public intervention that lay
behind these figures. The Conservative government rejected the ‘centralizing,
managerial, bureaucratic, interventionist style of government’. Government
had ‘to get out of the business of telling people what their ambitions should be
and how exactly to realise them’. ‘Optimism about the beneficent effects of
government intervention had largely disappeared.’ Government had ‘to put its
faith in freedom and free markets, limited government and strong national
defence’; in ‘the creative capacity of enterprise’.2
In sum, there is a long-term policy of reducing the size of the public sector
and, using several indicators, there is evidence of some success.
ALTERNATIVE S ERVICE DELIVERY SYSTEMS

A N D AG E N C I E S
Students of public policy-making can no longer limit debate to the hoary old
dichotomy between planning and markets. Life is no longer so simple. If you
believe the international best-seller, Reinventing Government, there are 36
service delivery options (Osborne and Gaebler 1992: Appendix). British gov-
ernment already uses a significant proportion of them and, briefly, I will
describe some recent experiments.
Contracting out, or market testing, is a long-standing feature of the British
public sector. It was introduced by Harold Wilson’s Labour government in 1968
when private firms were brought in to clean government departments, produ-
cing an estimated saving of 35,000 jobs and £500,000 a year. Subsequently, the
2
All the quotations in this paragraph are from: Thatcher 1993: 6, 14, 92, 15, 92, and 45–6.
See also the summary in Kavanagh 1990: 11–12.
1980s Conservative government of Margaret Thatcher made it compulsory for

many services in the National Health Service and local government. Walsh
estimates that the first round of compulsory contracting out under the Local
Government Act (1988) led to cuts in manual staffing levels in 57 per cent of the
contracts let. The cuts were large for refuse collection (16.9 per cent), street
cleaning (25.7 per cent), and building cleaning (18.5 per cent). He also estimates
there was an overall reduction in costs of about 6–7 per cent (Walsh 1991: 6 and
chs 12 and 13).
Contracting out is an established feature of the administrative landscape; we
already live in the contract state. This trend is challenging conventional
notions of local government. Clarke and Stewart (1991: 17–18) argue that the
British local government system was built on the assumption of self-sufficiency;
that is, it not only provided services but it also employed the staff and owned the
resources to do so. They suggest that local government is increasingly adopting
an enabling role; that is, it delivers services through other public and private
sector organizations. The role of local government is to facilitate the delivery of
services by others and oversee their performance.
Contracting for services is a distinguishing characteristic of the new rela-
tionship between the chief executive of an agency and his or her central
department. It is also a central feature of the experiments with quasi-markets;
for example, the purchaser–provider split in the National Health Service, and
the mixed economy of care for social services.
The Next Steps programme was the government’s response to the loss of
impetus in its management reforms of the 1980s. The Efficiency Unit pro-
posed that agencies should ‘carry out the executive functions of government
within a policy and resources framework set by a department’ (Efficiency Unit
1988: 9). By the end of 1992, there were 92 agencies employing 60 per cent of
the civil service. This figure is expected to rise to 75 per cent by the mid-1990s
(Mottram 1993 and Cm 2430 1993). Each agency had a framework document
setting out the contractual responsibilities of each side as well as an annual
business plan and a five-year corporate plan. In effect, and by intent, agencies
distance ministers and top civil servants from operational matters, leaving
them free to concentrate on policy.
The purchaser–provider split was the controversial flagship of the govern-
ment’s plans to reform the NHS. The NHS and Community Care Act (1990)
introduced a quasi-market in health by creating competition between hos-
pitals and other service providers. Thus, health authorities now buy services
from providers in the public, private, and voluntary sectors. Hospitals are one
obvious provider of services. They can opt out of health authority control and
become self-governing NHS Trusts. Also, general practitioners (GPs) can
buy services from hospitals. The underlying idea is that money follows pa-
tients in the same way it follows consumers. The resulting competition will not
only drive down costs but also provide patients with the services they want.
Although slow at first, implementation gathered impetus after the 1992
General Election (see Harrison 1994; Harrison and Pollitt 1994: ch. 6). The
same principles underpin the mixed economy of care that extends the
purchaser–provider split and the quasi-market principle to providing social
services (Wistow et al. 1994).
The effect of all these changes is to disaggregate public bureaucracies. In
place of the line bureaucracy delivering any service all over the country, there
is now a patchwork quilt of organizations. The conclusion of the Next Steps
report that ‘the Civil Service is too big and too diverse to manage as a single
entity’ has been generalized to the public sector (Efficiency Unit 1988: 4).
THE E UROPEANIZATION O F EV ERYTHING
The impact of the EU on British government is an emotive topic. Whether

the impact is judged good or bad is a separate issue not discussed here. The
key point is that there will be at least three major consequences for British
public administration in the 1990s: the erosion of sovereignty, the sectoraliza-
tion of policy-making, and the effects of the constitutional principle of
subsidiarity.
To be blunt, Britain lost sovereignty when it joined the EU to a far greater
extent than anticipated in 1972. For example, EU legislation takes primacy
over national legislation and the role of Parliament was undermined and its
control over EU legislation is, at best, weak. As the EU’s influence extends
post-Maastricht, there can be little doubt the British government will continue
to be an ‘awkward partner’ as it chafes at EU constraints whether on foreign
policy or the quality of bathing water.
The EU also added a new layer of government. Most distinctive is the
emergence of transnational policy networks. Such networks lead to routinized,
sectoral policy-making:
At the institutional level, the community is unequivocally supported by the self-
interest of the vertical alliances of policy specialists—interest associations, national
ministries and parliamentary committees, and the large contingent of specialised
lobbyists, bureaucrats and politicians operating at the European level. They all
profit from the availability of additional resources, and of additional points of
access to political decision processes, providing additional opportunities for play-
ing the game of influence and obstruction that is their raison d’etre
(Scharpf 1988: 270; see also Chapter 5, this volume).
Although the activities of these transnational networks are poorly documented,
they are a significant feature of EU policy-making. They also include local
and regional governments, thereby creating the opportunity for multi-level
bargaining games in which both the Commission and local authorities

have incentives to bypass national government. The consequences are an in-
creasingly complex policy process in which the vertical alliances of professional-
bureaucratic interests are accountable to no one and the problems of horizontal
coordination become ever more pressing (see Chapter 4, this volume).
And the mention of local government raises the translucent spectre of
‘subsidiarity’ or the constitutional principle that:
the responsibility for carrying out tasks should be held at the lowest level of
government competent to undertake them, and that where necessary higher
authorities should give support to enable them to fulfil the responsibilities that
are appropriately theirs under this doctrine (Norton 1991: 27).
Unfortunately, this definition makes the conceptual fog denser. It leaves three
questions unresolved. Who decides which government is competent? Which
responsibilities are ‘appropriate’? What are the criteria for these decisions?
There are no easy answers. The British government sees the principle as a
defence against an intrusive Commission. British local government sees it as
a defence against a centralizing central government. The Commission sees it
as a means for bypassing stubborn national governments to deal with more
pliant regional and local governments. These conflicting interests will fuel
debate about the proper distribution of functions between the several tiers of
government and will make any effort to redistribute functions highly charged.
However, the new Committee of the Regions and the Commission’s commit-
ment to reduce regional disparities seem certain to involve a greater role for
local and regional authorities and, as a result, increased conflict with national
governments.
THE NEW PUBLIC MANAGEMENT
The new public management (NPM) does not refer to any one idea but to the
currently fashionable set of ideas driving administrative reform. Indeed, it can
be defined so broadly that it covers all the topics discussed here. Some precision
is necessary and provided by Christopher Hood’s (1991: 4–5) discussion of
the seven components in the NPM doctrine: hands-on professional manage-
ment; explicit standards and measures of performance; greater emphasis on
output controls; disaggregation of public sector units; greater competition in
the public sector; stress on private sector styles of management; and greater
discipline and parsimony in resource use. He also suggests the origins of these
components lie in either the ‘new institutional economics’ or business-type
managerialism. I discussed the topics closely associated with the new institutional
economics—for example, disaggregating public bureaucracies, competition, and
quasi-markets—under the heading ‘Alternative Service Delivery Systems’.

Here, I concentrate on business-type managerialism; on the micro-level of the
roles and relationships of political administrators, whether official or elected.
The first thrust of managerial change in the public sector is captured by the
aphorism, the ‘3Es’; or economy, efficiency, and effectiveness. The key phrases
are ‘value for money’ and ‘better use of resources’. It is not necessary to tell
again the story of the efficiency scrutinies, MINIS, FMI, and Next Steps.3 More
important, the reforms were based on ‘an impoverished concept of manage-
ment’: that is, management is about setting clear objectives, and exercising
hierarchical control and coordination. Metcalfe and Richards (1991: 16–17)
comment that such reforms ‘would drag British government kicking and
screaming back into the 1950s’.
These management reforms were based on the conviction that the public
sector must emulate private sector management. However, as Ranson and
Stewart (1989) argue, just as the job of management varies in the private
sector, so there are distinctive tasks in the public domain and there are also
distinctive purposes and conditions. For example, a defining characteristic of
the public sector is the determination of collective values out of the mosaic of
conflicting interests, a process that extends beyond such values as managerial
efficiency to encompass equity and justice. NPM does not always recognize the
distinctive tasks, purposes, and conditions of public sector management.
There has been a narrow focus on efficiency at the expense of, for example,
broader notions of public accountability. The danger was recognized by
William Waldegrave when Secretary of State for Health:
Without remitting for one moment the pressure to get a better management
system, borrowing what is useful from business, let us watch our language a bit.
It just bears saying straight out: the NHS is not a business; it is a public service and
a great one (Waldegrave 1990, cited in Stewart and Walsh 1992: 516).
Management controls restrict civil service discretion and hollow out their jobs
from below. There are also signs that it is being hollowed out from above.
During the 1980s we lived in the era of ‘macho-ministers’. Fears were
expressed about the politicization of the civil service: ‘a thoroughly Thatch-
erised satrapy’ (Hennessy 1989: 623–35).
3
MINIS refers to Management Information System for Ministers, the financial management
system introduced by Michael Heseltine when he was Secretary of State for the Environment.
FMI refers to the Financial Management Initiative and was an early precursor of performance
management. Next Steps refers to hiving off sections of government departments as separate
executive agencies. The story of these initiatives is told entertainingly in Hennessy (1989: ch. 14).
The evidence for the party politicization of the civil service is weak. It is now
easier to recruit outsiders, with many more top jobs open to advertisement
and competition, and some view this trend with alarm. A sharp distinction
was drawn between policy (the strategic function of the minister) and its
administration (the operational function of the civil servant). Indeed, agencies
institutionalized the distinction. Sir Humphrey Appleby gave way to the ‘can
do’ civil servant.
The concept of management broadened as we entered the 1990s and the
era of the Citizen’s Charter. Sir Robin Butler, Head of the Home Civil Service,
describes the Citizen’s Charter as ‘the culmination of the movement to output
measurement’. Citizens are told what standard of service they can expect and
offered redress if it is not forthcoming. The consumers’ interests dominate the
providers’ interests: ‘people power’ (Butler 1993: 402). It is a little early for
such eulogies. Christopher Pollitt adopts a more measured tone when he
comments it ‘is not so much a charter for citizen empowerment as manager-
ialism with a human face’ (Pollitt 1993: 187).
Managerial control reduces civil service discretion. The Citizen’s Charter
requires civil servants to satisfy consumers. The analogy with the private
sector erodes the broader values of the public service ethos. The politics–
administration distinction and macho-ministers erode the contribution to
policy-making. In effect, the job of the individual civil servant is hollowed
out from above and below.
FROM SOLUTIONS TO PROBLEMS
There are no solutions to problems because, as Aaron Wildavsky (1980: 62)

noted in one those aphorisms at which he was so adept, ‘policy is its own
cause’. Each new solution breeds the next generation of problems. The trend
towards a hollow state in Britain will be no exception. This section discusses
several important, unintended outcomes. It does not discuss the likely conse-
quences of using individual policy tools; for example, I do not discuss the
strengths and weaknesses of contracting out. I concentrate on the conse-
quences of hollowing out the state.
Fragmentation
The experiment with alternative service delivery systems is an exciting one,

provided it is an experiment. A policy experiment should involve systematic
learning; that is, it must generate data so the policy-maker can identify and
correct errors. The current government programme is not so designed, an
omission of some importance because some unwelcome outcomes are in
the offing.
The inevitable corollary of the new systems is institutional fragmentation.
So, service delivery will depend increasingly on linking sets of organizations.
For example, the mixed economy of care in the social services involves
cooperation between numerous organizations in the public, private, and
voluntary sectors. In other words, organizational interdependence is ubiqui-
tous and steering complex sets of interdependent organizations will become
increasingly difficult for governments. As Hesse (1991: 619) notes:
advocates of decentralised self-guidance and control often fail to realise that
highly differentiated societies and pluralistic, fragmented institutional systems
create a growing need for collective steering, planning and consensus building.
Government is aware of this danger. Seminars and workshops puzzle over

‘strategic management’. An anecdote under Chatham House rules: ‘We do do
strategic management’, insisted the Permanent Secretary, ‘we pass review
papers to the minister and he sends them out again’. Or, in the words of
another senior official, ‘the medium-term is driven out by the short-term
demand to cut costs’. The same awareness exists for local government. Thus,
the idea of ‘the enabling authority’ posits an overarching strategic and coord-
inating role for the local authority. Awareness is not action, however, and the
drive for short-term cuts remains irresistible.
Fragmentation not only weakens coordination, it also reduces efficiency.
Fragmentation leads to functional and jurisdictional overlap, otherwise known
as duplication and waste, thereby increasing inefficiency. For example, the new
Trust hospitals are short of skilled personnel (nursing, ancillary, and medical).
So, the Trusts will pay, and are already paying, higher salaries to attract the
necessary staff. The resulting competition between the Trusts will drive up
labour costs, an outcome that the Conservative government, among others,
will see as increasing inefficiency.
Parenthetically, duplication and overlap can increase effectiveness. Landau
(1969: 356) identifies the benefits of redundancy:
It provides safety factors, permits flexible responses to anomalous situations
and provides a creative potential for those who are able to see it. If there is no
duplication, if there is no overlap, if there is no ambiguity, an organisation will
neither be able to suppress error nor generate alternative routes of action. In
short, it will be most unreliable and least flexible, sluggish, as we now say.
Or, in other words, the complex new service delivery systems will create
duplication and overlap. Agencies from the public, private, and voluntary
sectors will compete for clients and thereby increase the take-up of services.
This outcome may be welcome but it is costly on two grounds: the ‘inefficien-
cies’ of duplication, and paying for more people to receive the service.
Accountability
The hollow state erodes accountability. First, some governmental institutions

have never been effective instruments of representative democracy. Whatever
merits the European Parliament may have, and there must be some, not even
its most committed proponent can claim it holds the Commission to account.
Second, sheer institutional complexity obscures who is accountable to
whom for what. Policy networks, or professional–bureaucratic functional
alliances, are a characteristic feature of policy-making in Britain. Such net-
works restrict who contributes to policy-making and policy implementation.
They routinize the policy process. They are also a form of private government;
much of their work is invisible to the parliamentary and public eye. With the
growth of transnational networks linking UK networks to the EU, the policy
process becomes more complex and the lines of accountability ever more
difficult to identify.
To talk of multiplying service delivery systems is just another way of
describing institutional complexity. As Elgin and Bushnell (1977) argue,
complex systems decrease the ability of individuals to understand the system,
reduce public participation and limit access to decision-makers. One simple
illustration will do. Thus, the all-purpose local authority has been bypassed for
non-departmental public bodies and other kinds of special purpose agencies.
Under the Greater London Council (GLC), ratepayers paid their rates to four
bodies. After abolition of the GLC, they pay taxes to 17 bodies, two-thirds of
which are unelected (Travers 1986: 428).
Third, several of the new institutions were introduced to limit the influence
of elected representatives, especially local councillors: for example, urban
development corporations. The new agencies are another case in point. Imp-
ishly, Gerald Kaufman complains:
Bichard [Chief Executive, Benefits Agency] keeps writing to me, and I want him
to stop. Whenever I have a constituency case involving a social security problem,
I write about that case to the government minister responsible: these days, Peter
Lilley, Secretary of State for Social Security.
Lilley passes my letter to Bichard. Bichard then writes to tell me he is looking
into the case (Kaufman 1992: 21).
Kaufman is scoring off a political opponent, but there is a serious point to his
complaint: one objective of the new arrangements was to distance operational
management from the incessant demands of parliamentary accountability.
Unfortunately, the distinction between operational and policy issues is a
blurred and shifting line, leading inexorably to the problem of ‘Who is
accountable to parliament for what?’ As Grant Jordan comments:
There is a deliberate or accidental ambiguity. We are told ministerial account-
ability remains. But in reality it is now accountability to the Minister by the
Chief Executive rather than accountability of the Minister to the House of
Commons that is now on offer; these are different (Jordan 1992: 13).
Catastrophe
According to Hood and Jackson (1991), ‘government’s capacity to manufac-

ture social disasters has greatly increased’ and they identify five features of
NPM that increase the potential for misinformation: ‘corporatisation and
privatisation; deregulation and the accommodation of business; cost-cutting
and simple goal management; bureaucratic staffing by political loyalists and
policy making by opinion polling’.
In summary form, they argue NPM breaks up government organizations
into separate units creating both barriers to communication between the units
and incentives to distort and conceal information. The hands-off approach to
business and the relaxation of regulations encourages lax enforcement and
capture of the regulators. Cost-cutting leads to: loss of front-line staff; cutbacks
in measures to mitigate disasters because in normal times such cuts will attract
little attention; reduced maintenance, especially staff training; solutions to the
wrong problems because of the focus on limited objectives; and reduced
redundancy, which removes the flexibility so essential to coping with disasters.
Finally, the employment of executives on contract leads to a loss of bureau-
cratic experience. In sum:
NPM does appear to contain several of the organisational ingredients which have
been associated with socially-created disasters. At the worst, NPM could be a
disaster waiting to happen.
Central Capability
This combination of fragmentation and accountability points to another

meaning for the phrase ‘hollowing out the state’; it can also refer to a decline
in central capability. In other words, fragmentation constrains the centre’s
administrative ability to coordinate and plan. Diminished accountability con-
strains the centre’s ability to exercise political control. So, current trends erode
the centre’s capacity to steer the system—its capacity for governance. Sir
Robin Butler’s concern ‘to maintain a degree of cohesion across the service
as a whole’ addresses the problem of fragmentation and he argues that:
it is essential that it does not reach the point where individual Departments and
their Agencies become simply different unconnected elements in the overall
public sector with little in the way of staff transferability, and no real working
mechanisms for policy coordination (Butler 1993: 404, emphasis added).
The warning is clear but unheeded.
THE CASE F OR BUREAUCRACY
There can be a heady delight in perversity and to argue for a return to

bureaucracy is, in today’s political climate, perverse. Fortunately, there are
also some good reasons supporting the case for bureaucracy.
First, there is the case based on the current orthodoxy. Bureaucracy is a tool
just like any other service delivery system. If the aim is to experiment with
service delivery systems, if there is no one right way to deliver services, it
follows there may be conditions under which bureaucracy will be an effective
tool. The key question becomes ‘What might those conditions be?’ The answer
is ‘Where fragmentation produces suboptimal outcomes, bureaucracy provides
a central strategic capability.’
Second, there is the case based on evidence. The critics of bureaucracy,
whether of Left or Right, are factually wrong in many of their assertions.
Goodsell (1994, 2015) produces a cogent survey of the case for bureaucracy
that marshals much evidence to show that: most people are satisfied with the
services they receive from bureaucracy; public sector productivity has risen
continuously over the past two decades; the public sector has a better record
than the private sector as an equal opportunity employer; bureaucrats are
motivated and exercise initiative and discretion; bureaucrats are receptive to
change; and bureaucracies can and do change. Although Goodsell’s evidence is
for American government only, the equivalent defence can be mounted for
British bureaucracy. The cavalier neglect of evidence of many critics is clearly
illustrated by Osborne and Gaebler. Their object is ‘to bash bureaucracies’.
They want to promote the entrepreneurial spirit, to which end they tell many
stories about the entrepreneurial spirit in government bureaucracies. And here
lies the central paradox. The very bureaucracies they seek to bash are also the
sources of innovation (Osborne and Gaebler 1992: xviii)!
Third, there is the case based in political theory. For example, Wamsley
et al. (1984) argue that the ‘Public Administrator’ has an important role to play
in identifying the ‘public interest’. They define the ‘public interest’ as an ideal
and a process, meaning it is the product of a continuing dialogue that
addresses long-range views, competing demands, and all affected individuals
and groups. The ‘Public Administrator’ is a trustee of the public good. It is
their responsibility to look beyond the short term, to stimulate reasoned
debate, to involve citizens in that debate, and to expand the opportunities
for that involvement. These Platonic guardians of the public interest may pose
acute problems of political accountability but the official has always existed as
a counterweight to the politician. They are the repository of ‘specialised
knowledge, historical experience, time-tested wisdom and . . . some degree
of consensus of the public interest’, which acts as counterweight to short-term
political expediency and opportunism (Goodsell 1995: 155). They stand for
integrity and probity against partisan interest and corruption.
Finally, there is the case rooted in political necessity. Bureaucracies have

demonstrable advantages, including reliability, predictability, probity, cohe-
sion, and continuity. Above all, they provide direct, hands-on control of
services through the hierarchical, rule-based, disciplinary structure. These
characteristics favour intervention. Should any future government rail against
the constraints of fragmented service delivery systems and seek to steer, the
tool it will turn to will be bureaucracy. A government with redistributive aims
will have obvious incentives to intervene but if there is the potential for
catastrophe discussed earlier, then the political complexion of the government
will be irrelevant. Needs must where the devil drives and foundering service
delivery systems carry a high electoral penalty.
CONCLUSIONS
I freely admitted that talking about the hollowing out of the state was
speculation but it was speculation with a clear purpose; it signalled that
potentially dramatic changes were under way in British government. Govern-
ment is smaller. Both central and local government are losing functions to
other agencies and to the EU. Service delivery systems proliferate. The role of
officials is increasingly constrained by new management systems and political
controls. The obvious outcomes of these changes are fragmentation and
diminished accountability. There is also a less visible but more important
erosion of central capability. This erosion, coupled with the arrival of the
information polity, enlarges the potential for catastrophe. When catastrophe is
coupled with limited central capability, there will be a strong imperative for a
return to bureaucracy because governments will wish to strengthen their
capacity to steer the system.
So, we confront the conundrums in the panoply of recent change. Govern-
ance is not a choice between centralization and decentralization. It is about
regulating relationships in complex systems. There is no simple ideological
choice between planning and markets. There are many forms of service
delivery and we need to identify the conditions under which they work.
Private sector management does not necessarily serve the purposes, or work
in the distinct conditions, of the public sector. There are no solutions to
problems, only a process of policy succession that requires the capacity to
learn from mistakes. The process of hollowing out in British government
is not another way of heralding the minimalist state of Thatcherite aims. It
is more important; it is about redesigning governments to cope with scarcity
and devising complex solutions to problems that defeat the simple-minded
nostrums of both free markets and national plans. The lessons are there to be
learnt. I will end with another of Wildavsky’s aphorisms, ‘Scepticism depends
on dogma’ (Wildavsky 1980: 206). We have the dogma, but where is the
organized scepticism to find the lessons of the bold new era of the hollow state?
AFTERWORD
This Afterword walks a difficult line between updating the analysis and
replying to my critics as the two topics blur into one another. I consider it
more helpful to readers if the reply to critics is in one place, so I will return to
the topic of the hollowing out in Chapter 12.4 Here, I concentrate on describ-
ing what has changed empirically and demonstrating the continuing relevance
of my speculations. So, briefly, I revisit the topics of privatization, alternative
service delivery systems, the consequences of EU membership, and NPM.
In 1994, the UK economy was recovering from ‘Black Wednesday’
or Britain’s withdrawal from the European Exchange Rate Mechanism on
16 September 1992. This decision cost the UK economy some £3.3 billion and
caused an economic recession. However, by 1994, the economy was growing,
unemployment was falling, and inflation was low. Unfortunately for the
Conservatives, they gained little credit for the recovery. They polled less
than 30 per cent of the electorate and were about to lose power. So, my
original remarks were shaped by 18 years of Conservative rule and their
avowed intent of redrawing the boundaries of the state.
After New Labour’s electoral victory in 1997, Tony Blair kept much of the
neo-liberal agenda for reforming the public sector but his Chancellor of
the Exchequer, Gordon Brown, did not believe in the minimalist state.
Under his tutelage, public expenditure rose steadily in the 2000s to a peak of
47.7 per cent of GDP (Keynes and Tetlow 2014). Similarly, public employment
rose again in the late 1990s throughout the 2000s. To a significant degree,
there was a return to bureaucracy. However, with the advent of the Coalition
in 2010, both these trends were reversed. Public spending was cut and fell to
44.4 per cent by 2014. Public employment was cut from an average of some 20
per cent in the 2000s to 14.8 per cent in 2013 (Cribb et al. 2014: 36), although
health service and education continue to be protected.5
Although privatization continued at a lesser rate under both the Blair and
Cameron governments, and industries were sold to the private sector (for
example, the Royal Mail, the Tote), the most significant trend since 2010 was
4
Of course, I was not alone is diagnosing the hollowing out of the state. See, for example,
Frederickson 1996; Jessop 2004; Klijn 2002; Milward and Provan 2000; and, surprisingly in view
of his later strictures on this subject, Peters 1994.
5
The Institute of Fiscal Studies provides authoritative commentaries on the state of the
economy (see http://www.ifs.org.uk/). The Office of Budget Responsibility is another authorita-
tive source of data—see http://budgetresponsibility.org.uk/publications/. See also www.hm-
treasury.gov.uk/psf_statistics.htm.
the growth of contracting out. Bowman et al. (2015: 2–3) describe this growth
as the emerging ‘franchise state’. Citing Gash et al. (2013: 4), they estimate this
public service industry had a turnover of some £100 billion a year with some
£1 in every £3 going to independent providers (see also Raco 2016). A few
global firms have emerged that specialize in contract delivery and regulation
on a mass scale (for example, G4S, Atos, Capita, and Serco). Indeed, the
annual update on Open Public Services (Cabinet Office 2012: 13) was explicit:
In the world we are now entering, all those who serve the public will have a right
to be recognised as public servants—regardless of whether the organisations for
which they work are traditional public sector agencies, independent trusts,
employee mutuals, private enterprises, social enterprises or community groups.
Dedication to the provision of high-quality public services should be recognised
as the hallmark of the public servant, regardless of which particular type of
employer he or she happens to work for.
The state and these giant corporations are co-dependent. The corporations
rely on the taxpayers’ money, and the state that awards and monitors the
contracts has ‘stripped itself of institutional resources and intelligence previ-
ously used to deliver goods and services’ (Bowman et al. 2015: 5).
The cuts in public expenditure and public employment, and the growth of
the franchise state were underpinned by the so-called austerity narrative
(Blyth 2013; Johnson and Chandler 2015). The ‘structural current budget
deficit’ is the perceived problem. It was caused by the spending of the previous
Labour government, the secondary banking crisis in America, and world
recession. The governing elite agreed the most pressing problem facing British
government was the size of the public sector debt. So, the major parties agreed
we must have spending cuts to bring the deficit down; fiscal consolidation. As
with Margaret Thatcher’s Conservative government, this economic reality was
a brute fact; government must do less and public spending must be cut. An old
acronym returned: TINA—there is no alternative. The budgets for health and
education were ring fenced, so most of the cuts fell on welfare payments and
local government. However, the combined impact of the continuing Eurozone
crisis, slow growth, high unemployment, and low productivity meant that
deficit targets were not met. Further cuts in public spending were announced
over the life of the 2015 parliament. The deserving poor were protected (the
elderly). The undeserving poor (everyone else) paid for the polite euphemism
of ‘fiscal consolidation’. Cameron has presided over a ‘concerted assault on the
bottom third of society’ (Toynbee and Walker 2015: 3). Or, in the language of
broadsheet headlines, the recipe is ‘more poverty and worse public services’
(Guardian, 8 December 2013).
This austerity narrative is about not only fiscal consolidation but also that
age-old neo-liberal ambition for the minimal state. As Bale (2014) argues
‘the right—free-market, small-state, low-tax, tight-borders, tougher sentences,
eco- and Euro-sceptical—is where the solid centre of the [Conservative] party
now comfortably resides’. So, the franchise state is part of the blueprint for
creating and managing the minimal state. It involves a concentration of power
in the hands of the Minister and the central department. Intermediate insti-
tutions are abolished or bypassed for various forms of markets. Services are
outsourced or privatized. Managers are becoming empowered. The new quasi-
markets are subject to regulation by performance measurement regimes that
are also intended to foster choice by giving citizens more information. There
would appear to be a coherent ‘new governance’ narrative stemming from the
impact of neo-liberal, managerial, and neo-conservative ideas (Bevir and
Rhodes 2016). This blueprint has been the solid centre of government policy
since 2010. It also brings to the fore once again questions about the capacity
of the central state. As Bowman et al. (2015: 3) argue, the ‘franchise state is
socially wasteful and administratively inefficient’. The companies ‘game the
contractual system and taxation regimes’ and the state has ‘limited organisa-
tional capabilities’ to regulate such gaming (Bowman et al. 2015: 6).
In other words, the hollowing out thesis continues to pose questions about
the roles and boundaries of the state. The trends in public spending, public
employment, and privatization in the 2010s show the continuing relevance of
my argument, with an important qualification. Because I was writing about the
heyday of Conservative rule in the 1980s and 1990s, I did not allow for the
influence of the differing beliefs of the political parties. It is not a black and
white argument. The neo-liberal agenda is shared by New Labour, and the
Labour opposition under Ed Miliband adhered to the austerity narrative.
Nonetheless, neo-liberal beliefs in the minimal state are held mainly by the
Conservative Party whereas the Labour Party envisages a continuing, even
decisive role for the central state. The franchise state is hollowing out in new
clothes. As in the Thatcher and Major years, Conservative rule continues to
hollow out the central state.
The immediate retort to this argument is that the state has greater control
over less. But all these changes disaggregate public bureaucracies—central and
local—while providing limited capacity for regulation. It does not control the
franchise state. It does not even monitor the outcomes.
Nor is it obvious that British government had much influence over the issue
of British sovereignty in the EU. Indeed, it was the lack of such influence that
helped to fuel demands for the 2016 referendum and subsequent British exit
from the EU. Indeed, there was a case to be made that the influence of the EU
had grown. Clifton (2014) suggests that public service delivery is being
Europeanized because many services are increasingly treated as economic
and, therefore, within the purview of the Commission. For example, are the
subsidized lifeline ferry services to the Scottish islands an economic service
subject to competition rules? The Commission said they were and the Scottish
government had to introduce competitive tendering. To everyone else, the
ferries were an unprofitable social service for which there was little scope for
competition. No matter. The Commission has become a policy entrepreneur.
We enter the era of the ‘strait-jacketed state’. Similarly, Richardson (2012: 12)
argues that the EU has ‘acquired quite a high degree of sovereignty and by
so doing has begun to look very state like’. Morphet (2013: 201) concludes
that the EU has ‘shaped key areas of British public policy . . . and as the extent
of pooled powers has increased, then so has the level of influence’. Britain
remained a defensive, semi-detached member of the EU until Brexit. In
Morphet’s (2013: 209) characterization, Britain received policies made in the
EU rather than engaging with the process. That was not the statecraft of a
strong state.
The New Public Management became so all-embracing it lost any distinct-
ive meaning (Pollitt and Bouckaert 2011). It became a synonym for public
sector reform. Such reform was a constant as successive governments sought
for the ever elusive solution they could not articulate to problems they could
not define with precision and accuracy. A former senior civil servant opined:
Blair confuses the civil servants around him: On the civil service, he doesn’t know
what he wants. They say, in effect, ‘Tell me what you want and we’ll do it.’ But he
keeps saying different things. Richard Wilson finds it very difficult the way the
Prime Minister jumps around (Hennessy 2000b: 9).
So, initiatives come and go. Hood and Lodge (2007: 59) suggest we have
created the ‘civil service reform syndrome’ in which ‘initiatives come and go,
overlap and ignore each other, leaving behind residues of varying size and
style’. Tony Blair famously remarked on public sector reform:
You try getting change, you know, in the public sector and public services and,
you know, I bear the scars on my back after two years in government and heaven
knows what it’ll be like if it was a bit longer (Blair 1999).
His question was probably rhetorical because over the ensuing years there was
yet more frustration over the pace of change. If the aim of managerial reforms
was to reduce civil service discretion and increase their responsive to their
political masters, then the conclusion is probably ‘job done’. If the aim was the
‘3Es’ of economy, efficiency, and effectiveness, then, at best, the case is non-
proven. Pollitt and Bouckaert (2011: 155) describe the results of reform as a
‘half empty wineglass’ because we don’t have the data about efficiency or
outcomes.
So, the dilemmas persist. The Blair government experimented with joined-
up government, seeking to improve horizontal and vertical coordination to
counter fragmentation. King and Crewe (2013) catalogue the policy blunders
or catastrophes of government, numbering the lack of accountability as one of
the perennial issues. They were not alone (see, for example, Butler et al. 1994;
Timmins 2012). The debate on central capability raged around such topics as
the presidentialization of the prime minister, the growth of special advisers,

and the expansion of support for the prime minister, creating a department
that dare not speak its name. I discuss many of these debates in Chapter 9 (this
volume) and in Rhodes 2017, Volume II, Chapter 7. For now, I accept that
there were many attempts to centralize policy-making on the prime minister
and his court but I observe that prime ministers rarely found arrangements
that suited them or lasted. If the aim was enhanced central capability, then the
endless tinkering attests to the ineffectiveness of many arrangements.
Whether the preferred term is the franchise state, the straightjacketed state,
or the hollow state, the common thread is the argument about the conse-
quences of neo-liberal economic ideas, privatization in all guises, membership
of the EU, and the reform of public service delivery for our understanding of
the state, and of the capacity of the central state to govern. The reform
blueprint for both the Coalition and Cameron’s Conservative governments
once again emphasizes the salience of these trends for our understanding of
the roles and boundaries of the state (and see Chapter 12, this volume).
9
From Prime Ministerial Power

to Core Executive
Much work on the UK executive focused on long-running ‘chestnuts of the

constitution’, especially the controversy about the relative power of the prime
minister and the Cabinet (Heclo and Wildavsky 1977: 341–3).1 Mackintosh’s
(1968 [1962]) study crowned the debate with an impressive summary of the
historical evolution of Cabinet government:
It is impossible to establish conclusively the precise weight or influence of one
office or body out of an elaborate series all of which have some share in the
decision-making process. . . . But so far, the politics of the 1960s have strength-
ened rather than weakened or altered the lines of development which have led
contemporary British Government to be described as Prime Ministerial rather
than Cabinet Government (Mackintosh 1968: 627).
Over the last 40 years little systematic fieldwork-based research into the PM or
Cabinet government has been published (the major exceptions are Heady
1974 and Hennessy 1986 and 1995).2 Although this topic lives on as a
standard controversy much raked over by students and newspaper columnists,
in political science it has been an inactive field.
This chapter focuses on the British ‘executive’, especially the ‘core execu-
tive’. This term is used in a broader sense than usual. Textbooks on British
government typically refer to ‘the executive’ (for example, Norton 1984;
Kingdom 1991) and limit discussion to the power of prime minister and
Cabinet. In fact, what constitutes the executive varies from policy area to
1
The ideas in this chapter were first aired in Dunleavy and Rhodes (1990) and Rhodes (1993)
and brought together in Rhodes and Dunleavy (1995). This chapter is an edited version of
R. A. W. Rhodes (1995) ‘From Prime Ministerial Power to Core Executive’, in R. A. W. Rhodes
and P. Dunleavy (eds), Prime Minister, Cabinet and Core Executive. London: Macmillan,
pp. 11–37. Reproduced with permission of Palgrave Macmillan.
2
In the 2000s, see Blick and Jones 2010; Foley 2000; Hennessy 2000a; and Rose 2001.
policy area. Departments take important policy decisions with little or no

reference to the Cabinet and prime minister. Equally, central coordination is
not a function solely of the prime minister and Cabinet: for example, the
Treasury’s role in economic policy decisions.
The term ‘executive’ is used here to refer to the centres of political authority
that take policy decisions. In other words, the executive institutions are not
limited to prime minister and Cabinet but also include ministers in their
departments. The term ‘core executive’ refers to: all those organizations and
procedures that coordinate central government policies, and act as final arbiters
of conflict between different parts of the government machine. In brief, the ‘core
executive’ is the heart of the machine, covering the complex web of institu-
tions, networks, and practices surrounding the PM, Cabinet, Cabinet com-
mittees and their official counterparts, less formalized ministerial ‘clubs’ or
meetings, bilateral negotiations, and interdepartmental committees. It also
includes coordinating departments—chiefly, the Cabinet Office, the Treasury,
the Foreign Office, the law officers, and the security and intelligence services.
The label ‘Cabinet government’ was the overarching term for (some of) these
institutions and practices but it is inadequate and confusing because it does
not describe accurately the effective mechanisms for achieving coordination.
At best it is contentious, and at worst seriously misleading, to assert the
primacy of the Cabinet among all organizations and mechanisms at the
heart of the machine.
The term ‘core executive’ and the broad definition of the ‘executive’ are
working hypotheses. I use them to raise the issues of coordination and frag-
mentation in central government. The term ‘core executive’ directs attention to
the extent and efficacy of, and the various mechanisms for, coordination. The
‘executive’ focuses attention on the policy-making role of departments and
their relationship to the core executive. Above all, my terminology provides a
neutral description of the subject. It does not anticipate or prejudge the results
of empirical research.
This chapter answers two questions: ‘where are we now?’ and ‘where are we
going?’ Thus, the next section reviews the academic literature on the prime
minister and the Cabinet, not ministerial diaries and memoirs, arguing that the
debate about the relative power of these two institutions is stultifying. I set the
scene by briefly sketching the conventional wisdom on prime minister and
Cabinet, and identifying the defects of such a focus. This conventional wisdom
both restricts the questions posed about the core executive and ignores the
diversity of approaches. I identify and describe six variants of the prime minis-
terial government–Cabinet debate, concluding that the literature is theoretically
weak, conservative in its methods, and leaves many questions unanswered. The
third section asks, ‘Where are we going?’ It answers the question by arguing for a
focus on the ‘core executive’. I identify some promising theories, innovative
From Prime Ministerial Power to Core Executive 139
methods, and key research questions that need to be explored if the analysis of
the core executive is to blow fresh air on a musty topic.
WHERE ARE WE NOW? THE CONVENTIONAL

WISDOM DEFINED
What is the standard controversy surrounding the study of the British execu-
tive? Advocates of the prime ministerial power thesis argue that he or she is
more powerful than the Cabinet because the prime minister is leader of the
party; has the power to appoint and dismiss ministers; chairs the Cabinet and
controls its agenda; has more opportunity to amass considerable personal
popularity with the electorate through skilled use of the media; appears on an
international stage as a world leader; and, because of freedom from depart-
mental responsibilities, enables him or her to intervene over the full range of
government policy (see, for example, Benn 1980; Crossman 1963; Mackintosh
1968; Madgwick 1986).
Advocates of the Cabinet government thesis counter these claims by
pointing to the constraints on the prime minister. Thus, the party cannot be
ignored, it has to be listened to; ministers have their own bases of support in
the party and even the country at large; constitutional conventions require the
government to act collectively; public visibility can be two-edged with the
prime minister blamed when things go wrong; appearances on an international
stage serve merely to highlight how little such jamborees achieve; and the prime
minister lacks the expertise and advice necessary to intervene effectively in the
complex world of departmental policy-making (see, for example, Jones 1985;
Madgwick 1986, 1991; Norton 1988).
The arrival of Mrs Thatcher gave an additional twist to these arguments. It
is argued that she was a particularly dominant leader, providing an important
precedent for her successors. This increase in prime ministerial power was
supported by an enhanced role for, and increased numbers of, advisers at No.
10. Evidence of her dominance can be seen in her interventions in depart-
mental policy (for example, local government finance, football hooliganism)
and by, for example, the extensive use of prime ministerial powers of appoint-
ment of top civil servants. But, in this seemingly endless round of assertion
and counter assertion, it is argued that Mrs Thatcher’s domineering leadership
style isolated her from both party and Cabinet and the latter in particular was
instrumental in bringing her down (Alderman and Carter 1991; Jones 1995;
and Smith 1995). The size of the No. 10 unit cannot be compared, even
remotely, with that of a ministerial department. The evidence for the political
appointment of civil servants is scanty at best. Intervention brought its own
problems. Her initiative on football hooliganism embarrassed Mrs Thatcher

when her pet scheme had to be abandoned in the wake of the Hillsborough
disaster. Local government finance in general, and the community charge in
particular, contributed to her downfall, constituting perhaps, to continue with
a footballing rather than a fashionable cricketing metaphor, the most spec-
tacular own goal of the post-war period.3
There are several problems with this textbook approach to the study of
the British executive. King (1985a: 3–7) has cogently identified some of the
defects, arguing that the literature is ‘thin’ and ‘uniform’; ‘old arguments are
rehashed’; propositions are advanced ‘without testing them against reality’;
and ‘the same materials are endlessly recycled’. In other words, neither side in
the debate can marshal much by way of evidence to support their conflicting
claims. Original fieldwork is at a premium. To make matters even worse,
important questions are almost totally neglected. Again King (1985a: 3–7) has
pointed out that the ‘variety and fluctuation’ in prime ministerial involvement
and the effects of a changing context are ignored. Indeed, ‘important aspects of
the prime ministership are dealt with not at all or only in passing’.
Third, the debate adopts an institutional perspective betraying little or
no interest in other theoretical perspectives. For example, Mrs Thatcher’s
leadership style has greatly preoccupied commentators. However, this subset
of the literature does not draw upon existing theoretical perspectives to
interpret the phenomenon. There is nothing equivalent to Barber’s (1972)
or Greenstein’s (1975) systematic analyses of the personality of American
presidents (although see Berrington 1974; Foley 1993; and Iremonger 1970).
Nor has the literature on business leadership been explored even though it
provides a ‘contingency theory of leadership’ (Lawton and Rose 1991: ch. 7;
McCall 1977), which explores the ‘fit’ between leadership style and the context
within which a leader operates. The usefulness of comparing several different
interpretations of the same event was demonstrated as long ago as 1971 by
Allison in his analysis of the Cuban Missile crisis from three different view-
points. Students of the British executive have yet to follow his example. Finally
and paradoxically, by focusing attention on the prime minister-versus-Cabinet
debate, the conventional wisdom misrepresents the existing literature. There
have been several attempts to move beyond its confines. A more sympathetic
reading of the literature identifies several variants that suggest future avenues
of exploration and support the argument that attention should be switched to
the core executive.
3
In his resignation speech as Foreign Secretary, Sir Geoffrey Howe commented that his role
in EU negotiations was ‘rather like sending your opening batsmen to the crease, only for them to
find, as the first balls are being bowled, that their bats have been broken before the game by the
team captain’. House of Commons, 13 November 1990.
VARI E T I E S OF IN S T I T UTI O N A L I S M
There are six models that can be distinguished in the debate about prime
ministerial power, many of which advance an explanation (of how things work)
and a prescription (about how things ought to work). The models are prime
ministerial government; prime ministerial cliques; Cabinet government; minis-
terial government; segmented decision-making; and bureaucratic coordination.
Prime Ministerial Government
Prime ministerial government is conventionally seen as the exertion of

authority solely by the premier, or monocratic government. The prime min-
ister’s personal predominance in decision-making can be demonstrated in
three possible ways: by a general ability to decide policy across all issue areas in
which he or she takes an interest; by deciding key issues that subsequently
determine most remaining areas of government policy; or by defining a
governing ethos or ‘atmosphere’ that generates predictable and hard solutions
to most policy problems. As a result, other ministers’ freedom of manoeuvre is
constrained, making them simple agents of the premier’s will.
The first version of the argument has conventionally been criticized because
of the mismatch between the time and workload pressures inherent in the
PM’s office and the complexity of policy-making and administration in the
modern extended state. The job of any Cabinet minister is ‘a conveyor belt to
exhaustion and under-achievement all round, a predicament reflected in . . .
the finished policy, which is all too often defective and immensely difficult
to implement’ (Hennessy 1986: 184). These limitations acquire added force
given the demands on Downing Street. A hyperactive premier can consider-
ably enlarge the scope of issues he or she addresses compared with a more
relaxed or less hard-working incumbent. Yet international summitry, overseas
visits and visitors, and similar events consume much of the PM’s time, greatly
restricting her or his scope of attention and capacity to follow through
on issues.
The other two versions of the case for monocratic authority are not open to
such obvious objections. Prime ministerial control of key issues is plausible,
and visible, during economic or international crises, such as the pre-
devaluation period 1964–67 or limited wars (see Seymour-Ure 1984). How-
ever, such events are relatively rare and prime ministerial control over these
kinds of ‘key issues’ is less certain. The ideological authority version is more
limited in its application, because some recent PMs (such as Harold Wilson)
apparently had no such guiding principles. It fits Mrs Thatcher’s premiership
well enough (Young 1989).
Mrs Thatcher’s period of office converted the monocratic version of prime

ministerial power into an orthodoxy (Young and Sloman 1986; Wapshott and
Brock 1983), which many academic authors have also supported (Burch 1983;
King 1985b; and Minogue and Biddiss 1988):
What we have established through the testimony of many intimate witnesses is
the crucial importance of her personality to what her government does. Her
politics proceeds from her character. Her style of leadership turns heavily on her
being a woman . . . This is her time dominated by her character
(Young and Sloman 1986: 142).
[Thatcher] had no choice, given her aims and determination, but to lead in an
unusually forthright, assertive manner. Partly this was a matter of her personality:
she is a forthright and assertive person. But it was at least as much a matter of the
objective situation in which she found herself. She was forced to behave like an
outsider for the simple reason that she was one (King 1985b: 116).
[Thatcher] reaches out for decisions; she reaches out for people. She also reaches
out for ideas (King 1985b: 126).
She pushed out the frontiers of her authority ever since she took office in 1979
(King 1985b: 137; see also Foley 1993).
Yet if the Thatcher era apparently highlighted the strengths of the mono-
cratic approach to prime ministerial power, it also rather graphically demon-
strated its dangers and limitations. The resignations of Michael Heseltine over
the Westland affair (Linklater and Leigh 1986); Nigel Lawson over the role of
Sir Alan Walters as the PM’s economic adviser; and Sir Geoffrey Howe over
EU policy all illustrate that a domineering style can impose heavy costs on a
PM. It also proved too easy for observers adopting the monocratic line of
argument to collapse into hagiography, describing Mrs Thatcher as a leader
who ‘towered over all her contemporaries’; had an ‘unwavering purpose’;
evoked ‘admiration and detestation for one identical reason: she is “big”’;
fell ‘short of greatness, but radiates dominance . . . I do not believe that in our
lifetime we shall ever look upon her like again’ (Finer 1987: 140).
Prime Ministerial Cliques
An alternative interpretation (popular in other countries as well as the UK)

argues that the premier’s authority and influence are collective attributes of
her or his inner group of advisers. No single individual can hope to impose
leadership on the complex core executive of a large, modern nation state, let
alone upon the wider executive. Political leadership cannot be narrowly
conceived as personal initiatives or interventions in decision-making. It is
rather the product of a process of constructing, maintaining, and providing
political clout to a set of influential people in political and administrative

positions. An extended team or coalition can: generate ideas and applications
of the premier’s basic values; monitor a broad scope of government policies;
tell a broad range of actors and institutions about the premier’s intentions; and
regularly and reliably follow up on policy implementation to prevent the
premier’s input to decisions being ignored or forgotten.
Historically in the UK, this approach has been associated with arguments
about whether there is an ‘inner Cabinet’ of influential ministers and Cabinet
committee chairs. The general academic consensus is that there is no coherent
pattern, it is haphazard, and the debate has become tepid (James 1992: 194;
Walker 1970: 39–40; but see Rhodes 2017, Volume II, Chapter 7, for a
discussion of ‘court politics’).
A second version of the clique model posits an éminence grise or consiglieri
exerting a disproportionate influence upon the premier’s choice of policies and
individuals. In earlier periods, people nominated for this role included Horace
Wilson for Chamberlain or Lord Cherwell for Churchill. More recently media
and academic observers have focused on close and trusted advisers such as
William Armstrong, the Head of the Civil Service at the time of the Heath
government (Hennessy 1986: 80); members of the PM’s personal staffs, such
as Marcia Williams, Wilson’s Political Secretary (Haines 1977); and Thatcher’s
powerful Press Secretary, Bernard Ingham (Cockerell et al. 1985). Since the
mid-1970s, the premier’s Downing Street advisers emerged as distinct ‘players’
in some policy decisions: for example, Sir Alan Walter’s role in economic
policy-making. In addition, the PM has been closely linked to specific civil
service units in: the Cabinet Office, such as CPRS in the late 1970s and the
Efficiency Unit in the 1980s (see Lee 1974; Blackstone and Plowden 1988); and
the Treasury, such as the FMI Unit and later the Next Steps Unit; and
elsewhere, such as the Intelligence Services.
The episodic extension of Thatcher’s advisers to cover most key policy
areas, and the substantial upgrading of the personnel who serve as advisers
on economic affairs and foreign policy, and in the Policy Unit, has sparked a
controversy about the creation of a fully fledged Prime Minister’s Department
(Berrill 1985; Weller 1983; Jones 1980, 1983). The increasing salience of news
and media management in core executive operation in the 1970s and 1980s,
and especially the centralization of the government information services under
Bernard Ingham, extended the scope of the prime ministerial clique to include
a variety of presentational experts, news manipulators, and ‘spin doctors’
(Cockerell et al. 1985; Cockerell 1988; Margach 1978; May and Rowan 1982:
101–57). Paradoxically for a premier credited with much personal influence by
critics and admirers alike, Thatcher had an extended network of advisers,
image consultants, speech writers, and intellectuals, although their influence
tended to be episodic rather than continuous.
Critics of this trend argue that prime ministerial cliques undermine the
official allocation of ministerial briefs and departmental advice-giving, creat-
ing a parallel power network inside the executive. In effect, there is a counter
bureaucracy duplicating formal governmental structures. Other critics see the
premier’s clique as a way for a highly biased selection of external interests to
gain privileged access to the centre of decisions. In the late 1970s under
Callaghan, these external inputs mixed trade union leaders with corporate
business elites. In the 1980s, the networks plugged into Mrs Thatcher’s gov-
ernment were confined to major finance and industrial capitalists, together
with a few less conventional business entrepreneurs and assorted right-wing
think tanks or intellectuals.
The clique view expresses a long-running liberal fear that the top political
executive is not accountable for its policy-making. A related worry draws
attention to the dangers of a premier constructing a tightly knit set of advisers
insulated from outside networks or experiences. They can develop a strong
group consciousness and awareness of their elite influence. Janis (1972) argues
that groups can become divorced from outside networks and experiences and
develop a ‘group think’ syndrome in which policies are developed and pursued
for lengthy periods in the face of mounting external evidence of policy failures
or fiascos, which the elite group simply ignores or discounts. The community
charge, or poll tax, may perhaps be an example of such cut-off ‘groupthink’.
Cabinet Government
Proponents of Cabinet government were scarce in the 1980s. The justification

for collegial decision-making that it takes account of a diversity of depart-
mental interests remains unchanged. Until 1990, the continuing importance of
the Cabinet was asserted in a defensive fashion. Because of governmental
growth and the complexity of decisions, the executive became a ‘fragmented
set of overlapping decision arenas’ in which the Cabinet gives ‘the system a
focus but which itself takes only a small proportion of decisions in full session’
(Mackie and Hogwood 1985: 31–5). Similarly Barnes (1989) acknowledges
that in the 1980s the Cabinet’s actual discussions focused on legislative
timetabling, foreign affairs, public expenditure, and pre-budget discussions,
and the occasional ‘fire-fighting’ issues. Nonetheless, he argues that Cabinet
has a key residual role as court of appeal both for ministers radically out of
sympathy with a general line, and for a premier confronted by a ministerial
colleague who insists on ploughing her or his own furrow. Because of the
legal and constitutional pre-eminence of ministers in policy-making (see
pp. 145–7), a PM faced with a minister who refuses to toe the line agreed by
a majority of colleagues may be forced to take the issue to Cabinet to be
authoritatively resolved, as Thatcher did during the Westland affair (Dunleavy

1995a). Proponents of the Cabinet government thesis were more assertive
after Mrs Thatcher’s fall and many saw the lack of Cabinet support as a
decisive factor (Alderman and Carter 1991; and Jones 1995).
Ministerial Government
Ministerial government is the term coined by Jones (1975) to describe the

British executive. He argues against prime ministerial dominance, less in terms
of the vitality of Cabinet as a collective decision-making organ, and more in
terms of the capacity of political and administrative departmentalism to limit
the premier’s influence. The key problem of Cabinet:
has been how to organise itself to cope with the tremendous increase in the
amount, complexity and inter-relatedness of its business. . . . The over-riding
objective of the Prime Minister is . . . to hold together the Cabinet, which is
potentially fragmented (Jones 1975: 41, 57).
Similarly, Heclo and Wildavsky detect a key ‘debility’ where Cabinet members
are also ‘chief executives of their own departmental empires, empires where
their individual reputations are made and/or unmade . . . Everyone knows they
serve themselves by serving their departments’ (Heclo and Wildavsky 1974:
371, 369). A former Head of the Home Civil Service agrees:
The form and structure of a modern Cabinet and the diet it consumes almost
oblige it to function like a group of individuals, and not as a unity. Indeed for each
minister the test of his [sic] success in office lies in his ability to deliver his
departmental goals . . . No minister I know of has won political distinction by his
performance in Cabinet or by his contribution to collective decision-making. To
the country and the House of Commons he is simply the minister for such-and-
such a department and the only member of the Cabinet who is not seen in this
way is the Prime Minister (Wass 1984: 25).
There are both legal-constitutional and political reasons for the continuing
pre-eminence of individual ministers in making decisions within their
departments.
In constitutional theory, the minister continues to take precedence before the
ministry. In law it is the minister who is usually responsible for the actions of
the ministry . . . A minister has unlimited liability for political mistakes made by
his ministry. . . .
A minister’s position at the top of a hierarchy is an ambiguous eminence. The
minister is answerable for everything that happens within the ministry (and often
outside it), yet is remote from what is done by officials at the base (Rose 1987: 232).
This general position is reinforced in many areas of government decision-
making by legal constraints on the ways in which ministers exercise parts of
their responsibilities. In general, when ministers are exercising quasi-judicial
functions, and in some cases where they are appointing people to head quasi-
governmental agencies, overt interference by their colleagues (even by the
premier) is illegitimate, and could be embarrassing if publicized. Usually,
neither Cabinet nor the PM can seek to decide such issues, irrespective of
the weight of a premier’s known attitudes with the minister responsible. This
reliance on ‘Chinese walls’, on government in compartments, works best in the
case of such ‘enclave’ areas as the ‘independent’ Law Officers (Marshall 1986:
111–17; Marshall and Moodie 1967: 144–50). It is apparently ineffective in
other areas, such as the supposedly separate ministerial controls over the five
main security services—MI5, GCHQ, SIS, Defence Intelligence Staffs, and
Special Branch (Richelson and Ball 1985).
Politically, the foundations for ministerial roles are also strong. Most
ministers spare little thought for policy issues controlled by their colleagues,
unless their own department has a stake in them. Attempts by one minister or
department to acquire responsibilities or functions from another, or occasion-
ally to assimilate a department whose rationale has dwindled, are the chief
reason for outside interventions—and these efforts are rarely successful. The
PM’s role inside the government is unusual because he or she alone can
regularly or legitimately take an interest in ‘specific areas of micro-policy’ on
a wider front. But Callaghan’s experience of trying to influence domestic
policy-making is typical:
In each case it involved intervention in the normal Whitehall processes and often
upset the respective departmental Ministers and officials who believe that Prime
Ministers should not trespass on their policy cabbage patches
(Donoughue 1987: 7).
Ministerial responsibility plays an important role in defending policy turfs.
Collective decision-making has dwindled. The demands for governmental
solidarity policed from Downing Street expanded. Ministers had good reasons
to insulate themselves from the centre to safeguard their positions, reinforcing
departmentalism. Some observers suggest that Mrs Thatcher’s ministers
sought to counteract her efforts to intervene in policy-making across many
issues by keeping business away from formal Cabinet machinery or inter-
departmental committees (Burch 1989). Conflicts formerly pushed up to
Cabinet committees are now settled bilaterally between the ministers con-
cerned, often via correspondence alone. The strengthened role of junior
ministers in the 1980s may reflect these efforts to internalize more policy-
making in the department, or at least resolve issues with other departments
without prime ministerial scrutiny (Theakston 1987). The large majority in
parliament strengthened the interest of Conservative backbench MPs in
particular subject areas, usually amongst those MPs sitting on a select com-
mittee, and created parliamentary ‘clienteles’ to whom ministers needed to pay
attention.
There were other countervailing tendencies to Thatcher’s apparent mono-
lithic control during the 1980s. In the early years, strong Treasury control of
public expenditures reinforced monocratic authority (Dunsire and Hood
1989). But the economic boom of the mid-1980s led to a more relaxed attitude
towards public expenditure and encouraged a departmental fight-back. There
was a shift away from Treasury control of minutiae to global target setting and
non-interference, as long as targets were met. The Financial Management
Initiative’s stress on decentralized cost control reflected a similar trend
(Gray, Jenkins et al. 1991). So too did the developing pattern of cash limits
administration, and the replacement of rigid manpower controls by running
cost controls and manpower targets (Thain and Wright 1990). Finally, the
Next Steps proposals for hiving off 75–90 per cent of civil service manpower
into executive agencies could strike a further blow at Treasury controls (hence
the opposition of its public expenditure control divisions to the proposals). In
the 1990s, one scenario sees stripped-down policy-making departments taking
on much of the current role of the Treasury’s public spending divisions. The
departments would be the sponsors controlling the expenditures and targets of
the numerous agencies and quasi-government organizations. These reforms
imply the continuing diffusion of effective policy control to the sectoral
departments. Only the wholesale privatization of departmental activities is
likely to reduce radically the scope of ministerial responsibility and control
(Jones 1989: 254–8).
The Segmented Decision Model
The segmented decision model suggests that some of the conflicting claims
of the previous models can be simply resolved by agreeing that the premier
and the Cabinet operate in different policy areas, with ministers operating
below the interdepartmental level at which Cabinet machinery becomes
involved. Prime ministerial control is strong in strategic defence, foreign
affairs, and major economic decisions, but genuine Cabinet or ministerial
decision-making predominates over almost all other aspects of domestic policy.
In public policy, a Prime Minister is doubly constrained. Positive requirements to
emphasise party management and the presentation of self limit the time that can
be devoted to policy. The primary responsibilities of departmental ministers also
constrain the involvement of Downing Street in policy making. Where the Prime
Minister is most involved, British government is now inevitably weak: this is true of
the management of the economy as well as foreign affairs
(Rose 1980a: 49 emphasis added).
This segmented pattern helps to explain why most commentators have detected
increased prime ministerial influence in the modern period, while evidence
of the PM’s weak involvement in large areas of domestic policy continues
to accumulate. For example, Donoughue describes Callaghan’s decision to
target some key domestic decisions in which he wanted to inject a new policy
direction—including the sale of council housing, improving educational
standards, the Finniston inquiry into the British engineering profession, and
the Annan Commission on the BBC. Yet by the end of his term it was ‘difficult
to claim much evidence of success’ (Donoughue 1987: 124). Thatcher’s longer
tenure of office provides some similar instances of long-running but equally
ineffective prime ministerial involvement. To the earlier examples of football
hooliganism and local government finance can be added decision-making on IT
(Keliher 1995), the campaign against litter, and the common agricultural policy
of the EU, among many others. The premier has great influence over strategic
decisions, which, in turn, influence many specific issues. But the resulting system
cannot usefully be described as ‘prime ministerial government’ for three reasons.
First, although the premier may play a key role in ‘objectively important’
issue areas, some of the most critical aspects of domestic policy-making always
remain open to Cabinet or ministerial decision.
Second, the power-dependency relations (Rhodes 1981: ch. 5) between the
premier and key ministers in foreign affairs, defence, and economic policy-
making, do not fit neatly into the mould suggested by enthusiasts for prime
ministerial government. Premiers cannot directly or single-handedly deter-
mine basic policy directions in any of these areas, nor even easily influence the
range of decision options considered. Prime ministers can select and reselect
the personnel involved, and may be able to arbitrate particularly uncertain or
difficult decisions. But appointing key figures to major positions creates
power-dependency effects. A premier may not be able to impose her or his
line on an appointee short of dismissal. The succession of public disagree-
ments between Thatcher and Lawson over the direction of economic policy
during 1987–89 dramatized the PM’s inability to control a major department.
Nor can PMs normally assume a dominant position in all three strategic areas
simultaneously. They may not even be able to staff all the relevant Cabinet
committees with reliable supporters. While Thatcher successfully packed the
economic committees in her first term government, her control over foreign
policy was weaker before 1982; for example, she had to accept the decisions
leading up to the independence of Zimbabwe that brought Robert Mugabe
to power.
Third, UK policy-making is increasingly influenced by the EU (see Chapters
5 and 8, this volume). There are also numerous other international agreements
(for example, policies to combat global ‘warming’, control sea dumping,
or regulate the security of air travel). This change has tended to erode the
concept of a discrete category of ‘foreign affairs’ in which the Prime Minister,

Foreign Secretary, and Foreign Office play the predominant roles. The early
post-war defence and treaty-based commitments (like NATO) buttressed the
insulation of strategic policy from wide ministerial involvement. But trends
since the 1970s have worked in the opposite direction by involving more and
more ministers and departments in direct negotiations with overseas coun-
terparts about sector specific policies (W. Wallace 1986). There has been a
progressive transition to joint UK–EU control of some key policies previously
controlled in Whitehall—such as monopolies and mergers policy, or the
setting of clean water and environmental standards (Ward et al. 1990). Such
policy areas have also become more important in wider EU relations because
of the transition to a single European market in 1993. These changes have
extensively constrained prime ministerial control over external relations pol-
icy. In short, although the segmented decision model does not answer the
question, it does ask if government policy can be effectively shaped by any
elements of the executive, either acting alone or in combination. The external
constraints may leave little room for domestic choices.
The Bureaucratic Coordination Model
The bureaucratic coordination model claims explicitly that the core executive
has limited control over the rest of the government machine, and that Cabinet,
and even individual departmental ministers, play minimal roles. Most policy
choices are effectively defined by the processing of issues within Whitehall.
There are two versions of this thesis, one left wing and the other associated
with the ‘fatalist’ new right.
The left view is the simplest and best known, portraying the civil service as
‘an elite arrogating to itself political power’ (Sedgemoor 1980: 26–32 and ch. 4),
which is an obstacle to the introduction of effective socialist reforms by Labour
governments (see also Benn 1981, especially ch. 3; Castle 1973; Crossman
1975: 23–6, 342–3, and 614–21; Kellner and Crowther-Hunt 1980: chs 4 and 5).
The key mechanisms for undermining more radical ministers have been
the enormous growth in the effectiveness of interdepartmental committees;
civil service manipulation of information flows to ministers; and the ability
to bid issues past ‘troublesome’ departmental ministers to the PM for ‘safer’
resolution. Sedgemoor (1980: 34) stresses ‘the convergence of bureaucracy’:
that is, the shared interests between the civil service, the EU bureaucracy,
the CBI, and the TUC in sustaining routinized bureaucratic control of
issue-processing. De-radicalized Labour premiers play a superficially import-
ant role in this account—but only as the final arbiter or tie-breaker in
deadlocked inter-agency conflicts or as the stooges for unified bureaucratic
interests. They are not genuine controllers of the policy machine or initiators
of new policy directions.
The new right version of this viewpoint provides an analogous explan-
ation of why the Heath government or even Thatcher’s new right govern-
ment have failed to make major cuts in public spending. Civil service
obstruction of radical measures reflects a strong bureaucratic drive to maxi-
mize budgets and oversupply outputs. But the mechanisms of civil service
power are much the same as in the left account, stressing bureaucratic
monopoly of information; the ability to tone down, blunt, or delay initia-
tives; and efforts to marginalize political advisers and initiatives in a rapidly
moving flow of short-term problems and issues. Bureaucratic conservators
can also successfully orchestrate vested interests to oppose ministerial
proposals threatening to the status quo. Strong pressures are brought to
bear upon departmental ministers to opt out of difficult reform tasks and
instead ‘go native’ in their fiefdoms. The prime minister and other non-
departmental ministers, together perhaps with the Treasury, are the only
actors likely to keep up new right pressure for micro-policy changes. But
they are vulnerable to pressures to reflate public spending as part of the
political–business cycle, such as the consumer spending boom orchestrated
in the run-up to the 1987 election.
There are several attempts to move beyond the prime minister versus
Cabinet debate in an effort to capture the complex of relations at the heart
of the machine. The ministerial government, segmented decision-making, and
bureaucratic coordination models all point to a view of the executive in which
there are multiple actors whose relative power shifts both over time and
between policy areas. In other words, it is factually inaccurate to assert, by
using the phrases prime ministerial or Cabinet government, that these insti-
tutions invariably and inevitably either coordinate government policy or
resolve central conflicts. ‘Who coordinates?’ is an empirical question. It is
quite possible that the prime minister and or the Cabinet play this role but the
point must be documented, not asserted. Unfortunately, there is no coherent
theoretical alternative to the prime ministerial power versus Cabinet govern-
ment debate to guide the search for evidence. Indeed, there isn’t a great deal of
empirical research. Case studies are the dominant research method and there
are precious few of them. Any review of the literature compels the conclusion
that we know little about the British executive.
As one way out of this theoretical and methodological impasse, I propose
that we adopt a differentiated model in which the relevant executive varies
over both time and policy area. There is no single executive but multiple
executives. The phrase ‘core executive’, because it refers to a range of central
institutions, captures this essential variability, which brings me to the question
of how to study this core executive.
WH ERE ARE WE GOING?
This section is structured around the defects of the existing literature, focusing
on the theories, methods, and research questions that need to be explored if we
are to repair the gaps in our knowledge. And there are many such gaps.
Basic accounts of several executive institutions are lacking. The examples
include: ministers of state, permanent secretaries, central departments, think-
tanks, and audit agencies. This collective ignorance also encompasses consti-
tutional change and extends to the role of the several executive institutions in
the policy process. Rapid change in British government and politics further
compounds the problem.
The post-war period witnessed: the growth of the welfare state; the profes-
sionalization of government; retrenchment under the impact of economic
recession and new Right ideology; the widespread impact of the EU; the
fragmentation of bureaucracy with the allied spread of new methods of service
delivery; the changing relationship with, and expectations of, citizens (now
known as consumers); the impact of new technology; and the ‘new public
management’ with its sharp divorce of policy and administration. We must
assess the impact of these and other changes on the British executive before
we can begin to explain their variable effects on, and results for, executive
behaviour.
There is little theoretical literature on the executive. For example, the new
public management or ‘managerialism’ leans heavily on the teachings and
techniques of private sector management. Their relevance to the public sector
continues to be a matter of debate but there is still no management or
organization theory developed explicitly for the context and purposes of the
public sector (Rhodes 1991a). Both empirically and theoretically, therefore,
there is much to be done.
D EV EL O P I N G T H E O R Y
There is a lack of theory in the study of the executive.4 Recently, two theor-
etical approaches deserve attention: the ‘bureau-shaping’ theory of depart-
mental budgetary behaviour (see, for example, Dunleavy 1989a, 1989b, 1991,
1992) and policy network theory of policy formulation and implementation
(see, for example, Marsh and Rhodes 1992a; Rhodes 1988).
4
See Rhodes 2017, Volume II, Chapter 7 for an updated review of the theoretical literature on
the core executive.
The bureau-shaping model significantly modifies the classic rational choice
view of bureaucrats. In the classic version, bureaucrats are rational, self-
interested actors seeking to maximize their agencies’ budget. In the bureau-
shaping model, the bureaucrats remain rational and self-interested actors but
their behaviour varies both with the type of budget and the type of agency.
Dunleavy’s (1991: ch. 4) bureau-shaping model of bureaucracy distinguishes
between both types of budget and types of agencies. He identifies four types of
budget: core (salary and running costs); bureau (core plus capital expenditure
and transfer payments direct to individuals and organizations); programme
(core and bureau budgets plus funding supervised by agency); and super-
programme (all foregoing plus supervision of funds raised by other agencies).
He also distinguishes between delivery, regulatory, transfer, contracts, and
control agencies. He argues that budget maximization by officials depends on
their rank, type of budget, and type of agency. Thus, a rational middle rank
bureaucrat will seek to maximize the core budget because it will improve job
security and enhance career prospects. On the other hand, the rational top rank
bureaucrat will maximize the bureau budget because it boosts bureau prestige.
Moreover, the incentive to maximize will be strongest where there is a close
relationship between core, bureau, and program budgets; as in the case of
delivery agencies. In other words, there is great variation between bureaucrats
in the extent to which they have incentives to maximize their budgets. In place
of budget maximization as an explanation of official behaviour, Dunleavy
introduces the notion of ‘bureau-shaping’. He argues that:
rational bureaucrats oriented primarily to work-related utilities pursue a bureau-
shaping strategy designed to bring their bureau into a progressively closer approxi-
mation to ‘staff ’ (rather than ‘line’) functions, a collegial atmosphere, and a central
location (Dunleavy 1991: 202–3).
Thus, national level delivery agencies will become control, transfer, or contract
agencies and the central bureau will take on a small, central, elite character. In
short, rational bureaucrats work in varied settings and have a choice of
maximizing strategies. They do not just maximize their budgets.
This model not only explains variations in budget maximizing behaviour
but it can also be put to empirical use. Dunleavy (1989b, 1991: 188–91,
213–17) shows that his agency and budget typologies can be operationalized
to describe the organization structure of British central government. In add-
ition, the typologies are being elaborated: for example, following Hood (1983),
each agency type is said to have a distinct set of tools for interacting with the
outside world (Dunleavy 1991). Thus, regulatory agencies rely on making
rules (authority) whereas delivery agencies have their own staff to implement
policy (administrative organization).
Dunleavy would be the first to admit that the bureau-shaping model is at an
early stage of development. However, it already offers a way of comparing, for
example, central agencies throughout the EU. More ambitiously, it ought to be

possible, in principle, to describe and explain variations in budgetary behav-
iour between member states. Finally, it can predict and explain administrative
change: for example, Dunleavy (1989a: 268) claims the model explains why
policy level bureaucrats favour hiving off parts of their departments as separ-
ate agencies. In short, the bureau-shaping model has great potential and
deserves exhaustive testing.5
Recent work in Britain on policy networks argued that organizations are
dependent upon each other for resources and have to exchange them to
achieve their goals. These sets of interacting, interdependent organizations
were described as ‘policy networks’.6 Policy-making by policy networks is a
form of private government. The relevant central department, interest groups,
and other governmental bodies regularly and routinely decide on policy and
only occasionally refer it to either parliament or the Cabinet. Policy networks
are alliances of bureaucrat and professional across types and tiers of govern-
ment, including the EU (see Chapter 4, this volume). In effect, each policy
network is the executive for its functional area of government. For example,
Rhodes (1988: 237–55) argues that local and central government are inter-
dependent. Although the centre enacts the law and provides a high proportion
of local income, local government has the professional skills and organiza-
tional resources to implement the policy. They need each other if a policy is to
work. However, in the 1980s, the Conservative government adopted a unilat-
eral style of policy-making. The network became highly conflictual and local
government resisted central interventions. A policy mess ensued in which
neither level of government could achieve their objectives. In denying their
interdependence, they frustrated each other’s actions.
Such interdependence is characteristic of the relationships in the (multiple)
executives of British government and of their relationships with the core
executive. So, a focus on power-dependence provides the means for exploring
the nature of the executive and its relationship to the core executive. Rather
than comparing decision-making by departments, future research should
compare the different types of networks, each with a central department at
its heart; their relationships to the core executive; and the ways in which they
5
For the published version, the following text was deleted from my original typescript:
‘Dunleavy pays a price for his more complex view of bureaucratic motivations; the loss of
both parsimonious explanation and predictive capacity.’ Moreover, the evidential base for his
assumptions that the motives of senior bureaucrats are ‘to work in small, elite, collegial bureaus
close to political power centres’ (Dunleavy 1991: 202) is at best weak and probably non-existent.
He claims that his list of motives is based on ‘the most common pro and anti values cited in the
administrative sociology literature’ (Dunleavy 1991: 201). However, there is not a single citation
to support that assertion, so it cannot be considered adequate evidence.
6
See Chapter 3, this volume for a summary of this literature on policy networks. I have
deleted the summary in the published version of this chapter.
adjusted to the pressures of the 1980s. This approach could not only describe
the changing patterns of internal and external relationships of a department
but also analyse the impact of organizational change.
The advantage of both these bureau-shaping and policy networks approaches
is that they offer new vantage points from which to view ‘conventional’ topics
such as ministerial accountability. Instead of taking one of the chestnuts of
the constitution—that is, the power of the prime minister, collective Cabinet
responsibility, ministerial accountability to parliament—as the unit of ana-
lysis, they use either a classification of agency and budget types or of policy
networks.
DEVELOPING METHODS
Traditional approaches to the study of the executive—for example, telling the

story of a current event or describing institutional and legal arrangements—
are no longer sufficient because they are not theoretical. There are other ways
of studying the executive. Three methods have great potential.
First, Dunleavy et al. (1990) show how to use quantitative methods to
document the frequency and type of appearance in the Commons by prime
ministers since 1868 (see also Dunleavy et al. 1995). The same techniques
could be used to compare the relationships of ministers to Parliament. More-
over, I provide only one illustration of the usefulness of quantitative methods.
They can be used in many other areas: for example, the changing ministerial
composition of Cabinet committees; and prime ministerial popularity and
ministerial resignations (Dunleavy 1995b; Hudson 1984).
Second, it is not necessary to use only a single case study. Yin (1984: 48–9)
argues that multiple case studies are analogous to experiments because they
follow the logic of replication: that is, each case predicts similar results or
produces contrary results for predictable reasons. A multiple case research
design is more robust and allows us to generalize with greater confidence.
There are no multiple case studies of the British executive although several
studies show that, without access to Cabinet papers, illuminating individual
cases can be written (Bruce-Gardyne and Lawson 1976; Barnett 1969; Linkla-
ter and Leigh 1986).
Finally, we need an archive of oral administrative history. More than 50
senior officials retired in the 1980s. Most will only speak off the record.
Interviews with them, stored if necessary under the 30-year rule, would
provide an invaluable research resource for future generations of researchers.
It is commonly alleged that access is the key problem limiting research in
the field. However, researchers with a proven record can surmount most
problems. Also, the times they are a-changing. There is a willingness to help
and grounds for believing there will be greater access to civil servants and
more information about the policy process (Butler 1992).
Multiple sources of data are already available including:
• Hansard (parliamentary debates and questions, select committee
hearings);
• white papers, green papers, and other official publications (including
official statistics);
• media reports including television documentaries as well as newspaper
reports and investigations;
• memoirs, autobiographies, and diaries;
• biographies;
• interviews with past and present ministers and officials;
• seminars under Chatham House rules;
• Cabinet papers (available after 20 years with earlier access to official
papers for historians on a case-by-case basis); and
• other secondary sources, whether written by participants, journalists, or
academics.
In sum, there is already much material in the public domain (see James 1992 for
an illustration of the scope for syntheses based on published material).
Undoubtedly all of these research methods and sources have some serious
limits for studying the executive. The choice of methods has been too conser-
vative in the past. Available sources have not been fully exploited. Secrecy and
restricted access may be a problem, but there is still a great deal of work that
can be done.7
CONCLUSIONS
The state of the art in British core executive studies leaves plenty of room for
improvement but there are grounds for expecting some progress. Upon close
inspection, the institutionalist literature has proved to be diverse and to
support a fragmented or differentiated interpretation of the British executive.
Further progress requires a focus upon: the many institutions that constitute
the core executive; the relative strengths and weaknesses of the different
theoretical approaches; more fieldwork to provide case studies of the core
7
I deleted the section on future research from the original version of this chapter.
executive in action; and greater methodological sophistication to make full use

of the data that is already available.
AFTERWORD
Obviously this chapter does not cover anything written after 1995. The
literature review may be out of date, but not much else in this review is
dated. So, the literature can still be categorized as institutionalist while the
favoured topics continue to include, for example, the presidential prime
ministerial; the decline of Cabinet government; and the vertical and horizontal
coordination of central departments and their networks—joined-up govern-
ment. Some of the gaps in research were filled, but not all. The ESRC’s
‘Whitehall Programme’ produced much authoritative research (Bellamy
2011; Rhodes 2000a, 2000b). The debate about hollowing out and central
capacity rumbles on; quantitative analysis is still a minority sport (but see
Dowding et al. 2012). We have more case studies. There are comparative
studies of government departments (Marsh et al. 2001). Yet, executive studies
remain a small subfield with relatively few aficionados. I update the history of
the subfield in Rhodes 2017, Volume II, Chapter 7. But what about the core
executive approach—is it of continuing relevance?
After two decades of core executive studies, Elgie (2011: 71–2) concludes
that ‘the language of the study of British central government has been trans-
formed’ by the approach; and ‘the concept has travelled’ well to the study of
other countries, and continues to do so (see Eymeri-Douzans et al. 2015;
Heilman and Stepan 2016). However, some core executive studies are ‘less
innovative than they might at first appear’ because they can appear as an
‘updated version of the old prime ministerial vs. Cabinet government argu-
ment’. More significant, Elgie claims ‘the resource-dependency approach is
almost completely dominant’. There is one clear advantage to the approach; it
gets away from bald assertions about the fixed nature of executive politics.
I identified six varieties above: prime ministerial government; prime minis-
terial cliques; Cabinet government; ministerial government; segmented
decision-making; and bureaucratic coordination. Elgie (1997: 231) similarly
identifies six models. While only one pattern may operate at any one time,
there can be much fluidity as one pattern succeeds another. It also con-
centrates the mind on the questions of what kind of executive politics prevails,
and when, how, and why it changed. Focusing on the power of prime minister
and Cabinet is limiting whereas these questions open the possibility of
explaining similarities and differences in executive politics.
There have also been some promising theoretical developments. Burch
and Holliday (1996, 2004) see the core executive as a set of interlocking
networks. They identify eight tasks focused on the Cabinet system, including,
for example, domestic policy, EU policy, legislation, and the civil service and
machinery of government. They identify the network supporting each task,
noting that some are small and exclusive while others are large and open. They
conclude that:
The contemporary cabinet system is a complex set of organisations and posi-
tions . . . It operates at formal, semi-formal and informal levels and is structured
not only by distinct values and practices but also by a range of networks and
processes (Burch and Holliday 1996: 275).
The prime minister is the focal point of these core networks; the innermost
network linking the set of networks that comprise the core executive. The
prime minister is supported by enhanced central capacity that increases
the power potential of the prime minister. They suggest that the power of
the centre has increased but ‘the enhancement of central capacity within the
British system of government reflects contingent factors, including the per-
sonalities of strategically placed individuals (notably, but not only, the PM)’.
They note that such changes are ‘driven by prime ministerial whim’ and ‘if
they so desire, [prime ministers] try to shape the core in their own image’.
However, the extent to which they can do so ‘depends on the motivation and
skill of key actors, and on the circumstances in which they find themselves at
any given moment in time’ (Burch and Holliday 2004, 17 and 20). As
Oakeshott (1962: 127) observed, prime ministers:
sail a boundless and bottomless sea; there is neither harbour for shelter nor floor
for anchorage, neither starting-place nor appointed destination. The enterprise is
to keep afloat on an even keel.
In sum, the argument for a broader focus than just prime minister and Cabinet
has stood the test of time. The core executive as a set of interlocking networks
in which the roles of actors and organizations are contingent and no one
pattern of executive politics prevails. To this point, I have told the story of the
core executive during the policy network years. However, the network formu-
lation underpins the analysis of the core executive as court politics (see Rhodes
2017, Volume II, Chapter 7). There is a challenge to the orthodoxy and it
comes from interpretive theory and ethnographic methods. It is to these
interpretive years that I turn in Volume II.
10
The New Governance

Governing without Government
INTRODUCTION
Over the past 15 years vogue words and phrases for reforming the public
sector have come and gone. ‘Rayner’s Raiders’ and the ‘3Es’ of economy,
efficiency, and effectiveness gave way to the ‘new public management’ and
‘entrepreneurial government’. This chapter focuses on one of these words:
‘governance’.1 It is widely used, supplanting the commonplace ‘government’,
but does it have a distinct meaning? What is it supposed to tell us about the
challenges facing British government?
Unfortunately, even the most cursory inspection reveals that ‘governance’ has
several distinct meanings. A baseline definition is essential, therefore, and where
else to look other than a textbook. Finer (1970: 3–4) defines government as:
• ‘the activity or process of governing’ or ‘governance’;
• ‘a condition of ordered rule’;
• ‘those people charged with the duty of governing’ or ‘governors’; and
• ‘the manner, method or system by which a particular society is governed’.2
Current use does not treat governance as a synonym for government. Rather,
governance signifies a change in the meaning of government, referring to a
new process of governing; or a changed condition of ordered rule; or the new
method by which society is governed.
1
Originally published as R. A. W. Rhodes (1996b) ‘The New Governance: Governing without
Government’, Political Studies, 44: 652–7, which was a revised version of Rhodes 1995c, a lecture
delivered to the RSA/ESRC Joint Initiative on The State of Britain, RSA, London, 24 January.
Reprinted with permission of John Wiley and Sons.
2
Low (1904) is an early example of the use of ‘governance’ in the analysis of British
government but the term is not in the index nor defined in the text; it is a synonym for
government.
The New Governance: Governing without Government 159
So far, so simple; but the problems of definition become acute when
specifying this new process, condition, or method. There are at least six
separate uses of governance:
• as the minimal state;
• as corporate governance;
• as the new public management;
• as ‘good governance’;
• as a socio-cybernetic system;
• as self-organizing networks.
Of course, words should have clear meanings but there is a more interesting
theme to my discussion.
The 1980s heralded a new chapter in the debate about ways of governing.
Analysing ‘governance’ will help to pin down the nature of this experiment
and to identify trends and contradictions in the evolution of the British state.
I argue that British government can choose between ‘governing structures’ (see
Chapters 11 and 12, this volume). To markets and hierarchies, we can now add
networks. None of these structures for authoritatively allocating resources and
exercising control and coordination is intrinsically ‘good’ or ‘bad’. The choice is
not necessarily or inevitably a matter of ideological conviction but of practical-
ity; that is, under what conditions does each governing structure work effect-
ively. Bureaucracy remains the prime example of hierarchy or coordination by
administrative order and, for all the recent changes, it is still a major way of
delivering services in British government; for example, the Benefits Agency
remains a large bureaucracy. Privatization, market testing, and the purchaser–
provider split are examples of government using market or quasi-market ways
of delivering services. Price competition is the key to efficient and better quality
services. Competition and markets are a fixed part of the landscape of British
government. It is less widely recognized, especially by British government, that
it now works through networks characterized by trust and mutual adjustment,
for example, to provide welfare services. British government is searching for a
new ‘operating code’. This search involves choosing between governing struc-
tures. Network governance is one such structure (see Chapter 11, this volume).
USES OF GOVERNANCE
Governance as the Minimal State
This use is a blanket term, redefining the extent and form of public interven-
tion and the use of markets and quasi-markets to deliver ‘public’ services. To
employ Stoker’s (1994: 6) apt phrase, ‘governance is the acceptable face of
spending cuts’. The extent of any change is a matter of dispute. Indisputably,

the size of government was reduced by privatization and cuts in the civil service.
However, public expenditure remained roughly constant as a proportion of
GDP; public employment rose slightly in local government and the national
health service; and regulation replaced ownership as the preferred form of
public intervention with the government creating ten major regulatory bodies.
Whatever the results in practice, the ideological preference for less government
was stated loudly and often (see Kavanagh 1990: 1–12). Governance encapsu-
lates that preference, but says little else, being an example of political rhetoric.
Governance as Corporate Governance
This specialized use refers to ‘the system by which organisations are directed
and controlled’ (Cadbury Report 1992: 15). Thus:
the governance role is not concerned with running the business of the company,
per se, but with giving overall direction to the enterprise, with overseeing
and controlling the executive actions of management and with satisfying
legitimate expectations for accountability and regulation by the interests beyond
the corporate boundaries. . . . All companies need governing as well as managing
(Tricker 1984: 6–7).
The Chartered Institute of Public Finance and Accountancy (CIPFA) adapted
this use to the public sector:
Developments such as compulsory competitive tendering, the creation of discrete
business units within internal markets and the introduction generally of a more
commercial style of management are bringing about a different culture and
climate, which represents a departure from the traditional public service ‘ethos’,
and its values of disinterested service and openness. The fact that public services
are currently undergoing significant change enhances the need for extra vigilance
and care to ensure that sound systems of corporate governance are both set in
place and work in practice (CIPFA 1994: 6).
Its report identifies three fundamental principles that apply equally to organ-
izations in the public and private sectors. They recommend openness or the
disclosure of information; integrity or straightforward dealing and complete-
ness; and accountability or holding individuals responsible for their actions by
a clear allocation of responsibilities and clearly defined roles. Although a
narrow use of the word, the concerns of corporate governance are echoed
when discussing accountability in the ‘new public management’ and ‘good
governance’. Also, this use reminds us that private sector management prac-
tice has an important influence on the public sector.3
3
My thanks to Andrew Dunsire (University of York) for pointing out this use and providing
several helpful references. Personal correspondence, 28 April 1994.
Governance as the New Public Management
Initially the ‘new public management’ (NPM) had two meanings: manager-
ialism and the new institutional economics (see Hood 1991; Pollitt 1993).4
Managerialism refers to introducing private sector management methods to
the public sector. It stresses: hands-on professional management, explicit
standards and measures of performance; managing by results; value for
money; and, more recently, closeness to the customer. The new institutional
economics refers to introducing incentive structures (such as market compe-
tition) into public service provision. It stresses disaggregating bureaucracies;
greater competition through contracting out and quasi-markets; and con-
sumer choice. Before 1988, managerialism was the dominant strand in Britain.
After 1988, the ideas of the new institutional economics became more
prominent.
NPM is relevant to this discussion of governance because steering is central
to the analysis of public management and steering is a synonym for govern-
ance. For example, Osborne and Gaebler (1992: 20) distinguish between
‘policy decisions (steering) and service delivery (rowing)’, arguing bureaucracy
is a bankrupt tool for rowing. In its place they propose entrepreneurial
government based on ten principles:
Most entrepreneurial governments promote competition between service pro-
viders. They empower citizens by pushing control out of the bureaucracy, into
the community. They measure the performance of their agencies, focusing not
on inputs but on outcomes. They are driven by their goals—their missions—not
by their rules and regulations. They redefine their clients as customers and offer
the choices . . . They prevent problems before they emerge, rather than simply
offering services afterwards. They put their energies into earning money, not
simply spending it. They decentralize authority, embracing participatory man-
agement. They prefer market mechanisms to bureaucratic mechanisms.
And they focus not simply on providing public services, but on catalysing all
sectors—public, private, and voluntary—into action to solve their community’s
problems.
Clearly NPM and entrepreneurial government share a concern with compe-
tition, markets, customers, and outcomes. This transformation of the public
sector involves ‘less government’ (or less rowing) but ‘more governance’ (or
more steering) (Osborne and Gaebler 1992: 34).5
4
Inevitably meanings proliferate and the term has been extended to cover the corporate
takeover of public services and post-bureaucratic public management. See: Dunleavy 1994;
Yeatman 1994; and Barzelay with Armajani 1992.
5
I do not use ‘steering’ as a synonym for policy decisions in this chapter but use it to refer to a
mode of control which involves setting a norm and correcting deviations from it (see Dunsire
1990). Also, it is useful to distinguish steering (the process) from directedness (the effect).
Andrew Dunsire, personal correspondence, 28 April 1994.
Governance as ‘Good Governance’
Government reform is a worldwide trend and ‘good governance’ is the latest

flavour of the month at the World Bank (1992), shaping its lending policy
towards third world countries.6 For the World Bank, governance is ‘the
exercise of political power to manage a nation’s affairs’ and ‘good governance’
involves:
an efficient public service, an independent judicial system and legal framework to
enforce contracts; the accountable administration of public funds; an independ-
ent public auditor, responsible to a representative legislature; respect for the law
and human rights at all levels of government; a pluralistic institutional structure,
and a free press (Leftwich 1993: 610).
Leftwich identifies three strands to good governance: systemic, political, and
administrative. The systemic use of governance is broader than government
covering the ‘distribution of both internal and external political and economic
power’. The political use of governance refers to ‘a state enjoying both legitim-
acy and authority, derived from a democratic mandate’. The administrative use
refers to:
an efficient, open accountable and audited public service which has the bureau-
cratic competence to help design and implement appropriate policies and manage
whatever public sector there is (Leftwich 1993: 611).
And to achieve efficiency in the public services, the World Bank seeks to:
encourage competition and markets; privatize public enterprise; reform the
civil service by reducing over-staffing; introduce budgetary discipline; decen-
tralize administration; and make greater use of non-governmental organiza-
tions (Williams and Young 1994: 87). In short, ‘good governance’ marries the
new public management to the advocacy of liberal democracy.
Governance as a Socio-cybernetic System
‘Socio-cybernetics’ is protected by the cloak of language, but I try to avoid

most of its neologisms.7 For Jan Kooiman (1993a, 1993b), governance:
can be seen as the pattern or structure that emerges in a socio-political system as
‘common’ result or outcome of the interacting intervention efforts of all involved
6
This section draws on the work of my late colleague at the University of York, Adrian
Leftwich, and I would like to acknowledge his help.
7
For ease of exposition, I focus on Kooiman (1993c) as the best recent collection of articles
about this approach. However, I must also mention the work of Vickers (1968) and Dunsire
(1986), both pioneers in applying cybernetics to British government.
actors. This pattern cannot be reduced to one actor or group of actors in
particular (Kooiman 1993b: 258).
In other words, policy outcomes are not the product of actions by central
government. The centre may pass a law but subsequently it interacts with
local government, health authorities, the voluntary sector, the private sector,
and, in turn, they interact with one another. Kooiman distinguishes between
the process of governing (or goal-directed interventions) and governance,
which is the result (or the total effects) of social-political-administrative
interventions and interactions. There is order in the policy area but it is not
imposed from on high emerging from the negotiations of the several affected
parties. Also,
These interactions are . . . based on the recognition of (inter)dependencies. No
single actor, public or private, has all knowledge and information required to
solve complex dynamic and diversified problems; no actor has sufficient overview
to make the application of needed instruments effective; no single actor has
sufficient action potential to dominate unilaterally in a particular governing
model (Kooiman 1993a: 4).
So, all the actors in a particular policy area need one another. Each can
contribute relevant knowledge or other resources. No one has all the relevant
knowledge or resources to make the policy work. Governing confronts new
challenges:
Instead of relying on the state or the market, socio-political governance is directed
at the creation of patterns of interaction in which political and traditional
hierarchical governing and social self-organisation are complementary, in
which responsibility and accountability for interventions is spread over public
and private actors (Kooiman 1993b: 252).
Central government is no longer supreme. The political system is increasingly

differentiated. We live in ‘the centreless society’ (Luhmann 1982: xv); in the
polycentric state characterized by multiple centres. The task of government is
to enable socio-political interactions; to encourage many and varied arrange-
ments for coping with problems; and to distribute services among the several
actors. Such new patterns of interaction abound: for example, self- and co-
regulation, public–private partnerships, cooperative management, and joint
entrepreneurial ventures.
This use is not restricted to national governance, encompassing also the
international system. For example, James Rosenau distinguishes government
from governance by suggesting that government refers to ‘activities that
are backed by formal authority’ whereas governance refers to ‘activities
backed by shared goals’. Governance is ‘a more encompassing phenomenon’
because it embraces not only governmental organizations but also ‘informal,
non-governmental mechanisms’. So, you get governance without government
when there are ‘regulatory mechanisms in a sphere of activity which function

effectively even though they are not endowed with formal authority’ (Rosenau
1992a: 3–6).
The socio-cybernetic approach highlights: the limits to governing by a
central actor, claiming there is no longer a single sovereign authority. In its
place, there is the multiplicity of actors specific to each policy area; inter-
dependence among these social-political-administrative actors; shared goals;
blurred boundaries between public, private, and voluntary sectors; and multi-
plying and new forms of action, intervention, and control. Governance is the
result of interactive social-political forms of governing.
Governance as Self-organizing Networks
The system of government beyond Westminster and Whitehall was trans-

formed ‘from a system of local government into a system of local governance
involving complex sets of organisations drawn from the public and private
sectors’ (Rhodes 1988, 1991b). This use sees governance as a broader term
than government with services provided by any permutation of government
and the private and voluntary sectors. Interorganizational linkages are a
defining characteristic of service delivery. I use the term network to describe
the several interdependent actors involved in delivering services. These net-
works are made up of organizations that need to exchange resources
(for example, money, information, expertise) to achieve their objectives, to
maximize their influence over outcomes, and to avoid becoming dependent on
other players in the game (Rhodes 1988: 42–3). As British government creates
agencies, bypasses local government, uses special purpose bodies to deliv-
er services, and encourages public–private partnerships, so networks become
increasingly prominent among British governing structures. Indeed, Metcalfe
and Richards (1991: 220) define public management as ‘getting things done
through other organisations’ and criticize managerial reform in Britain
for concentrating on internal management. Governance is about managing
networks (see also, for example, Agranoff 1990; Friend et al. 1974; Hanf and
Scharpf 1978; Kickert 1993b; and Marsh and Rhodes 1992a).
Network management is not specific to the public sector. To define public
management as getting things done through other organizations is not an
implicit argument against the use of markets and quasi-markets. Networks
are a widespread form of social coordination, and managing interorganiza-
tional links is just as important for private sector management (Thompson
et al. 1991: chs 14–24). Powell (1991) argues that networks are ‘a distinctive
form of coordinating economic activity’. In similar vein, Larson (1992)
explores ‘network structures in entrepreneurial settings’, concluding that ‘the
network form of governance’ highlights ‘reputation, trust, reciprocity and
mutual interdependence’. So, networks are an alternative to, not a hybrid of,
markets and hierarchies, and they span the boundaries of the public, private,
and voluntary sectors:
If it is price competition that is the central co-ordinating mechanism of the
market and administrative orders that of hierarchy, then it is trust and co-
operation that centrally articulates networks (Frances et al. 1991: 15).
More important, this use of governance also suggests that networks are self-
organizing.8 At its simplest, self-organizing means a network is autonomous
and self-governing:
The control capacity of government is limited for a number of reasons: lack of
legitimacy, complexity of policy processes, complexity and multitude of institu-
tions concerned etc. Government is only one of many actors that influence the
course of events in a societal system. Government does not have enough power to
exert its will on other actors. Other social institutions are, to a great extent,
autonomous. They are not controlled by any single superordinated actor, not
even the government. They largely control themselves. Autonomy not only
implies freedom, it also implies self-responsibility. Autonomous systems have a
much larger degree of freedom of self-governance. Deregulation, government
withdrawal and steering at a distance . . . are all notions of less direct government
regulation and control, which lead to more autonomy and self-governance for
social institutions (Kickert 1993a: 275).
In short, integrated networks resist government steering, develop their own
policies, and mould their environments.
Governance: A Stipulative Definition
It would seem that governance has too many meanings to be useful, but the
concept can be rescued by stipulating one meaning and showing how it
contributes to the analysis of change in British government. So, governance
refers to self-organizing, interorganizational networks.
Any stipulative definition is arbitrary but my definition incorporates sig-
nificant elements of the other uses, most notably governance as the minimal
state, as a socio-cybernetic system and as self-organizing networks. I list below
the shared characteristics of ‘governance’.
8
This notion is closely related to that neologism to end all neologisms: autopoiesis. Autop-
oietic theory argues that organizations seek ‘self-referential closure’ with their environments; that
is, an organization’s relationship with its environment is internally determined and change is
internally generated. For a short introduction, see Morgan 1986: 235–45. On the several variants
of autopoietic theory, see Jessop 1990: ch. 11; Kickert 1993; and Luhmann 1986.
(1) Interdependence between organizations. Governance is broader than

government, covering non-state actors. Changing the boundaries of
the state meant the boundaries between public, private, and voluntary
sectors became shifting and opaque.
(2) Continuing interactions between network members, caused by the need
to exchange resources and negotiate shared purposes.
(3) Gamelike interactions, rooted in trust and regulated by rules of the
game negotiated and agreed by network participants.
(4) A significant degree of autonomy from the state. Networks are not
accountable to the state; they are self-organizing. Although the state
does not occupy a privileged sovereign position, it can indirectly and
imperfectly steer networks.9
This list clearly demonstrates the common ground between my definition and
the other uses. However, the key test of any stipulative definition is the
contribution it makes to understanding change in British government in the
1990s. The following analysis of change in British government treats networks
as a governing structure—as an alternative to markets and hierarchies. It
explores the extent to which they already exist and discusses the consequences
for British government.
SELF-ORGANIZING N ETWORKS AND BRITISH

GOVERNMENT
Governance contributes to the analysis of British government because it raises

new questions about recurrent problems. I explore its potential contribution
under the headings of: hollowing out the state; the contradictions of the new
public management (NPM); and the rise of intergovernmental management
(IGM).10
9
Andrew Dunsire in his discussant’s note on Rhodes (1995c) regretted that my definition
left out ‘some element’ of steering by the state:
I would begin from Rhodes’ formulation but add the means by which the government actor
may (however indirectly and imperfectly) steer the transient dynamics of network oper-
ations away from undesired configurations and towards desired ones—with no privileged
position or authority save what is readily acknowledged.
I concede this point implicitly in my discussion of intergovernmental management and steering,
so I amended this list of network characteristics to accommodate Dunsire’s point before the
article was published.
10
Many of the theoretical issues were explored by the ESRC’s Whitehall programme, which
analyzed the changes in British central government in the post-war period. See Rhodes 2000a.
This chapter will not document the pace and extent of change in British
central government in the 1980s and 1990s, but there have been significant
changes, which pushed back the boundaries of the state; reasserted political
authority; improved monitoring and evaluation; reformed public sector man-
agement, increased the transparency of the public sector; reformed the struc-
ture; and changed the civil service culture (see Rhodes 1997c; Wright 1994;
and Chapter 8, this volume). Obviously, the Conservative government revisited
old problems. The reform of government structure, improving management
in government, strengthening central capability, the gap between central policy
objectives and local implementation, and the accountability of quangos all have
long histories. However, the Conservative government evolved a distinctive
strategy for reforming the public sector. Initially, for example, it railed against
the plethora of special purpose bodies in British government, only to use
them extensively later to bypass local authorities and to fragment service
delivery systems. The recurrent motifs in its reforms are competition and
markets. Without denying the persistent nature of many of the problems of
British government, this article focuses on the distinctive changes made by the
Conservative government.
Hollowing Out the State
I discussed ‘the hollowing out of the state’ in detail in Chapter 8, this volume,
and I will not repeat that discussion here. In brief, the public sector is
becoming both smaller and fragmented by privatization, new service delivery
systems, the EU, and public management reform and these trends raise the
problems of fragmentation, steering, and accountability. As networks multi-
ply, so do doubts about the centre’s capacity to steer. Kettl (1993: 206–7)
argues that, as a result of contracting out, government agencies found them-
selves ‘sitting on top of complex public–private relationships whose dimensions
they may only vaguely understand’. They had only ‘loose leverage’ but remained
‘responsible for a system over which they had little real control’.
The hollowing out of the state is another way of describing the problems of
managing interorganizational networks in British government. Interdepend-
ence, fragmentation, the limits to central authority, agency autonomy, and
attenuated accountability are all features of governance. Governance is rele-
vant to British government because self-organizing interorganizational net-
works are already part of the landscape of British government.
The Contradictions of NPM
NPM as managerialism extols the virtues of private management practice.

There are important lessons to be learnt from private management, but they
are not the lessons proselytized by supporters of managerialism. The new
governance points to four weaknesses in NPM.
First, and most obvious, managerialism adopts an intra-organizational
focus. It concentrates on the ‘3Es’ and value for money; on hierarchical control
and the clear distribution of authority and responsibility. Although appropri-
ate for managing line bureaucracies, and there are many left in British
government, this approach pays no attention to managing interorganizational
links; to negotiating shared purposes where there is no hierarchy of control.
Second, managerialism is obsessed with objectives; it resurrects management
by objectives for the 1980s and 1990s. Vickers (1968: 30) compares such goal-
seeking to the behaviour of rats in a maze, arguing that the distinctive feature of
human decision-making is regulation, or maintaining relationships in time. This
shift of emphasis is crucial for managing network relationships. For example, the
diplomatic skills of preserving the relationship between the local authority social
services department and the private sector provider of domiciliary care are more
significant than strict adherence to specific contractual targets. Preserving trust
overrides all other considerations.
Third, NPM focuses on results. In an interorganizational network, no
one actor is responsible for an outcome; there may be no agreement on either
the desired outcome or how to measure it; and the centre has no means of
enforcing its preferences. There is ‘the problem of many hands’ where so many
people contribute that no one contribution can be identified; and if no one
person can be held accountable after the event, then no one needs to behave
responsibly beforehand (Bovens 1990: 115). The style of management varies
with the governing structure. NPM may suit line bureaucracies but it is
inappropriate for managing interorganizational networks and, more import-
ant, such networks undermine NPM with its intra-organizational focus on
objectives and results.
Finally, there is a contradiction between competition and steering at the
heart of NPM. For example, Training and Enterprise Councils (TECs) were
described as a market solution to remedy the defects of industrial training. In
reality, TECs were the nodal point of a network coordinating multiple stake-
holders. The problem was the low level of interdependence between the several
stakeholders and the resulting network was unstable and lacked the trust
necessary to develop a negotiated equilibrium. The language of markets and
competition serves only to compound the problem of steering.
The Rise of IGM
Effective governance requires a re-examination of the government’s toolkit.

Steering (the process of setting norms) is separated from directedness (the
outcome of that process). The government needs tools to bridge that gap.
Intergovernmental management claims to provide them.
According to Deil Wright (1983: 481), IGM has three distinctive features:
problem-solving, intergovernmental games, and networking. It is about coping
with several jurisdictions to solve particular problems and building networks of
communication to produce such useful results. According to Agranoff (1990:
23–4), it is about separate organizations developing joint actions and finding
‘feasible courses of joint management activity’, although others argue that the
scope and boundaries of the subject remain unclear (see also Marando and
Florestano 1990; and Chapter 4, this volume). However, American and Dutch
research has begun to identify the kinds of techniques that work.
Klijn et al. (1995) argue that networks do not respond to managers as
system controllers. The effective manager plays a facilitative role; that is,
does not seek to achieve his or her own objectives. He or she can pursue two
broad strategies: game management or identifying the conditions that will
sustain joint action; and network structuring, which involves changing the
rules of the game. For example, promoting win–win situations in which
everybody gains some benefit will foster joint action, even though many
network actors will not achieve their initial aims. Alternatively, changes in
the distribution of resources within the network can be used to encourage
some kinds of behaviour, to introduce new actors to the network or to
undermine other actors.
In a similar vein, Agranoff (1990: 25–6) identifies 12 management
approaches to IGM. They include: ‘grantsmanship’, or the several members
of the network acquiring grants from several sources for numerous purposes;
‘process revision’, or ‘smoothing grant management through managerial process
changes, such as joint applications’; ‘bargaining and negotiation’; ‘problem-
solving’ through ‘mutual adjustment’; ‘co-operative management’, or manage-
ment by agreement; and ‘political games’ such as lobbying (see also Bogason
1995 and White 1989).
There is some evidence of equivalent behaviour in the UK. Friend et al.
(1974: 43–4) analyse the problem of Birmingham overspill and the agreement
to expand Droitwich. They identify decision networks or ‘open networks of
communication among people acting either within or across the interface
between them’ and stress the importance of reticulists who occupy ‘nodal
positions in the wider decision network’ and make key judgements about
linkages or what to communicate to whom. Rhodes (1986a: 392–3) uses the
notion of a policy network to explore the relationship between central and
local government and describes both the rules of the game and strategies used
by central and local government in the intergovernmental network. The
strategies for managing the relationship include: incorporation, consultation,
bargaining, avoidance, incentives, persuasion, and professionalization. The
gamelike quality of network management is not specific to federal systems.
Bureaucracy remains an important governing structure in Britain but

administrative orders do not work for all policy areas in all circumstances;
they are as likely to provoke avoidance and confrontation as cooperative
action. Market solutions to long-standing problems are the distinctive innov-
ation of the Conservative government. Policies such as contracting out deliver
significant cost and labour savings. But as for bureaucracy, market solutions
do not work for all policy areas in all circumstances. Networks are a third
governing structure. They are not better than either bureaucracies or markets.
They have different characteristics and suit some policy areas some of the
time. Reciprocity and interdependence, not competition, characterize network
relations. If there is one phrase that captures the nature of IGM, it is ‘mutual
adjustment’. So, managing interorganizational networks is both gamelike and
requires strategies rooted in trust. Planning, regulation, and competition need
to be supplemented with facilitating, accommodating, and bargaining, the
keys to effective network management (see Dunsire 1993; Kaufman et al.
1986; and Chapter 5, this volume).
CONCLUSIONS
Governance defined as self-organizing interorganizational networks does help

us to understand change in British government. First, it identifies the extent to
which networks already deliver services and the resulting problems of steering
and accountability. Such problems are not new but the reforms of the 1980s
and 1990s deliberately fragmented service delivery systems, generating func-
tional imperatives for interorganizational coordination. These networks are
self-organizing and the centre’s capacity to regulate them remains undevel-
oped; it has only ‘loose leverage’. Second, it shows the limits to managerial
reforms, which stress either intra-organizational control and management by
objectives or competition, masking the need for trust and cooperation with the
language of the market. Third, it suggests that networks require a distinctive
managerial style based on facilitation, accommodation, and bargaining.
Focusing on governance can blur, even dissolve, the distinction between
state and civil society. The state becomes a collection of interorganizational
networks made up of governmental and societal actors with no sovereign actor
able to steer or regulate. Governance as self-organizing networks is as distinct
a governing structure as markets and hierarchies. A key challenge for govern-
ment is to enable these networks and seek out new forms of cooperation. The
central lesson of this development is that:
the outcomes of administrative action are in many areas not the outcomes of
authoritative implementation of pre-established rules, but rather the results of a
‘co-production’ of the administration and its clients (Offe 1984: 310).
The challenge for British government is to recognize the constraints on central
action imposed by the shift to self-organizing networks; and to search for new
tools for managing such networks. Game-playing, joint action, mutual adjust-
ment, and networking are the new skills of the public manager (and see
Chapter 4).
However, there is a danger in these remarks. They treat networks as a tool of
government. But they are not just a mechanism for delivering services. As
important, they are a challenge to democratic accountability. Some specific
problems of accountability were noted earlier (see, for example, Chapters 3
and 4, this volume) but there are bigger problems. The conventional account
of policy networks treats them as an instance of private government, arguing
that networks:
destroy political responsibility by shutting out the public; create privileged oli-
garchies; and are conservative in their impact because, for example, the rules of
the game and access favour established interests
(Marsh and Rhodes 1992a: 265; see also Lowi 1969: 85–97 and 287–97).
An alternative interpretation suggests citizens could be regaining control of
government through their participation in networks as users and governors,
thereby creating a ‘postmodern public administration’:
Networks of publicly interested discourse which transcend hierarchical institu-
tions provide a feasible model for public administration. Some policy networks,
interagency consortia, and community task forces exhibit potential for discourse.
In these nascent forms are found think tank experts, legislative staff, policy
analysts, public administrators, interested citizens, process generalists, even elect-
ed officials participating together to work out possibilities for what to do next
(Fox and Miller 1995: 149).
In a similar vein, Rosenau (1992b: 291) argues that governance empowers
citizens:
Given a world where governance is increasingly operative without government,
where lines of authority are increasingly more informal than formal, where
legitimacy is increasingly marked by ambiguity, citizens are increasingly capable
of holding their own by knowing when, where, and how to engage in collective
action.
There are important limits to the new role of citizen as user. There are
significant constraints on authentic discourse within the network. Govern-
ments still restrict access to information and there are clear limits to the
knowledge of citizens. There is an obvious conflict between the tenets of
accountability in a representative democracy and participation in networks
that can be open without being formally accountable. These differing views of
networks pose different challenges for the public manager. Is their role to
regulate networks (in the sense of maintaining relationships)? Do they act as
guardians of the public interest? Do they still have the authority and legitimacy
to claim a privileged position in the network? Can they be privileged actors in

the network without undermining the discourse?
These speculative comments may only introduce the topic of the relation-
ship between governance and democratic accountability but, nonetheless, they
are essential to demonstrate the extent of the challenge posed by governance as
self-organizing interorganizational networks. The study of networks raises
issues of equivalent importance to the study of bureaucracy and democratic
accountability, but it gives these issues a distinctive twist. For example, the
assumption of institutional hierarchy, which underpins so many discussions
of bureaucratic accountability, no longer holds. Accountability can no longer
be specific to an institution but must fit the substantive policy and the several
institutions contributing to it. The ‘problem’ of a network governing structure
cannot be reduced to developing a new style of public sector management.11
Interorganizational networks are already widespread. This trend is not
widely recognized and has important implications not only for the practice
of British government but also for democratic accountability. Governance as
self-organizing networks is a challenge to governability because the networks
become autonomous and resist central guidance. They are set fair to become
the prime example of governing without government.
AFTERWORD
According to the LSE Public Policy Group (2011: 68), the Publish or Perish
website (www.harzing.com) is ‘a most valuable programme that combats
many of the problems of interpreting Google Scholar outputs’. As of
31 December 2015, it records this paper as my most cited item with 3,855
citations. It became a central chapter in Understanding Governance (1997a). It
was part of a series of articles I wrote in the mid-1990s, including Rhodes
(1997b), which, frankly, I should have included in Understanding Governance.
It would have avoided several misunderstandings of my argument (as in Peters
and Pierre 2000; Torfing et al. 2012). The revised version of Rhodes (1997b)
follows as Chapter 11. It makes little sense to reconsider this chapter separately
from the articles and the book written at the same time, so I reserve all
my comments for Chapter 12.
11
Hirst (1994) recommends associational democracy in which ‘voluntary self-governing
associations’ are the locus of both democracy and service delivery. Other governing structures
include, for example, solidarity or ‘acting according to common values and duties while
neglecting the price’. See Gretschmann (1986: 395). I note these normative accounts of governing
structures in this chapter to broaden the discussion beyond management issues.
11
It’s the Mix That Matters

From Marketization to Diplomacy
If economics is the dismal science, then politics studies the ‘sour laws of
unintended consequences’ (Hennessy 1992: 453). This chapter focuses on
the unintended consequences of marketizing public services and other public
sector reforms. It analyses the mix of markets, hierarchies, and networks (or
governing structures) in the differentiated polity; and explores the limits to
marketization and the prospects for ‘diplomacy in governance’.1
British government changes. The tradition of the strong executive encap-
sulated in the Westminster model founders on the complex maze of institu-
tions that deliver services. Interdependence confounds centralization. More
control is exerted, but over less. Services continue to be delivered, but by
networks of organizations that resist central direction. There are plenty of
organizations that government can only imperfectly steer. We live in a ‘centre-
less society’ (Luhmann 1982: xv and 253–5), referred to here as ‘the differenti-
ated polity’. This organizing perspective provides ‘a framework for analysis, a
map of how things relate; a set of research questions’ (Gamble 1990: 405; see
also Greenleaf 1983: 3–8; Tivey 1988: 3). It follows, therefore, that an organ-
izing perspective is always partial; it is not falsifiable, it is more or less accurate,
and it never provides a comprehensive or even definitive account. It is
1
An abbreviated version of an article published as R. A. W. Rhodes (1997) ‘It’s the Mix that
Matters: From Marketization to Diplomacy’, Australian Journal of Public Administration, 56:
40–53. Reprinted with permission of John Wiley and Sons. The first section of the paper
described policy networks, hollowing out, and governance, topics already covered in detail in
this volume. So, I deleted this summary to avoid undue repetition. An early version of this paper
was delivered on a lecture tour for the Institute of Public Administration Australia, 28 October–8
November 1996. It was delivered first on 30 October 1996 to the State Conference of the Institute
of Public Administration Australia (Queensland Division). The Afterword to this chapter draws
on Jenny Fleming and R. A. W. Rhodes (2005) ‘Bureaucracy, Contracts and Networks: The
Unholy Trinity and the Police’, Australian and New Zealand Journal of Criminology, 38:
192–205. © 2005 Sage Publications. Reprinted with permission.
a map and such maps ‘can guide . . . even when they are and are known to be
grossly inaccurate’. Maps can be corrected on the way. But such:
emendations and additions were only able to afford the guidance that they did in
fact afford by being ancillary to the original map. Lacking that map, there would
have been no focus, no way of organizing, often no way of characterising the
items in the list of emendations and additions
(Macintyre 1983: 32; Loughlin 1992: 37–8; Rhodes 1997a: ch. 9).
Simplification is essential because the system ‘is too extensive, too complex,
too fast changing to be observed except through the simplifying lens provided
by the interpreters’ (Tivey 1988: 16). The key criteria for evaluating an
organizing perspective, therefore, include but are not limited to its factual
accuracy. More important, it also identifies what is worthy of study. In other
words, the questions it poses are as important as the answers, because they
focus on features of the polity all too commonly ignored.
The Westminster model is an organizing perspective, which captures
some essential features of British government and, through sheer longevity,
it has become the conventional or mainstream view. It focuses on: parliamen-
tary sovereignty; strong Cabinet government; accountability through elec-
tions; majority party control of the executive (that is, prime minister,
Cabinet, and the civil service); elaborate conventions for the conduct of
parliamentary business; institutionalized opposition; and the rules of debate
(Gamble 1990: 407; Weller 1989). Birch (1964: 65) provides perhaps the best
short summary:
This [Liberal] view . . . comprised four distinct but interrelated doctrines. First,
there was the theory of representation . . . the eventual aims of which were crudely
expressed in the popular slogan ‘one man one vote; one vote one value’. Second,
there was the doctrine of Parliamentary sovereignty, combined with the belief
that any conflict between the two Houses the views of the Commons ought to
prevail . . . Third, Liberals insisted that ministers of the Crown were accountable
to Parliament for their actions. . . . Only in this way . . . could the political system
provide for responsible as well as representative government . . . Fourth, Liberals
attached great value to certain legal principles that came to be known as ‘the Rule
of Law’ (Birch 1964: 65).
Birch’s contribution also draws attention to another language of the Consti-
tution: ‘the other language is used by civil servants, the Speaker, Ministers of
the Crown and opposition leaders who hope soon to become Ministers’ (Tivey
1988: 58, quoting Birch 1964: 165). The essence of this language lay in its
emphasis on the continuing power and responsibility of the Government as
the guardian of national well-being.
The Westminster model focuses on institutions—that is, the rules, proced-
ures, and formal organizations of government. It has a shared set of meth-
odological assumptions, which involve using the inductive tools of the lawyer
It’s the Mix That Matters: From Marketization to Diplomacy 175
and the historian to explain the constraints on both political behaviour and
democratic effectiveness. There is a strong normative or reform strand, fos-
tering representative democracy. As Gamble (1990: 409) highlights, the West-
minster model is idealist, seeing ‘institutions as the expression of human
purpose’ and focusing, therefore, on the interaction between ideas and insti-
tutions (see, for example, Johnson 1975: 276–7). Finally, the Westminster
model makes some important if implicit assumptions about power. As
Smith (1996: 6–9) argues, the model focuses on behaviour, motivations, and
individuals. Power is an object that belongs to the prime minister, Cabinet, or
civil service. So, ‘power relationships are a zero-sum game where there is a
winner and a loser’ and power is ‘ascribed to an institution or person and fixed
to that person regardless of the issue or the context’. Personality is a key part of
any explanation of an actor’s power (and for a more detailed discussion of
these characteristics of the Westminster model, see Rhodes 1997a: ch. 1).
The Westminster model was, and remains, part of mainstream political
science. I do not present a caricature (see, for example, Norton 1983, 1996).
There is a growing diversity of approaches, although the hold of the West-
minster model ‘has not disappeared, nor has it been replaced by a coherent
alternative’ (Gamble 1990: 419). This chapter presents an alternative to the
Westminster model. It replaces strong Cabinet government, parliamentary
sovereignty, and ministerial responsibility with policy networks, the core
executive, hollowing out, and governance. The shorthand phrase for this
organizing perspective is the ‘differentiated polity’.
THE ‘DIFFERENTIATED P OLITY ’
This section defines briefly the differentiated polity organizing perspective (see
Rhodes 1988: 387–413 and figure 5.3; and Rhodes et al. 2003). Differentiation
refers to the process of functional and institutional specialization and the
consequences of that process. A ‘differentiated polity’ is characterized, there-
fore, by functional and institutional specialization and fragmenting policies
and politics. The most common, but by no means the only, form of special-
ization in British government is the functional policy network. The results of
the process are an increase in complexity and loss of central steering capacity.
In short, just as the Whitehall view is a necessary corrective to the Liberal view,
so the differentiated polity is a necessary corrective to both these maps. I tell
this story so we can see things differently by pointing to new connections in
governance and new aspects of governance. The central concepts of the
differentiated polity are policy networks, hollowing out, the core executive,
and governance. I have defined all these terms already in earlier chapters.
Here, I focus on the characteristics of governance to show that networks are a

pervasive feature of British government in the 1990s, and an alternative
governing structure to markets and hierarchies.2
THE CHARACTERISTICS OF GOVERNANCE
With the formal definition of terms to one side, I can now discuss the
characteristics of governance, identify the dilemmas raised by each character-
istic, and provide an example (see Table 11.1).
Institutional Complexity
The fragmentation of British government is plain for all to see. For example, in
2009, there were 766 non-departmental public bodies sponsored by the UK
Government employing 111,000 people in 2009 and spending £46.5 billion.
Add in privatization, services contracted out to the private and voluntary
sector, and functions run by the EU and the extent of service fragmentation
is still understated (see Chapter 8). I can best illustrate by sketching the
implementation structure for AIDS policy in the York–Selby area of North
Yorkshire. Thirteen organizations planned the service and 39 organizations
were involved in delivering services. Yet there were 24 HIV positive individ-
uals in the area and only six had developed AIDS (Battista 1994).
Table 11.1 The characteristics of governance.

Characteristic Dilemma Example
Institutional Confusion and uncertainty AIS (Battista 1994)

complexity
Dependence Command vs. Poll tax (Butler et al. 1994)
interdependence
Game playing Marketization vs. trust Community health services (Flynn et al. 1996)
Self-governing Accountability Wessex RHA (CPA 1994)
Steering Rubber levers Department of National Heritage (Taylor 1997)
Source: Rhodes 1996a, 1996b; and Stoker 1997.
2
King (2001: 99) interprets the changes as a shift from a power-hoarding to power-
fractionated system where ‘to fractionate’ is to break into fragments. The label differentiated
polity is sufficiently well-established for another label to be unwarranted.
The obvious, if not the only, consequence of this complexity is the dilemma
of confusion and uncertainty for both producers and users (see, for example,
the case of Mrs K in Chapter 6, this volume). Most people’s map of British
government does not admit to this degree of institutional differentiation.
Finding your way around the system is not straightforward; there is no one
point of contact. The question of ‘who provides what for whom’ admits of no
easy answer, although it is clear that the simple nostrums of the Westminster
model do not apply.
Dependence
Services are now delivered by sets of organizations that depend on each other for
resources. In sharp contrast to a bureaucracy, for most welfare state services
there is no hierarchy of authority; no one person can legitimately issue com-
mands and expect compliance. Central departments, local authorities, health
authorities, voluntary organizations, and the private sector must cooperate with
one another to provide the legal authority, finance, expertise, and organization
necessary for delivering a service. To ignore the brute fact of interdependence is
to risk the withdrawal of cooperation and fuel active noncompliance.
There can be no finer example of the dilemmas of dependence than local
government finance. The government adopted a command operating code
and told local authorities their spending must fall in real terms. The policy
objective was clear. The government was determined to judge by the sheer
volume of local government legislation and punitive means used (see Rhodes
1997a). The stated policy objective was not achieved. Between 1979 and 1995,
local government current spending in real terms rose by 33 per cent, and by at
least 2 per cent a year. A neutral observer may think the government’s record
‘uninspiring’. With the poll tax, it moved from uninspired to outright failure.
The tax-fuelled violent street demonstrations, a massive outcry from the
government’s own supporters, and a dramatic increase in both administrative
costs and non-payment of tax bills, all contributed to its eventual abolition.
But, from the government that refused to throw money at problems, the most
dramatic response was the increase in Value Added Tax from 15 per cent to
17.5 per cent to pay for a subsidy to reduce poll tax bills. Butler et al. (1994:
165, 175) estimate the subsidy at some £9 billion. The previous Labour
government set up the Consultative Council for Local Government Finance to
negotiate a reduction in local spending, and succeeded (Rhodes 1986a: ch. 4).
The simple, even obvious, conclusion is that, in networks where central and
local government are interdependent, commands provoke non compliance
whereas negotiation produces co-operation. The poll tax vividly illustrates the
high costs of imposition.
Game Playing
Introducing the purchaser–provider split and contracting out in the National

Health Service (NHS) illustrates the inevitability of game-playing or strategic
behaviour. Community Health Services (CHS) were one of the three arms of
the NHS with hospitals and general practice. CHS encompass chiropody,
health visiting, family planning, school health, district nursing, community
midwifery, immunization and vaccination, health education, speech therapy,
and community dentistry. The services were provided by professionals based
in either hospitals or health centres and clinics. Flynn et al. (1996) studied the
CHS in three district health authorities (DHAs).
The characteristics of CHS made them difficult services to contract out.
These features were: ‘their closeness to the lay sector of informal care; their
locality orientation; and the diffuse, non-acute health needs’. CHS were
‘difficult to define and measure with the kind of precision necessary for the
setting of formal contracts between purchasers and providers’ (Flynn et al.
1996: 24). So:
virtually none of the conditions relevant to the virtues of contracting-out . . . apply
to NHS contracts: service specification is extremely problematic; implementation
will entail ‘haggling’; . . . monitoring is complex and administratively costly; and
penalties for non-compliance carry high administrative and political costs
(Flynn et al. 1996: 21).
Contracting out of community health services faces three problems even more
acutely: ‘specifying the amount to be purchased, fixing a price for the services
and evaluating the quality provided’ (Flynn et al. 1996: 19).
Game playing took two forms. Purchasers alternated between adversarial
and collaborative styles of negotiation, although ‘there were frequent refer-
ences to the necessity and merits of joint working, shared understandings
and co-operative strategies’ (Flynn et al. 1996: 115). However, GP fund
holders:
displayed a greater commitment to relational contracting in their purchasing of
CHS, placing a high value on local networks, trust and loyalty with regard to
members of the primary health care team (Flynn et al. 1996: 91).
Flynn et al. (1996: 61) found that:

the process of linking needs, resources, services and outcomes was one that
required much closer linkage between purchasers and providers. The most
successful examples were . . . where the purchasers had stepped outside the con-
tractual relationship between senior managers in order to cultivate collaborative
network relationships between intermediate-level managers in the purchaser
organisation and service heads in the provider trust.
Moreover, markets cost because ‘the network structure which characterises

primary care may be fractured by the aggregate effect of individual GP fund
holder purchasing decisions’ (Flynn et al. 1996: 79). So, although
the contracting system and the experience of bargaining and negotiation seems
to permit or even encourage adversarial . . . modes of relating to each other,
underlying this both parties agreed on the priority of maintaining local relation-
ships with CHS (Flynn et al. 1996: 86).
So, ‘fund holders are prepared to shop around, but only on the margins’
(Flynn et al. 1996: 92).
The CHS displayed an uneasy mix of adversarial, collaborative, and net-
working strategies, which created a dilemma. The government promoted both
collaboration and competition. For example:
We need constructive co-operation between different parts of the NHS as well as
the beneficial impact of competition. Improving health care is not a question of
choosing one or the other. We have to find the appropriate balance between the
two (cited in Flynn et al. 1996: 22).
The contradiction mattered because central government’s instructions were
potentially corrosive, creating an atmosphere of mistrust and undermining
professional values. Thus:
the government’s instructions to promote contestability, accountability and
value for money ultimately came to dominate their commissioning and con-
tracting approach. The effect was often to corrode rather than nurture common
values and commitments, to create an atmosphere of mistrust and to exacerbate
inherent problems of uncertainty in the contract process
(Flynn et al. 1996: 115).
So, there are ‘potentially corrosive effects of market relations and commodity
exchange values on professional networks which depend on co-operation
reciprocity and interdependence’ (Flynn et al. 1996: 136–7). In short, contracts
undermined trust, loyalty, informality, and cooperation.
This danger was acute when key actors in primary health care, such as GP
fund holders:
avoided bureaucratic or punitive styles in favour of much looser or more informal
approaches . . . they appeared to endorse a network approach in which loyalty to
colleagues in the primary health care team—and appreciation of their local
knowledge was essential (Flynn et al. 1996: 136).
Four aspects of Flynn et al.’s study are important. First, the characteristics of
CHS posed serious problems for contracting. Second, health authorities
adopted varying strategies: bureaucratic and negotiated. Third, much behav-
iour was not competitive but cooperative, rooted in trust and networking
(Flynn et al. 1996: 136, 139–41). Finally, competition corroded cooper-
ation (for similar conclusions for the social care services, see Wistow et al.
1996: 173–4).
Self-organizing
At its simplest, self-organizing means a network is autonomous and self-

governing. Integrated networks resist government steering, develop their
own policies, and mould their environments. The literature on policy-making
and implementation provides innumerable examples of such resistance and
bottom-up policy-making. The only point at issue is not whether there is a
control deficit, but its extent. The trend to fragmented service delivery pro-
duces greater control deficits and more scope for bottom-up policy-making
(see, for example, Marsh and Rhodes 1992b).
The dilemma is ‘who is accountable to whom for what’, and it is no easier to
resolve than my earlier question of ‘who provides services’. The Committee of
Public Accounts (CPA) (1994) reported a graphic illustration of the failure of
accountability. The (former) Wessex Regional Health Authority implemented
a regional information system but the NHS management executive did not
know about the significant overspend until six months after the end of the
financial year. The CPA estimated the Wessex RHA wasted at least £20 million
implementing a scheme it abandoned in 1990. It argued the NHS management
executive and the regional health authority were kept at arm’s length for three
and a half years and it represented ‘a serious failure . . . to secure accountability’
and ‘to act with sufficient urgency’. The waste was compounded by conflicts
of interest:
the authority appointed a Director of [Wessex Integrated Systems Ltd], with
whom the Regional Health Authority had a contract to supply computer services,
to act as Regional Information Systems Manager. The Authority also allowed a
secondee from IBM to advise them on the purchase, without competition, of an
IBM computer for £3.3 million, at a time when it could have been purchased for
£0.5 million to £1 million less.
Such instances may be isolated occurrences but the Committee on Standards
in Public Life (Nolan) made 50 recommendations for improving the account-
ability of local public spending bodies, a category that excludes both the NHS
and non-departmental public bodies (Cm 3270, 1996). One fact is clear.
Contrary to the tenets of the Westminster model, the accountability of services
to either elected local representatives or ministers is seriously diminished by
networks.
Steering
The combination of institutional differentiation, interdependence, game-

playing, and self-organization limits the capacity of the centre to steer. Policy
outcomes are not the product of actions by central government. The centre
may pass a law but subsequently it interacts with local government, health
authorities, special purpose bodies, the voluntary sector, the private sector, and,
in turn, they interact with one another. There is order in the policy area that is
not imposed from on high but emerges from the negotiations of the several
interdependent parties. No one actor has the information and expertise to solve
complex problems, or the influence or policy instruments to make unilateral
decisions (see Kooiman 1993a: 4; Klijn et al. 1995; Kickert et al. 1997).
So, all the actors in a particular policy area need one another. Each can
contribute relevant knowledge or other resources. No one has all the relevant
knowledge or resources to make the policy work. Central government is not
just another actor but equally it is no longer supreme. British government is
increasingly differentiated—it has become the polycentric state.
In the 1990s, British government adopted a strategy of ‘more control over
less’. It privatized the utilities. It contracted out services to the private sector. It
introduced quasi-markets through purchaser–provider splits when services
could not be privatized. It bypassed local authorities for special purpose
bodies. It removed operational management from central departments and
vested it in separate agencies (see Rhodes 1997c). Central departments rarely
delivered services themselves; they were non-executant. Government policy
fragmented service delivery systems. It compensated for its loss of hands-on
controls by reinforcing its control over resources. Decentralizing service
delivery has gone with centralizing financial control. Such hands-off controls
may not provide enough leverage for the centre to steer the networks. As
networks multiply, so do doubts about the centre’s capacity to steer. This shift
from line bureaucracies to fragmented service delivery systems can be sum-
marized as the shift from government to governance.
However, this analysis of governance and its networks differs from my
earlier analysis of networks. The functional policy networks were based on
central departments or parts of them. Government policy in the 1980s and
1990s sought to weaken and control policy networks and to bypass local
government by creating special purpose bodies. By fragmenting Britain’s
institutional structure, the government created service delivery networks
with two distinctive features. First, the membership of networks became
broader, incorporating both the private and voluntary sectors. Second, the
government swapped direct for indirect controls. Central departments are no
longer either necessarily or invariably the fulcrum, or focal organization, of a
network. Power relations may remain asymmetric. The government can set
the limits to network actions. It still funds the services. But it has also increased
its dependence on multifarious networks.
The Department of National Heritage (DNH) is an exemplar of steering
and the new governance.3 Created in 1992, it is responsible for heritage,
sport, tourism, broadcasting, media, arts, the national lottery, libraries, and
museums and galleries. It is a non-executant central department with a total
budget of only £1 billion. Its operating style is the ‘strategic, policy-oriented
and contracted-out form’ (Cm 2511 1994: 12) and a favourite phrase is ‘setting
the framework’. Its distinctive characteristic is that:
it operates largely through a network of public bodies and agencies which receive
much of their support from public funds but which are managed and operated at
arm’s length from government (Cm 2211 1993: v).
Its role was, therefore, to provide the policy framework and to act as a catalyst:
‘to mobilise resources and skills from across the public and private sectors’; to
use ‘our power of regulation effectively’; and to provide ‘other assistance and
advice’ and to ‘stimulate as well as influence private sector activities’ (Cm 2811
1995: v). In short, it worked through established networks and developed the
appropriate policy levers. To the conventional lever of patronage, Taylor
(1997) adds four levers: ministerial activism; systematic review and scrutiny;
policy guidance; and financial resources.
Ministerial activism refers to redefining the arm’s length principle. The
DNH wanted explicit understandings about the use of government funding,
and if it could not get satisfactory funding agreements it was prepared to direct
that ‘the arm should be made shorter’ (Taylor 1997). Systematic review
requires sponsored bodies to produce a corporate and business plan (includ-
ing a financial agreement) supplemented by an annual review of performance.
Policy guidance refers to public statements of intent issued in the form of press
releases and White Papers, not legislation. Finally, the DNH’s sponsored
bodies are dependent on it for finance, providing it with a direct lever when
distributing exchequer grants and indirect leverage when setting the policy
framework for distributing national lottery money. In sum:
the DNH took over some long-established policy networks, some of recent origin
and some where a network hardly existed. Where a network already existed the
DNH officials saw their role as to accommodate that network to its ethos, where
the network was new or embryonic: their task was to substantially create (or at
least systematise) both the policy and the network by exploiting the uncertainty
(caused by the DNH’s creation) of the boundaries of its policy networks
(Taylor 1997).
3
This example is from Taylor (1997). I thank him for permission to use his work.
Taylor (1997) argues the effectiveness of the DNH’s policy levers undermines
the governance thesis about the loss of central steering capacity but provides
no evidence of the impact of his policy levers. The dilemma is that the
DNH influences and persuades. It does not direct, but relies on indirect
management. So, it pulls its policy levers, uncertain they will have the desired
effects. Intent and impact are not synonyms and the DNH confronts the
problem that it is pulling ‘rubber levers’.
The DNH understands network logic. It rejects the command central
operating code for indirect management. It keeps fuzzy boundaries between
its networks. Network closure aids the private government of a policy
area and lies at the heart of self-organizing networks. Indeterminate bound-
aries and openness of communication are essential if the alleged benefits
of networks—such as expertise, jurisdictional competition, deconcentration,
and responsibility—are to be realized (Rhodes 1988: 404). But the power of
the DNH is the power to persuade, negotiate, and guide. It epitomizes
indirect management and heralds the new era of intergovernmental
management.4
CHA NGING LANGUAGES: FROM MARKETIZATION

TO DIPLOMACY
Governance as self-organizing interorganizational networks has characteris-

tics distinct from markets and hierarchies and poses new dilemmas. British
government can choose between ‘governing structures’; that is, between mar-
kets, hierarchies, and networks. Bureaucracy remains the prime example of
hierarchy or coordination by administrative order and, for all the recent
changes, it is still a major way of delivering services in British government;
for example, the Benefits Agency remains a large bureaucracy. Privatization,
marketing testing, and the purchaser–provider split are examples of govern-
ment using market or quasi-market ways of delivering services. Price compe-
tition is the key to efficient and better quality services. Competition and
markets are a fixed part of the landscape of British government. It is less
widely recognized that British government now works through networks
characterized by trust and interdependence; for example, to provide health
and social care. British government is searching for a new ‘operating code’.
4
For further discussion of intergovernmental management and citations, see Rhodes 1997a:
chs 3 and 9; and for the work of the ‘governance club’ at Erasmus University, see Kickert 1993;
Kooiman 1993c; Klijn et al. 1995; Kickert et al. 1997. See also Chapter 5, this volume.
This search involves choosing not only between governing structures but also
the mix of structures and strategies for managing them. Networks are a
widespread form of social coordination, and managing interorganizational
links is just as important for private sector management. However, if networks
are a distinctive form of economic coordination, it is important to recognize
their distinctive characteristics. Three characteristics recur: interdependence;
resource exchange; and trust (see Chapter 5, this volume).5
Because of their distinctive characteristics, networks need different policy
instruments and management styles. For example, Alexander (1995: 276) iden-
tifies six strategies for managing interorganizational coordination. The strategies
are: cultural-persuasive (for example, public relations); communicative (for
example, information exchange); functional (for example, coalition formation);
cooperative (for example, resource exchange); control (for example, monitoring
and enforcement) and structural (for example, reorganization). He concludes
‘there is no universal algorithm’ for managing networks (see also Agranoff 1990;
Kickert et al. 1997; and Chapter 4, this volume). However, any strategy must
reflect the characteristics of networks. The unilateral imposition of a solution
will probably not work and it will incur high costs. So, managing interorganiza-
tional networks is gamelike, employs an indirect style of management, needs
strategies rooted in trust, and uses the ‘art of rhetoric’ (Thompson 1993: 57) or
argument, debate, and persuasion.
The language used to describe the changes is important. Managerialism and
neo-liberalism provided the fashionable, legitimating operative concepts in the
1990s and they continue to do so today (see Bevir and Rhodes 2016), but they
do not describe accurately what is happening. Transaction cost economics
resorts to such concepts as ‘relational contracting’ to describe long-term
relations governed by norms evolved by the parties during the contract.
Such uses stretch the meaning of contract too far. I prefer the language of
networks because it provides a better description of the social context of
exchanges, identifies the mix of governing structures, and suggests different
managerial strategies for coping with that mix. At the end of this embedded
process of negotiating, there is a contract but it is not self-liquidating; it is a
formal juncture in a continuing, informal process.
The five examples of AIDS, local government finance, community health
services, Wessex RHA, and the DNH cover central government, special purpose
bodies, local government, and the NHS. They illustrate some facet of network
5
For a more detailed discussion of the distinctive characteristics of networks in economic
transactions as well as in delivering public services, see Flynn et al. 1996: 139–41; Kramer and
Tyler 1996: chs 4 and 16; Larson 1992: 98; Powell 1991: 268–74; Rhodes 1997a: ch. 3; Thompson
et al. 1991: Introduction and chs 2–3; and Thompson 1993: 54–60. See also Chapter 3, this
volume.
behaviour. Except for the DNH, government policy was not to create networks
and its preferred management strategy was managerialism with its emphasis
on competition, objective setting, targets, and performance indicators. Networks
and network behaviour were an unintended consequence of marketization.
Institutional differentiation—whether by contracting out, public–private part-
nerships, or bypassing local government for special purpose bodies—creates
imperatives for interdependent actors to work together and multiply net-
works. Marketization corrodes networks and prompts defensive behaviour.
Resorting to hierarchies, with their command operating code and instructions,
fuels non compliance and recalcitrant, conflictual behaviour. No governing
structure works for all services in all conditions. The issue, therefore, is not
the superiority of markets and hierarchy over networks, but managing
networks in the conditions under which they work best. For example, Flynn
et al. (1996: 147) conclude that networks thrive when:
professional discretion and flexibility, together with collaborative teamwork, are
deemed to be core values and organisational prerequisites for most community
practitioners. Their clan-like structure, and their promotion of the virtues of
co-operation and interdependence, necessitates both a management approach
and a purchasing strategy based on high trust, and soft or relational contracting.
In a similar vein, Powell (1991: 272) argues that networks fit where: actors
need reliable information; quality cannot be specified; and commodities are
difficult to price.
Managing networks, therefore, is central to any new operating code for
steering the differentiated polity. My objective is to add the language of
networks to the language of contracts and competition. So, I focus on the
distinctive characteristics of networks and the art of rhetoric. The concept of
trust and diplomacy are central to this discussion as a way of managing
interdependence by negotiation.
Trust
Williamson (1993: 485) argues trust is ‘a diffuse and disappointing concept’. It

is part of a family of concepts including fairness, honesty, and reciprocity
(Larson 1992: 96); and loyalty, consistency, and reputation (Thompson 1993:
59). The discussion will be clearer if I distinguish between economic and social
conceptions of trust.
Williamson (1993) exemplifies the economic conception of trust and he
argues for a distinction between calculative, personal, and institutional trust.
Calculative trust is synonymous with risk and Williamson argues that ‘the
practice of using “trust” and “risk” interchangeably should be . . . discontinued’
(Williamson 1993: 486; see also March and Olsen 1989: 27). Personal trust is
nearly ‘non-calculative’ and characterized by ‘the absence of monitoring . . .

favourable or forgiving predilections . . . and discreteness’ (Williamson 1993:
484). Institutional trust refers to ‘the social and organisational context within
which contracts are embedded’ (Williamson 1993: 486). Williamson (1993:
484) suggests contract relations are characterized by calculative trust and that
personal trust is largely peripheral to commercial transactions.
Networks are a distinctive coordinating mechanism and, therefore, a sep-
arate governing structure from markets and hierarchies. The social conception
of trust is ‘the most important attribute of network operations’. It is the central
coordinating mechanism of networks in the same way that commands and
price competition are the key mechanisms for bureaucracies and markets
respectively (Frances et al. 1991: 15). Shared values and norms are the glue
that holds the complex set of relationships together; trust is essential
for cooperative behaviour and, therefore, the existence of the network. So,
economic transactions are both socially embedded (Granovetter 1985) and
characterized by non-calculative trust. For example, Tyler and Degoey (1996:
347) conclude that non-calculative trust is present even in commercial relations
and a central influence on organizational behaviour. They suggest that trust
should be seen as ‘benevolent intentions’ rather than ‘calculated risk’ (Tyler
and Degoey 1996: 345). Similarly, network studies show that non-calculative
trust is important for transactions, including economic transactions (see Flynn
et al. 1996).
However, it would be a mistake to over-emphasize the dichotomy between
economic and social conceptions of trust. As Powell (1996: 63) points out,
trust is ‘neither chosen nor embedded but is instead learned and reinforced,
hence a product of ongoing interaction and discussion’. Similarly, Larson
(1992: 97) argues that:
actors are over socialised when portrayed as governed exclusively by values
and norms and under socialised when portrayed as isolated, rational economic
units . . . A middle ground must be found.
Maintaining trust, therefore, is a reciprocal and endless task (Barber 1983).
This conclusion is borne out by Flynn et al. (1996: 136) who argue that trust
became important because of the difficulties in specifying contracts and
participants’ experience of assertive purchasers whose style ‘engenders or
exacerbates suspicious attitudes and feelings of mutual distrust’. So:
where there is no competitive market, and/or there are doubts about compliance
with quality requirements, purchasers can either rely on trust or they can try to
stipulate specifications in great detail (Flynn et al. 1996: 115).
The response in Flynn et al.’s (1996: 136) case study authorities was clear:
it was evident that for the different groups within and between agencies (includ-
ing GP fund holders), their commissioning and contracting behaviour was
significantly influenced by their willingness to trust the other party in a whole
range of circumstances.
Their conclusions echo Fox’s (1974: 362) analysis of trust in industrial relations.
He argues that, in high-trust relationships, participants:
share certain ends or value; bear towards each other a diffuse sense of long-term
obligations; offer each other spontaneous support without narrowly calculating
the cost or expecting any equivalent short-term reciprocation; communicate
freely and honestly; are ready to repose their fortunes in each other’s hands;
and give each other the benefit of any doubt that may arise with respect to
goodwill or motivation.
In contrast, in low-trust relationships, participants:
have divergent ends or values; entertain specific expectations which have to be
reciprocated through a precisely balanced exchange in the short term; calculate
carefully the costs and anticipated benefits of any concession; restrict and screen
communications in their own separate interests; seek to minimise dependence on
each other’s discretion; are quick to suspect ill-will; invoke sanctions against
others, and default on obligations.
As a working axiom, therefore, contracts are low trust and networks are high trust.
Diplomacy
Diplomacy is ‘the art of getting what we want, applied to foreign politics.6

Outside this special field we are all diplomats: in our business, in our family life
and in our love affairs’ (Vare 1938: 23). It is a way of managing interdepend-
ence by negotiation. Nicholson (1950) also identified seven diplomatic virtues:
truthfulness; precision; calm; good temper; patience; modesty; and loyalty
(to the government one serves). There is a charming quality to Nicholson’s
account. The budding diplomat is advised that ‘above everything, do not allow
yourself to become excited about your work’ (Nicholson 1950: 116); ‘patience
and perseverance are also essential to any successful negotiator’ (Nicholson
1950: 117); and ‘personal vanity breeds self-satisfaction which leads to a loss of
adaptability and a decline in imagination’ (Nicholson 1950: 120). Nicholson
(1950: 126) then added:
‘But’, the reader may object, ‘you have forgotten intelligence, knowledge, discern-
ment, prudence, hospitality, charm, industry, courage and even tact’. I have not
forgotten them. I have taken them for granted.
6
The earliest discussion of diplomatic skills in public management I have found is Keeling
(1972: Chapter 5) who identifies three species of systems in public services: administration,
management, and diplomatic systems.
The term may not be fashionable, but Nicholson identifies a distinct style.7
I addressed the annual conference of the Queensland Division of the Institute
of Public Administration Australia. I contrasted the style of the ‘head kicker’—
Australian for macho-manager—with that of the diplomat. As I mingled after
the address, three female public servants working for the Queensland govern-
ment approached me, congratulated me on my talk, ‘but’, they commented,
‘they won’t listen to you. That diplomacy (pause) it’s girlie talk’. My instant
response was to laugh. On reflection, I realized that language about sitting
where the other person sits and helping other people to realize their objectives
was seen as ‘soft’. But such objections are specious and ignore much accumu-
lated experience. Francois de Callières ((1963) [1716] 103) commented:
Now, if I were in the place of this Prince, wielding his power, subject to his
passions and prejudices, what effect would my mission and my arguments have
on me? The more often he puts himself in the position of others, the more subtle
and effective will his arguments be.
Watson (1982: 225), as almost the last words in his book on diplomacy,
comments ‘experience teaches that compromise is required’.
Indeed, the literature on diplomatic negotiations contains uncanny parallels
with intergovernmental relations in a nation state. For example, Craig and
George (1983: 157) argue that negotiation is necessary when there are ‘com-
mon interests and issues of conflict’: ‘without common interests there is
nothing to negotiate for; without conflict there is nothing to negotiate about’.
Without common interests there is no interdependence in the differentiated
polity and functional conflicts are endemic. One way of resolving conflict in
international relations lies in negotiation; hence Nicholson’s definition.
Such negotiation involves several interrelated tasks and purposes. For
example, Watson (1982: 123–5) identifies the following tasks:
finding out or guessing intelligently what one power needs to know about another . . .
sifting and collating the information received . . . and of producing a coherent picture
of the issues and developments abroad on which decisions are needed . . . determining
the options available to a government and submitting them for decision . . . commu-
nicating and explaining a government’s decision to another government.
The emphasis lies not in imposing one’s objectives on another but on finding
out about the other. The diplomat must persuade ‘another government
to accept and perhaps actually help to promote the policies which it is
the ambassador’s function to advocate’ (Watson 1982: 125) and the main
technique is ‘the maintenance by continual persuasion of order in the midst of
change’ (Watson 1982: 223).
7
Watson (1982: 12) essays a similar list: ‘Nor have I tried to enumerate the personal qualities
of honesty, perceptiveness, tact, a sense of timing, a flair for entertaining, a flair for poker and so
on which a long line of distinguished statesmen and ambassadors have compiled as desirable for
diplomats.’
To return to the present, diplomacy may be an old-fashioned word but the
arts of negotiation and persuasion are not specific to it. Such skills lie at the
heart of steering interorganizational networks. The idea of diplomatic man-
agerial skills is not new, but it has been overlooked.
CONCLUSIONS
Markets, hierarchies, and networks are not found in their pure form. All are
treated here as ideal types to find out, for example, the extent of price
competition or the degree of trust in a given set of relationships. Also, no
one service will employ only one resource allocation mechanism. CHS involve
hierarchy (instructions from the government), markets (contracting), and
networks (GP fund holders and primary health care). It is the mix of govern-
ing structures that distinguishes services one from the other. These governing
structures may mix like oil and water.
Contracting can erode trust in local networks. Trust and negotiation can
reduce the scope for competitive pricing. Also, the policy levers that can be
effectively used vary between governing structures and it is important to
distinguish between them. There is little to be gained by using the label ‘contract’
to cover both trust-based agreements and price-based contracts. Both will
contain elements of price competition. Both will need some trust between the
parties. But the relative priority accorded to price and trust varies and our
language, the map, should be able to distinguish between the two. It is the mix
that matters and that mix is no longer markets and hierarchies. Marketization
has contributed to the spread of networks by increasing the extent of functional
differentiation in public service delivery systems. Its operative concepts have
restricted the toolkit available to government for managing networks. Networks
are pervasive. Government is picking up the skills of indirect management, but
slowly. This chapter aims to hasten that process by providing a language for
exploring and managing the mix of governing structures in the differentiated
polity. The new public management, whether in the guise of managerialism or
institutional economics, is no longer the challenge confronting government. The
challenge is diplomacy in governance.
A F T E R W O RD
As with earlier chapters, this Afterword sets out the context in which the
chapter was written and updates the empirical story when relevant. I wrote
this chapter at the same time as I wrote Understanding Governance (1997) and
I returned to the topic of ‘the mix’, on several occasions. This section summar-
izes these discussions (see Fleming and Rhodes 2005; Rhodes 2003b, 2006b).
I reply to my critics in Chapter 12.
Continuity and Change
Obviously, as with Chapter 8, my examples are dated but, as I will argue in this
Afterword, the general argument remains valid. The first point to note is that
subcentral governments in the UK experienced mixed fortunes. It is important
to separate English local government from the devolved assemblies in
Scotland, Wales, and Northern Ireland. There was genuine decentralization
to the latter, creating a quasi-federalism or ‘dis-United Kingdom’ (Rhodes
et al. 2003; and Chapter 12, this volume) and new dimensions to intergovern-
mental management and bargaining (see Holden 2010; Mitchell 2009).
For English local authorities, the 1990s and the 2000s were a period of
‘hypercentralization’ (Wilson and Game 2011: 166) characterized by the ‘vice-
like grip’ of the Treasury and the service sponsoring departments (Travers
2007: 78). Stoker (2004: 216–22) accepts there has been a significant degree of
centralization but prefers to describe it as ‘steering centralism’ in which the
centre steered rather than commanded, especially when councils had ‘earned
autonomy’ through improved service delivery.
The fortunes of local government did not improve under the Coalition
government of 2010–15. Austerity meant offloading the cuts to the periphery:
Between 2010 and 2015 council expenditure and employment had fallen faster
than in any period since 1945. Average real spending by councils in England was
down by 15 to 20 per cent in real terms (Travers 2015: 240).
Not only was central funding slashed but the council tax was capped.
Despite the centre’s rigorous control over local expenditure, it was not all
change. There was some noteworthy continuity. First, as ever, there were costs
to centralization. As Stoker (2004: 220–2) comments, it is difficult in central-
ized systems for the government both to trust the information it receives and
to cope with the sheer volume of such information:
Managers at the periphery use the space created through information overload to
pursue their own schemes, to get on and do their own thing. The space is there
both when the centre is not looking and when it is looking but cannot see through
the fog of information that surrounds it (Stoker 2004: 221).8
Using my preferred terminology, hierarchies, with their command operating
codes, are clumsy and cannot fine-tune policies, so non compliance, recalci-
trance, conflict, and even failure become endemic.
8
See Lowndes and Gardner 2016, for an example of how one local authority used that space.
Second, fragmentation was a persistent feature of reforms as the Conserva-

tives pursued their neo-liberal agenda of creating the minimalist (or 36-per-
cent) state through contracting out, public–private partnerships and bypassing
local government for special purpose bodies. As noted in Chapter 8, this
volume, the companies in the ‘franchise state’ game the system producing
administratively inefficient outcomes that the government has no capacity to
regulate (Bowman et al. 2015: 6; Raco 2016). When fragmentation and aus-
terity collide, the results are perverse. For example, Bradbury (2016) concludes
that ‘social care professionals and clients considered the market governance
model [of social care] to be inadequate, unworkable and unfair’ and the
National Audit Office (2014) found that spending by local authorities fell by
8 per cent in real terms while demand for care was rising. So, the National
Health Service struggles with ever-rising demand yet cannot discharge pa-
tients into social care because government cuts to local government spending
mean that such provision is not available.
And the mix is about to change again. Since May 2015, the Conservative
government offered ‘radical devolution to the great cities of England’. The
Chancellor of the Exchequer, George Osborne (2015), accepted that ‘the old
model of trying to run everything in our country from the centre of London
is broken’. He offered to transfer additional powers in housing, transport,
police, and social care to cities in a metropolitan area that agreed to ‘a directly
elected metro-wide mayor’.9 The package led one seasoned commentator to
observe:
By 2019/2020, with the Government’s proposed phasing out of Revenue Sup-
port Grant and 100% local retention of business rates, spending on day-to-day
services for large numbers of councils would be almost entirely funded by a
combination of business rates and council tax. Councils have glimpsed
the future and, almost unbelievably, that future is grant-free. They will be
locally funded to a degree that the most localist of reformers over the past
few decades could hardly have conceived—even if the means and consequences
aren’t necessarily those that they might have advocated (David Wilson, corres-
pondence, 5 February 2016).
So, the general argument about unintended consequences remains valid (and
for many more examples, see King and Crewe 2013). The costs of centraliza-
tion (hierarchies), contracting (markets), and fragmentation (networks) are an
old story. As I have argued throughout this chapter, no governing structure
works for all services in all conditions. It is an ever-shifting, oscillating balance
between the governing structures with governments struggling to manage any
one structure let alone their mix.
9
Subsequently his ideas were set out in the Queen’s Speech of 27 May 2015 in the Cities and
Local Government Devolution Bill, which became law in January 2016.
The Sour Laws and the Unholy Trinity
The heart of my analysis is captured by two phrases; ‘the sour laws of

unintended consequences’, especially those consequences that arise from ‘the
unholy trinity of markets, hierarchies, and networks’.
I use the phrase ‘the sour laws of unintended consequences’ in two ways.
First, I use it as a humorous adage rather like Murphy’s Law to warn that the
intentions of policy-makers are just that: intentions not outcomes. Second,
I use it more precisely to refer to policy outcomes that are unanticipated or
unintended. An intended outcome can realize some or all of the intentions of
the policy-makers but it can also have unforeseen side effects, or unanticipated
consequences. An intended outcome can fail to realize the intentions of the
policy-makers and have the opposite effect to those intended, or unintended
consequences. I accept that both unanticipated and unintended consequences
can be beneficial to some people, but that is not my concern here (and, for
those so inclined, there is a definitional survey in Perri 6 2010).
The causes of such consequences have been much discussed. As well as
sheer stupidity and political venality, the causes include incomplete informa-
tion, limited cognitive and technical skills for analysing data, the complexity of
problems, theories that cannot predict, and hypotheses that we either cannot
or have not tested. From my standpoint, the key reason is that policy-making
practices produce complex specificity in context.
Conventionally, we focus on the intentions of central policy-makers but
they are not the sole stakeholders in any policy. There are multiple stake-
holders; for example, street-level bureaucrats, social movements, everyday
makers (see, for example, Bang and Sørensen 1999; Lipsky 1980; Maynard-
Moody and Musheno 2003; and Vinzant and Crothers 1998). So, the inten-
tions of central policy-makers are not the only yardstick against which to
assess a policy. For example, resistance to central policy creates unintended
consequences for those policy-makers, but they are the intended outcomes for
those resisting. Policy is ever different and changing because people’s actions
and practices intersect and interact spinning off to create and recreate webs of
inordinate complexity that are the product of no one person’s intentions but
become part of the beliefs and practices of all. This spinning-off or generative
effect is ever present and mostly unanticipated. My use of the expression
‘unintended consequences’ focuses on the performance of everyday actions
and practices by individual actors, not institutions or central policy-makers,
and their consequences.10
10
Unintended consequences are not just the result of poor design or recalcitrant implemen-
tation, they are inevitable because of ‘the circularity of social knowledge’: ‘New knowledge
(concepts, theories, and findings) does not render the social world more transparent, but alters
its nature, spinning it off in novel directions’ (Giddens 1990: 153). There is a ‘juggernaut’, or to
Turning to the unholy trinity and the mix of markets, hierarchies, and
networks, my analysis focuses on the beliefs and practices of the actors in a
policy arena and the unintended consequences of their actions. Thus, Fleming
and Rhodes (2005) conducted 27 interviews with police officers about man-
agement reform. We explored how the three governing structures interacted.
We reported the views of the police on bureaucracy, contracts, and networks
in their own words. Here, I provide a brief summary of their views.
On Bureaucracy
The continuing importance of bureaucracy—of authority, hierarchy, and rules
pervaded the interviews.
We have guidelines for the procedure on how to deal with a mentally disturbed
person; animals on the loose; pulling vehicles over; arresting someone—just
about everything a police officer does is prescribed by a practical guideline . . .
Sometimes we have to create new ones to deal with new legislative requirements
or new situations. It’s about procedure and policy—how we do things
(Interview 10).
The traditional ‘command and control’ style persists:
action and results are highly valued by police officers . . . they are competitive
about arrests . . . they view success as someone behind bars . . . there is a desire to
right wrongs . . . that’s what motivates them . . . they are not motivated by a school
principle who says they have conducted their community policing duties well
(Interview 1).
There was much agreement among senior officers that the organizational
structure was still based on ‘a rationalised, centralised model—where areas
such as traffic and crime prevention are considered as specialised units’.
Two important inferences can be drawn from this material. First, beliefs in
the efficacy of rules, uniforms, and authority persist after decades of reform
and appear as essential organizing principles in these accounts. Second, such
beliefs persist because they accord with the experience of the officers. For
them, bureaucracy works because it imposes order. So, this data can be
interpreted as evidence of the effectiveness of bureaucracy.
On Contracts
Managerialism in both its guises of performance measurement and
contracting out littered the conversation of interviewees. There was much
use my preferred term, ‘generative’, quality to the impact of knowledge on policy (see Rhodes
2017, Volume II, Chapter 10).
‘management-speak’ and the language of competition abounded and infil-

trated their worldview. Many believed that they were ‘driven by contracts’
(Interview 23) and were often fearful that those contracts would be awarded
elsewhere (Interview 25).
As ever, views differed. For some, the shift to the new management style had
yet to take place:
The thing is collectively we haven’t realised yet we are a business. We have to
make decisions. [Senior management] don’t make decisions and when they are
forced by circumstance to do so they shoot from the hip. . . . (Interview 12).
For others, change was gradual but the force was getting there.
On Networks
Community policing is about partnerships, consultation, and building trust.
We found formal consultative links, issue specific links, and informal activity.
Formal consultative links covered, for example, domestic violence, working
with and in schools, and community consultation. Issue specific links referred
in the main to such agreements as memoranda of understanding or MOUs.
The force had some MOUs covering partnerships with local taxi firms, mental
health, and prisons and corrective services. Even when there was no formal
consultative body or MOU, the police still had informal contacts with various
sections of the community, and informal understandings with government
agencies.
There were low levels of awareness of the extent of police involvement in the
community, but there was commitment from those who see community
networking as the future:
We need to work towards an inter-agency approach—it will be difficult but if you
are determined to make it work there is no physical reason why it shouldn’t work
if you persevere. We need a cooperative focus (Interview 24).
However, there was a clear stereotype that the police focused on crime and saw
networking and crime prevention as soft:
I think your biggest problem will be the culture. It’s still isolated, a ‘boy’s own’
club—community policing means beat policing to them and they don’t do that
well. They don’t like all this touchy feely stuff (Interview 16).
Police don’t want to get into the crime prevention stuff though. No one
wanted to do these jobs—they wanted to leave it to ‘the warm and fuzzies’.
Police wanted ‘to wear their underpants on the outside and save the world—
they wanted to make the person pay’ (Interview 18). There were ‘lots of
platitudes but little action. The reactive stuff always takes precedence over
the proactive stuff ’ (Interview 3).
The future will not lie with either markets, or hierarchies, or networks but
with all three. The trick will not be to manage contracts or steer networks
but to mix the three systems effectively when they conflict with and undermine
one another. The cooperative behaviour of a network can collapse under the
impact of competition or of changed priorities. Such changes are a continuing
problem in crisis-driven organizations like the police. Some officers appreciated
this dilemma and recognized the need to fit their managerial strategies to the
context.
Command and control is situational. In my team, I don’t have subordinates.
I have team members. Years ago a constable wouldn’t speak to a superintendent—
this is not the case now. I invite their ideas and input and encourage them to talk to
me. If they are happy I have a productive working team. However, as I said, it’s
situational. Fighting fires is a good example. As a commander, when I want
something done, it isn’t up for negotiation. We have to rely on command and
control (Interview 7).
The central story of police reform will be the efforts to match management
style to the situation. Police officers confront several unintended consequences
arising from the coexistence of the three governing structures. There is a
tension between cooperative behaviour working with the community and
internal competition for resource allocations linked to performance measure-
ment. There is the paradox between public complaints about police ineffi-
ciency and public demand that police do all manner of non-police work and
attend non-urgent calls. The call for openness in dealings with the community
flounders on cost reduction strategies because it is not politically feasible to
publicize, for example, that domestic burglary will not be investigated. Per-
formance indicators are inescapable but they do not span organizational
boundaries; they do not fit community policing. Finally, police leadership on
all these matters is compromised by political demands for responsiveness.
Police officers seek to resolve these dilemmas by balancing the unholy trinity
of the ever-changing mix of markets, hierarchies, and networks. Police reform
is truly an example of ‘it’s the mix that matters’.
Conclusion
12
What Is New about ‘Network Governance’

and Why Does It Matter?
Pigeon-holed as an ‘idealist’, Collingwood (1978 [1939]: 56) foreswore the ‘occa-

sional remonstrance’; ‘I became used to it: otherwise I might have been too much
annoyed to keep that rule against answering critics which everyone must keep who
has work of his own to do.’ So far in this volume, I have listened to Collingwood
and concentrated on what I want to say. Now I become deaf to his argument.1
The phrase ‘the differentiated polity’ is my preferred summary term for my
account of British government (see Chapter 11), although it is also described
as ‘the Anglo-Governance School’ (Marinetto 2003) and ‘the Governance
narrative’ (Newman 2005: 8). The Westminster model of British government
is best understood by exploring such core ideas as a unitary state, parlia-
mentary sovereignty, strong Cabinet government, ministerial accountability,
majority party control of the executive, and institutionalized opposition (see
Rhodes and Weller 2005). The differentiated polity narrative challenges the
Westminster model’s account of British government. Its core ideas are policy
networks, the core executive, hollowing out, and governance. It argues there
has been a shift from government by a unitary state to governance through
and by networks. Differentiation became more extensive in the 1980s and
1990s, which saw significant changes in the functional and territorial special-
ization of British government. The arguments that networks have multiplied
as an unintended consequence of marketization; that the degree of inter-
national interdependence is greater; and that, as a result, the core executive’s
1
This chapter is substantially new but it incorporates material from earlier replies to critics,
mainly R. A. W. Rhodes (2007) ‘Understanding Governance: Ten Years On’, Organization Studies,
28 (8): 1243–64. © 2005 Sage Publications. Reprinted with permission; R. A. W. Rhodes (2003) ‘What
Is Governance and Why Does It Matter?’ In J. E. S. Hayward and Anand Menon (eds), Governing
Europe. Oxford: Oxford University Press, pp. 61–73. Reprinted by permission of Oxford University
Press; and R. A. W. Rhodes (2011) ‘The Stateless State’. In Mark Bevir (ed.), The Sage Handbook of
Governance. London: Sage, pp. 203–17. © Sage Publications. Reprinted with permission.
capacity to steer is reduced or hollowed out serve to reinforce the interpretation

that centralization and control are incomplete and Britain is best viewed as a
differentiated polity—a disUnited Kingdom (Rhodes, Carmichael et al. 2003).
Commenting on my research into British central government, Peter Riddell,
then political editor of The Times, said:
The language in which political scientists operate is divorced from that of

practitioners and commentators. Every time I see the word ‘governance’ I have
to think again what it means and how it is not the same as government. Terms
such as ‘core executive’, ‘differentiated polity’ and ‘hollowed out executive’ have
become almost a private patois of political science, excluding outsiders, rather like
the jargon of management reform in the civil service. The current generation of
political scientists should look back a century to the elegance and clarity—though
not the views—of Dicey and Bryce, and even perhaps the wit of a Bagehot.
I defend the ‘private patois of political science’, replying to critics of the

notions of governance, the core executive, hollowing out, and the differenti-
ated polity. I defend my guide to understanding British government; to what
we are trying to understand and how we understand it. I do so because I am
convinced the old vocabulary for describing Westminster and Whitehall is at
best a partial description of how British government works. We need a new
language to capture the changes that have and continue to take place. And
here lies both a puzzle and a danger. The puzzle is that the new vocabulary is
not acceptable until approved by everyday use but it cannot be so approved
until we start using it. Fortunately, we have started using it. The word
‘governance’ has spread rapidly both because changing social theories have
led people to see the world differently, and because the world itself has
changed. New theories and practices have drawn attention away from the
central institutions of the state. The focus has shifted to the activity of
governing, and much of the activity of governing now involves private and
voluntary organizations as well as public ones.
In defending this patois, my objective is not to repair the wounded pride of
political science. It matters how we understand British government. Such
understandings are not the privilege of the chattering classes. If our existing
map of our institutions and how they work is faulty, we mislead citizens and
undermine representative democracy. Such maps are about how we are
governed, and politicians with faulty maps will make promises they cannot
keep, not because they are venal but because, unwittingly, they travel in the
wrong direction. I am trying to make corrections to the existing map of British
government so citizens and politicians alike know what journeys they can and
cannot take. Perhaps even a former political editor of The Times can look
beyond the received views of the commentariat to see that Westminster and
Whitehall elites confront a changing world that they are in part creating even
as it confounds their policies. Increasingly, this point is accepted by politicians:
What is New about ‘Network Governance’? 201
‘Question—If you hear the expression “the Westminster model”, what

does that mean to you?
Answer—Absolutely nothing
Question—What does the term “governance” mean?
Answer—Governance means the plurality of different agencies and
public delivery mechanisms that have changed the way that particular
aspects of our life are dealt with and moderated, or mediated, including
resources.’ (Both quotes from David Blunkett, former Secretary of State
for Education, interview 19 January 2007.)
So, the patois is spoken outside the confines of the groves of academe. We need
such new maps urgently. The ideas of the differentiated polity and the dis-
United Kingdom both anticipated the limits to the reach of our political class.
Neo-liberalism prized competition over cooperation and added divisiveness to
differentiation. Anti-politics flourished. The expectations gap between leaders
and led became wider. These changes were dramatized by the Scottish refer-
endum of 2014 and the EU referendum of 2016, which left our political class
floundering. As I have consistently argued, the reach of our political class was
over stated by commentators and practitioners alike. The differentiated polity
is an idea for our times. I return to this argument in Rhodes 2017, Volume II,
Chapter 12.
I concede three points to my critics at the outset. First, when trying to repair
a gap in the map of British government, there is always the danger of
appearing one-sided. I seek to counter the dominant view of British govern-
ment, which stresses that Britain is a unitary state with a strong executive. That
latter argument has been put so often by so many that it needs no restatement
by me. I do not dispute the British executive can act decisively. Obviously, the
centre coordinates and implements policies as intended at least some of the
time. But defenders of the Westminster model attach too little importance to
the sour laws of unintended consequences. Governments fail because they are
locked into power-dependent relations and because they must work with and
through complex networks of actors and organizations. To adopt a command
operating code builds failure into the design of the policy. Such centralization
will be confounded by fragmentation and interdependence, which, in turn,
will prompt further bouts of centralization.
Second, I plead guilty to using rhetorical devices. I still like phrases such as
‘from government to governance’, ‘the hollowing out of the state’, ‘the sour
laws of unintended consequences’, and ‘the differentiated polity’, although
they provoke my critics. Indeed, in Rhodes (2017, Volume II), I have added
such phrases as ‘genre blurring’, ‘plausible conjectures’, ‘greedy institutions’,
and ‘court politics’ to my repertoire. I prefer to state arguments baldly, and
then qualify them. I think my reasons for using such rhetorical devices are
transparently obvious, and further my aim of provoking new ways of seeing
British government.
Finally, I hope at least some of what I write is vexatious. It is odd that musicians
like Miles Davis or Bob Dylan are valued for their chameleon qualities. Their
willingness to experiment is valued over consistency. In academia, consistency is
prized above all. I cannot deny that my work shifts from a modernist-empiricist
to an interpretive approach. I mix mere description with interpretive ethnog-
raphy and the search for instrumental knowledge. Such dissociative tendencies
can edify the reader by finding ‘new, better, more interesting, more fruitful ways
of speaking about’ politics and government (Rorty 1980: 360). Along with Geertz
(1973: 29), my aim is to get better at vexing colleagues.
I start by revisiting the key concepts at the heart of the differentiated polity.
I reply to criticisms of the notions of policy networks, the core executive, and
governance. This chapter maps the shift from this modernist-empiricist
account of networks and to an interpretive approach. It is the bridge between
Volume I and Volume II. So, I turn to the decentred critique of the differen-
tiated polity revisiting the idea of the state and exploring the potential of
storytelling. I suggest that state authority is constantly being remade, negoti-
ated, and contested so hands-on steering has many limits. I discuss storytelling
as a hands-off way of steering. Finally, I outline the advantages of viewing the
UK as a differentiated polity and I summarize the decentred approach to
provide an introductory link to Volume II (Rhodes 2017).
K E Y CR I T I C I S M S
There are many criticisms of the differentiated polity and its core notions of
policy networks, the core executive, and governance. I will reply to my critics
under each of these headings. I will not even try to reply to every criticism—
with the benefit of 20/20 hindsight, much of the debate is unduly esoteric, even
self-absorbed. I focus only on the main charges (and for a listing of my critics,
see the Appendix).
Policy Networks
There are dead ends in the study of policy networks; for example, typologies of
networks. Amazingly, some debates were resolved. For example, the policy
network model seeks to explain why some groups are more powerful than
others in a network and why some networks are more powerful than others.
The critics deny the model is explanatory. Dowding (1994) claims the model
lacks an explanatory theory of power because it does not have ‘a modelling of
the bargaining process which can go beyond the mere labelling or shorthand
description contained in the policy network approach’ (see also Bennington

and Harvey 1994; Kassim 1993, and Chapter 4, this volume).
Despite the endless recycling of this criticism (see, for example, Wallace,
Pollack, and Young 2015: 56), it is wrong because the critics fail to consider the
power-dependence model that underpins my version of policy network the-
ory. This debate is not arcane. Power-dependence is a central feature of policy
networks. The distribution, and type, of resources within a network explains
the relative power of actors (individuals and organizations). The different
patterns of resource-dependence explain differences between policy networks.
There must be a theory to explain differences within and between networks.
Subsequently, Dowding (1995: 145 and n. 48) accepts that Rhodes’s (1981,
1986a, 1986b) version of policy networks is founded on the power-dependency
model and this ‘bargaining model and game theory can be fruitfully applied to
understand the nature of policy networks’. He concedes, ‘I am rightly accused of
ignoring the power-dependence model.’
Dowding (1995: 142–5) further complained that differences between types of
policy networks can be explained by the characteristics of the actors, denying
that the characteristics of networks are the centrepiece of the explanation. But
the network is a structure of rules and relationships; it is the chessboard on
which the pieces move. Occupying a position in a network gives access, often
privileged access, to resources and to the bargaining games governed by rules of
the game evolved in that network. So, the characteristics of actors stem from
their position in the network. Bargaining is governed by shared network rules.
The network both constrains and enables actors. It seems obvious that resource
exchange was always the centrepiece of my explanation, given the network was
always conceptualized as a structure of resource-dependent organizations. But,
equally, the inherited practices of networks shape resource exchanges. Both are
essential to the explanation of differences between networks and their outcomes.
Given that treating a network as a set of constraining rules is a central feature
of rational choice approaches to policy networks (Scharpf 1997), Dowding’s
objection was misplaced.
The problem is that the debate about power-dependence and network
structure became embroiled in a longer-running spat between proponents of
rational choice and of critical realism. Marsh and Smith’s (2000) dialectical
approach posits three dialectical relationships in the analysis of policy net-
works: between network structure and agents; between the network and its
socio-economic context; and between networks and the policy outcomes.
Dowding (2001) retorts by lecturing them on formal modelling and achieving
little beyond demonstrating the obvious; that he prefers rational choice
explanations (cf., for example, Dowding 2001 and Marsh and Smith 2001).
The two sides have irreconcilable differences of epistemology, theory, and
method. Despite our long-standing partnership, I did not co-author these
papers with Marsh for a simple reason. I did not, and still do not, agree with
his ‘realist’ dialectical approach. One of my referees commented waspishly on

such critical realist commentaries:
I find the ‘critical realist’ school most amusing. It seems to me that its supposed
adherents have not read their own founding texts . . . nor can they explain what it
is or why it’s useful. ‘Critical’ has become a default adjective for every subdisci-
pline these days, as though it deals with epistemological questions by a wave of
the hand, as well as legitimating politicised (semi-Marxist) analysis at the
same time.
It was neo-Marxism in drag (and see Rhodes 2017, Volume II, Chapter 12 for
further discussion).
Dowding (1995: 146) may have exonerated me from being a mere metaphor
but he insisted that the model should not have been developed ‘away from
considering the resources of actors in a game over policy outcomes’. I agree with
this part of his assessment and my later work on networks favoured a decentred,
actor-focused analysis of the games people play in networks (see Chapter 6, this
volume). With this shift of emphasis to meaning, the competing webs of
meaning, and their historical roots, many earlier critical exchanges are no longer
relevant. My emphasis falls, therefore, on where we go from here. I structure my
remarks around the criticisms that seem to me to have some force; namely, the
context of policy networks, and explaining change and the role of ideas. My aim
is to open new directions of research into networks and governance.
The Context of Policy Networks

This set of criticisms focuses on whether my approach places the analysis of
networks in a broader socio-economic context. For example, Marsh et al. (2003:
306) argue network governance ‘neglect the broader socio-economic structural
context within which politics takes place’. This ‘broader socio-economic struc-
tural context’ constrains and facilitates ‘the actions, and the likely success, of
individuals and interest groups in the British polity’. I take the general point that
changes in networks and governance must be placed in a broader context. Why
would I not? It is there for all to read in my initial account of the differentiated
polity; you sit around and ask for context—can’t you read?
Rhodes (1988: 48–77, 372–87 and figure 5.1) sets policy networks in the
context of an unstable external support system, the decline of the mixed
economy, the growth of the welfare state, the spread of functional differenti-
ation and professionalization, a social structure characterized by de-aligned,
multiple, and non-class cleavages, and a stable dominant, elite political trad-
ition. Further, I suggested that inexorable, impersonal forces such as the
functional differentiation of the modern state or the marketization of the
public sector explain the shift from hierarchy to a new governance of markets
and especially networks. Whatever the weaknesses of this formulation, it is
not convincing to claim that I neglected the broader context. Equally it is

inaccurate to describe the proffered explanation as ‘essentially . . . neopluralist’
and (unhelpfully) ‘pluralist’ (Marsh et al. 2003: 308 and 332). Networks are
explicitly described as an example of the oligopoly of the political marketplace
(see Chapter 1, this volume). This oligopoly is characterized by closed, struc-
turally inequal, private government, and elite pluralism (Marsh and Rhodes
1992a: 259, 263, 268). In this instance, as Barthes (1977: 145–6) observes, I am
a mere ‘scriptor’ who cannot provide the authoritative reading of his own text.
Rather, it is ‘eternally written here and now’ by each reader. Authorial
intention is no more when it comes to analysing the context in which networks
operate. That said, I no longer agree with my analysis of the broader context.
Latterly, I have focused on a decentred account of the state and the analysis
of traditions and dilemmas (see, this volume Chapter 6; and Rhodes 2017,
Volume II, Chapter 2).
Explaining Change
The most common and recurrent criticism of policy network analysis is that it
does not, and cannot, explain change (for a summary of the argument and
many more citations, see Richardson 2000). Policy network analysis stresses
how networks limit participation in the policy process; decide which issues
will be included and excluded from the policy agenda; shape the behaviour of
actors through the rules of the game; privilege certain interests; and substitute
private government for public accountability. It is about stability, privilege,
and continuity. So, there is force to the argument that the policy networks
literature in general pays too little attention to change and the role of ideas in
change. My work with Mark Bevir seeks to develop a decentred analysis of
British government that addresses this set of criticisms.
Decentred analysis produces detailed studies of people’s beliefs and prac-
tices. It begins from the insight that to understand actions, practices, and
institutions, we need to grasp the relevant meanings, the beliefs, and prefer-
ences of the people involved. I ask what the meaning of British governance is
to elites and to all who participate in, for example, a policy network. The
approach denies we can read off people’s beliefs from their institutional
position or their social class. A decentred study of a tradition or an institution
unpacks the ways in which each is created, sustained, and modified through
the beliefs, preferences, and actions of individuals in many arenas (and for a
detailed exposition, see Bevir and Rhodes 2003, 2006a, and Rhodes 2017,
Volume II, Chapter 2). It encourages us to recognize that the actions of these
individuals are not fixed by institutional norms or a logic of modernization,
but, on the contrary, arise from the beliefs individuals adopt against the
background of traditions and in response to dilemmas.
If historical analysis is the way to explore traditions and the context of

networks, then ethnographic analysis is the way to explore the beliefs and
practices of people, whether shared or contested. Policy network analysis should
make greater use of such ethnographic tools as: studying individual behaviour in
everyday contexts; gathering data from many sources; adopting an ‘unstruc-
tured’ approach; focusing on one group or locale; and, in analysing the data,
stressing the ‘interpretation of the meanings and functions of human action’
(paraphrased from Hammersley 1990: 1–2). The toolkit is varied. It is not
limited to participant observation, although it is a defining method. It also
encompasses textual analysis, historical archives, official documents, biograph-
ies, oral histories, recorded interviews, and informal conversations as well as
statistical and survey techniques (Shore 2000: 7–11). The task would be to write
‘thick descriptions’ or our ‘constructions of other people’s constructions of what
they are up to’ (Geertz 1973: 9, 20–1). We should focus on the social construc-
tion of policy networks through the ability of individuals to create meaning.
Rhodes’s (2003a, 2005, 2007a, 2011a) ethnographic studies of British government
provide several examples of decentred analysis (and, for examples of organiza-
tional anthropology, see Bate 1997; Linstead 1997; and Van Maanen 1988).
So, decentred analysis places agency and meanings at the heart of network
governance. It focuses on the diverse practices of governance, practices that
are themselves composed of multiple individuals acting on changing webs of
beliefs rooted in overlapping traditions. Patterns of governance arise as the
contingent products of diverse actions and political struggles informed by the
beliefs of agents as they arise in the context of traditions; situated agents (Bevir
and Rhodes 2006a: 4–11). This approach focuses on inherited beliefs and
ideas, on the games people play, and on the role of both in explaining how
the practices of network governance change.
The Core Executive
The origins of the idea of the core executive lie in conversations between
Patrick Dunleavy and me at the annual conference of the Political Studies
Association of the UK at the University of Aberdeen in April 1987. In skeletal
form, it was in Beyond Westminster and Whitehall (1988) that I characterized
the centre of British government as segmented:
executive authority is neither the sole preserve of prime ministers nor exclusive to
political leaders . . . decision making is fragmented between policy networks with
sporadic prime ministerial interventions. Ministers responsible for domestic
departments are, to a substantial degree, sovereign in their own turf. Coordin-
ation is achieved (if at all) and conflicts resolved (or at least suppressed) in and
by Cabinet and its multifarious committees, supplemented by bureaucratic
mechanisms. The means of bureaucratic co-ordination include the Treasury and
the public expenditure survey, interdepartmental committees, the Cabinet Office
and the official committees which ‘shadow’ ministerial Cabinet committees. Frag-
mentation of policy making has generated, therefore, a variety of co-ordinating
mechanisms and networks; a complex ‘central executive territory’ . . . Indeed, the
continuous growth of, and change in, the central executive territory attest to the
elusive nature of the goal of effective central co-ordination (Rhodes 1988: 76).
Patrick Dunleavy and I subsequently developed this notion of the segmented
executive to cover ‘the core executive’ (Dunleavy and Rhodes 1990, and
Chapter 9, this volume). The core executive approach rejects any notion of
dominance by any one actor or set of actors. The emphasis falls on fluidity and
shifting allegiances. Power no longer resides with any position. Rather it is
contingent and relational; that is, it depends on the relative power of other
actors. So, ministers depend on the prime minister for support in getting funds
from the Treasury. In turn, the prime minister depends on ministers to deliver
the party’s electoral promises. Both ministers and the prime minister depend
on, as Harold Macmillan allegedly put it, ‘events, dear boy, events’; for
example, on the health of the Chinese economy for the economic growth to
fund public services. This power-dependence approach focuses on the distri-
bution of such resources as money and authority in the core executive and
explores the shifting patterns of dependence between the several actors (see
Rhodes 1995a; Smith 1999; and Chapter 3, this volume). Thus, Norton (2000:
116–17) argues, ‘Ministers are like medieval barons in that they preside over
their own, sometimes vast, policy territory.’ Crucially, ‘the ministers fight—or
form alliances—with other barons in order to get what they want’ and ‘they
resent interference in their territory by other barons and will fight to defend it’.
So, the core executive is segmented into overlapping games in which all players
have some resources with which to play the game and no one actor is pre-
eminent in all games. The core executive is the set of networks that police the
functional policy networks (see also Burch and Holliday 1996; and the After-
word to Chapter 9, this volume).
From the outset the approach attracted criticism. Andeweg (1997: 59)
queries whether ‘coordination should be the defining function of Cabinet
government’. He suggests other functions such as ‘the provision of democratic
legitimation to government, or the creation of a channel for political account-
ability, or simply decision-making; the “authoritative allocation of values”. He
also claims the term coordination is unclear; ‘what is and what is not included
in that term’. In other words, the functional approach does not dispel ‘the fog
around the edges of the concept of Cabinet government’. Elgie (2011: 72–3)
concludes that core executive studies are ‘less innovative than they might at
first appear’ because they are an ‘updated version of the old prime ministerial vs.
cabinet government argument’. Yet, Elgie claims that ‘the resource-dependency
approach is almost completely dominant’. So, after two decades of core
executive studies, Elgie (2011: 71–2) concludes that the ‘the language of the
study of British central government has been transformed’; and ‘the concept has
travelled’. The main challenges to this orthodoxy were the presidentialism
thesis, and prime ministerial predominance, both of which assert that the core
executive remains overweening.
Presidentialism
Several commentators reject the differentiated polity’s contention that the core
executive is subject to many constraints. They argue it remains strong, and
claim the prime minister is now analogous to a president As one example
among many, Poguntke and Webb (2005: 5 and 7) argue that presidentializa-
tion has three faces: the executive face, the party face, and the electoral face.
Presidentialism occurs when there is a shift of ‘political power resources and
autonomy to the benefit of individual leaders’ along each face and ‘a concomi-
tant loss of power and autonomy of collective actors like cabinets’. They argue
these various shifts ‘generate a greater potential for, and likelihood of, this
“presidential” working-mode’ irrespective of regime (Poguntke and Webb
2005: 347). In other words, not just in Britain but in parliamentary govern-
ments worldwide, power is increasingly centralized on the core executive,
which has grown bigger, coordinates the other central networks, and inter-
venes both regularly and effectively across policy sectors.
The empirical evidence supporting such claims is inconclusive at best.
Centralization, pluralization, and personalization represent not a concentra-
tion of power, but an endless search for effective levers of control by a core
executive less powerful than many commentators and insiders claim (see Bevir
and Rhodes 2006a: ch. 6, and 2006b for a survey of the evidence, commentary,
and citations). The most obvious example of the limits to prime ministerial
power is the case of Gordon Brown as Chancellor of the Exchequer; ‘a great
crag standing in the way of a thoroughly monocratic government’ (Hennessy
2002: 21). Recognition of Brown’s authority requires us to shift from tales of a
Blair presidency to stories of at least a dual monarchy: ‘Brown conceived of the
new government as a dual monarchy, each with its own court’ (Rawnsley 2001:
20). British government was not presidential but a duumvirate, dominated by
two men presiding over territory ever more jealously guarded. Brown was
‘immovable’, ‘dominating his own territory’ with ‘jagged defences designed to
repel any invader, including the Prime Minister’. So, ‘they were not interested
in submerging their differences in outlook, but in making an exhibition of
them’ (Naughtie 2002: 352). Brown was reported as saying to Blair: ‘There is
nothing you could ever say to me now that I could ever believe’ (Peston 2005:
349). Brown became ‘the official opposition to Blair within the very heart of
the Cabinet’ (Peston 2005: 13 and 353).
Blair and Brown’s oscillating relationship exemplifies the shifting fortunes

and contingency of high politics. As Sartori (1997: 102) suggests in his
comparative analysis of parliamentary systems, prime ministers can be first
above equals, first among unequals, and first among equals. The British prime
minister can be each of these, but not at the same time. We live in a land where
barons vie for favour in the court of a would-be president as dependent on
them for support as they are on him for favours. Whichever shorthand
expression is preferred to describe British high politics, presidentialism is
not it. So, how does the ‘predominant’ prime minister fare?
Prime Ministerial Predominance

This thesis is associated with the work of Richard Heffernan and Mark
Bennister. For Heffernan (2003: 348) the proposition that power is relational
and based on dependency is ‘only partially accurate. Power is relational
between actors but it is also locational. It is dependent on where actors are
to be found within the core executive, and whether they are at the centre or the
periphery of key core executive networks.’ He agrees the core executive is
segmented, but disputes that power is as fragmented and dispersed as I have
suggested. An inherently unequal distribution of resources affords leaders
unique advantages, creating the potential for prime ministerial predominance.
Since power-dependence characterizes core executive relationships, it fol-
lows that attention should focus on the distribution and dispersal of resources
and shifting patterns of dependence between multiple actors. Prime ministers
command many ‘institutional resources’, including patronage, prestige,
authority, political centrality, and policy reach, knowledge, information, and
expertise, Crown Prerogative (for example, to delegate powers and responsi-
bilities to ministers and departments), and control of the agenda (Heffernan
2003: 356–7). They also have ‘personal resources’ such as reputation, skill,
and ability; association with actual or anticipated political success; public
popularity; and high standing in the party (Heffernan 2003: 351; 2005: 16).
It follows that the more resources a prime minister has, or can accumulate,
the greater their potential for predominance. But ministers also have access
to many resources that are not available to the prime minister, including ‘a
professional, permanent and knowledgeable staff, expert knowledge and relevant
policy networks, time, information and, not least, an annual budget’ (Heffernan
2005: 614).
So, the unit of analysis in core executive studies cannot be solely the prime
minister, nor can it be just the Cabinet; power is more widely dispersed.
Prime ministers remain key actors who, because of their access to institutional
and personal resources and their position at the centre of key networks
(Heffernan 2003, 2005), have the potential to exercise significant power. The
experiences of Tony Blair, John Howard, and Kevin Rudd, who at key points
in time were regarded as ‘predominant’ prime ministers, remind us that

unpredictable forces shape, constrain, and sometimes undermine leaders’
ability to get their own way. So, we must examine the relations between leaders
and their colleagues in Cabinet, the party room, and other ‘followers’ on
whom they also depend.
Indeed, both Bennister (2007: 328) and Heffernan (2005: 607) agree the
core executive approach need not necessarily abandon the idea of a strong
executive government; the two approaches are not mutually exclusive. Even
Heffernan’s (2003: 350) initial version of the argument had many qualifica-
tions. He suggests the prime ministerial authority is ‘contingent and context-
ual’. Prime ministers have the ‘potential’ to be predominant, ‘but only when
personal resources are married with institutional power resources, and when
the prime minister is able to use both wisely and well’. The prime minister’s
personal resources are ‘never guaranteed. They come and go, are acquired and
squandered, are won and lost’ (Heffernan 2003: 356). Moreover, ‘there is . . . a
vast and sprawling system of networks, committees and taskforces where most
work is undertaken’ (Bennister 2007: 335). Later versions of the prime min-
isterial dominance argument introduce more significant qualifications:
Prime ministers can be sure-footed or clumsy, be associated with policy
success or failure, have a low or a high party standing, a solid or a weak
parliamentary reputation, become electorally popular or unpopular. He or she
can preside over a happy or unhappy parliamentary party and can face weak
or powerful intra-party rivals. Prime ministers can be lucky or unlucky
and face strong or weak inter-party opponents. Often, an underperforming
economy, or some other such record of policy failure, can prove the instru-
ment of the prime minister’s downfall (Heffernan 2005: 616–17). As
I suggested in Chapter 9 (this volume), and as Elgie (1997) also suggested,
monocratic or prime ministerial government remains one of the varieties of
court politics to be found in the core executive with collective or Cabinet
government, ministerial government, segmented government, and so on.
While one pattern of executive politics may operate at any one time, there
can still be fluidity as one pattern is succeeded by another. Take, for example,
the rapid decline in public support for Kevin Rudd that followed his decision
to abandon his commitment to an emissions trading scheme. This policy
change created an opportunity for those angered by Rudd’s domineering
leadership style to harness discontent among ministers and the Labor caucus.
Allegiances, including those of Rudd’s former supporters, shifted to his dep-
uty, Julia Gillard. Over night, Rudd was ousted in a party-room challenge that,
lacking support, he didn’t contest. Indisputably, ‘predominance can . . . ebb
and flow’ (Bennister 2007: 340). It was the fate also of Margaret Thatcher
and Tony Blair, both for a time predominant prime ministers.
Few would have difficulty accepting that the prime minister is the ‘principal
node of key core executive networks’ (Heffernan 2005: 613; and see Burch and
Holliday 1996). So, I read the later Heffernan (2005) and Bennister and
Heffernan (2011) as an important set of qualifications to the prime ministerial
dominance argument. It is significant that they wrote their first version during
the heyday of the Blair ‘presidency’, while their qualifications reflect his later
decline. Now, we need to move beyond the increasingly stale debate about
prime ministerial predominance, which is now generating more heat than
light (see Parliamentary Affairs 66 (3) 2013). In his most recent article, in reply
to Dowding (2013), Heffernan (2013: 642, 643) emphasizes that the prime
ministers can have ‘more or less political capital’ and their ‘power waxes and
wanes’. These qualifications bridge the gap between their approach and the
core executive approach. Heffernan (2005: 616–17) downplays prime minis-
terial predominance, and opens the way for a convergence. The questions that
should be of central concern focus on changes in the standing of the prime
minister in central networks, and the fluctuating personnel and fortunes of
those networks.
So far, so modernist-empiricist. Decentred theory does not define the core
executive in functional terms by its core tasks in the system. The core executive
is a descriptive concept that captures the fluid and varying actors involved in
central decision-making. Core executives are characterized less by their insti-
tutions and functions than by their court politics. The court, or the core
network of the core executive, is the term conventionally used to refer to the
interactions of a leader and his immediate entourage. Court politics refers to
the beliefs, practices, and traditions of the networks of actors with the formal
authority of political and administrative leadership whose statecraft is a matter
of ruling (by rhetoric and manoeuvre), and rationalities (Rhodes 2014).
Studying court politics is about telling stories about the contending beliefs
and practices of governing elites; it is about providing our narratives of the
elite’s narratives. The notion of court politics is consistent with a decentred
approach because it focuses on the beliefs and practices of individuals. These
webs of belief and actions are located in inherited traditions and practices,
which constrain their actions; they are situated agents. The approach calls for a
political anthropology of the executive’s court politics (for observational
studies of prime ministers, ministers and Cabinets, and historical studies of
‘High Politics’, see Rhodes 2014; and Rhodes 2017, Volume II, Chapter 7).
Governance and the Decline of the State
There is a rather odd challenge to the governance narrative that questions

whether it is an accurate description (Colebatch 2009; Hughes 2010). Whether
the number of networks has grown or whether such networks are new are,
frankly, deeply uninteresting questions that miss the point. My main concern
is the spread of new ideas about markets and networks and the consequent
changes in the role of the state. Such sceptics are dealt with brusquely and
briskly by Torfing et al. (2012: 31–2). They argue there have been three
‘irreversible changes’: in the expectations of stakeholders about their involve-
ment in collaborative policy-making; in the shift of public bureaucracies to
‘open organisations . . . engaged in joint problem-solving and collaborative
service delivery’; and in the belief that network governance is ‘a legitimate
alternative to hierarchy and markets’. The new ideas had consequences.
Most critics have focused, correctly, on the changing role of the state and
challenged the idea that there has been a hollowing out or decline of the state.
They see a transformation rather than a weakening of the state.2 One example
must suffice and I examine the critique by Pierre and Peters (2000: 78, 104–5
and 111; Peters and Pierre 1998, 2009; Torfing et al. 2012) because their views
are typical and, undeniably, they have been persistent. They argue the shift to
network governance could ‘increase public control over society’ because
governments ‘rethink the mix of policy instruments’. They continue, ‘coercive
or regulatory instruments become less important and . . . “softer” instruments
gain importance’; for example, for steering instead of rowing. In short, the
state has not been hollowed-out but has reasserted its privileged position to
govern by regulating the mix of governing structures, such as markets and
networks, and deploying indirect instruments of control. There has been no
decline of the state. They argue the changes are not a zero-sum game and
governance has increased state control over civil society (Pierre and Peters
2000: 78).
It would seem that I am a ‘scriptor’ again, but one who is reluctant to
surrender his intentions. Of course, in replying to one’s critics, one should be
fair-minded, looking for common ground and ways forward. Such commend-
able virtues can be sorely challenged by some of the more egregious misrep-
resentations one encounters. There is some ground clearing to do before we
can move forward.
First, I agree with Scharpf (1997: 38 and 40) that, although hierarchical
coordination ‘remains a relatively rare phenomenon’, self-coordination among
units takes place in ‘the shadow of hierarchy’ because, for example, hierarchical
structures ‘define the context within which negotiations take place’. There is
nothing new here because I rehearsed this argument about the continuing
importance of hierarchy (in Rhodes 1986b: 4–7 and again in Rhodes 1999a:
114–16) as well as arguing for the continuing importance of bureaucracy in
Australian (Davis and Rhodes 2000) and British government (Rhodes 1994;
and Chapter 8, this volume).
2
See, for example, Bell and Hindmoor 2009; Jessop 2000; Jordan et al. 2005; Kjær 2004;
Newman 2005; Marsh 2008a, 2008b, 2009, 2011; Marsh, Richards and Smith 2003; Pierre and
Peters 2000; Saward 1997; Skelcher 2000; Taylor 2000; Torfing et al. 2012; and Walters 2004. For
a full listing, see the Appendix.
Second, some claim my views are ‘extreme’ (Torfing et al. 2012: 3). It is
difficult to reconcile such assertions with what I said. From the outset, I argued
that governments had to choose between three main governing structures of
bureaucracy, markets, and networks:
British government can choose between ‘governing structures’. To markets and
hierarchies, we can now add networks
(Chapter 8, p.159, this volume, and Rhodes 1996: 653).
No governing structure works for all services in all conditions. The issue, there-
fore, is not the superiority of markets and hierarchy over networks, but managing
networks in the conditions under which they work best
(Chapter 11, p. 185, this volume, and Rhodes 1997b: 48–9).
Indeed, the title of the 1997 article, ‘It’s the Mix that Matters’, might suggest
that I saw the state’s key task as steering through some mix of markets,
hierarchies, and networks. Torfing et al. (2012) and Peters and Pierre (2000,
2009) find it impossible to get beyond the eye-catching phrase ‘from govern-
ment to governance’ to grasp the essentials of my arguments. For example,
Torfing et al. (2012: 14) define interactive forms of governance as:
the complex process through which a plurality of social and political actors with
diverging interests interact in order to formulate, promote, and achieve common
objectives by means of mobilizing, exchanging, and deploying a range of ideas,
rules, and resources.
The definition accords no special place to ‘command and control’ despite their
stress on the core role of the state. Rather, they stress that complexity,
common objectives, and decentring are the three key features of this defin-
ition. Governments . . . ‘often play a crucial role as facilitator and manager . . .
but there is no privileged centre in public policy-making, but a number of
competing actors and arenas’ (Torfing et al. 2012: 15, emphasis added; see also
Peters and Pierre 2009: 92). Moroever, Ansell and Torfing (2016: 552) concede
that the argument about self-organization is a common theme, not an extreme
position. Given that I also said the state does not occupy a privileged sovereign
position, that the relationship is asymmetric, that centralization must coexist
with interdependence; and that the state can imperfectly steer (Rhodes 1997a:
199), it is difficult to see how their account differs from mine. Frankly, it would
be hard to get a slip of Rizla paper between my views and those of many of my
critics (cf. Peters 1994), although I concede that my emphasis fell on the role of
non-state actors because others focused on the state to the exclusion of all else.
It is clear from my account of the British state (Chapter 8, this volume) that
there is a new meta-narrative around austerity and ‘the new governance’.
Today’s neo-liberal states have come into being and are perpetually reconsti-
tuted through constant reform initiatives. Many of the individual reform
initiatives have their roots in neo-liberal ideas about increased privatization,

deregulation, and reductions in government spending. A recurrent theme in
my work is the changing role of the state, asking whether the state has been
rolled back to create the minimalist state or whether it is rolling out to extend
its influence by outsourcing and incorporating others in public governance. Of
course it is both, and my original version of hollowing out did not allow for
both these trends (and see Rhodes 2017, Volume II, Chapter 12 for further
discussion).
Metagovernance
There have been two much discussed waves of theory about governance;
network governance (see Chapters 2–4, this volume), which was followed by
metagovernance. I consider this second wave here.
Metagovernance refers to the role of the state in securing coordination in
governance and its use of negotiation, diplomacy, and more informal modes
of steering. As with network governance, metagovernance comes in several
varieties (Sørensen and Torfing 2007a: 170–80). They share a concern, how-
ever, with the varied ways in which the state now steers organizations,
governments, and networks rather than directly providing services through
state bureaucracies, or rowing. These other organizations undertake much of
the work of governing; they implement policies, they provide public services,
and at times they even regulate themselves. The state governs the organiza-
tions that govern civil society; ‘the governance of government and governance’
(Jessop 2000: 23). Moreover, the other organizations characteristically have a
degree of autonomy from the state; they are often voluntary or private sector
groups or they are governmental agencies or tiers of government separate
from the core executive. So, the state cannot govern them solely by the
instruments that work in bureaucracies.
Torfing et al. (2012: 156–9 and ch. 7) suggest the traditional role of the
public service is ‘supplemented’ (not replaced) with that of the ‘meta-governor
managing and facilitating interactive governance’. Their task is to ‘balance
autonomy of networks with hands-on intervention’. They have various spe-
cific ways of carrying out this balancing act. They become ‘meta-governors’
managing the mix of bureaucracy, markets, and networks (see also Koliba
et al. 2011, xxxii and ch. 8; and Rhodes 1997b and Chapter 11, this volume).
There are several ways in which the state can steer the other actors involved
in governance (see, for example, Jessop 2000: 23–4; 2003; Torfing et al. 2012:
ch. 7). First, the state can set the rules of the game for other actors and then
leave them to do what they will within those rules; they work ‘in the shadow of
hierarchy’. So, it can redesign markets, re-regulate policy sectors, or introduce
constitutional change. It can supplement such hands-on measures with, sec-
ond, hands-off steering through storytelling. It can organize dialogues, foster
meanings, beliefs, and identities among the relevant actors, and influence what
actors think and do. Third, the state can steer by the way in which it distributes
resources such as money and authority. It can play a boundary spanning role;
alter the balance between actors in a network; act as a court of appeal when
conflict arises; rebalance the mix of governing structures; and step in when
network governance fails. Of course, the state need not adopt a single uniform
approach to metagovernance. Finally, they have a small ‘p’ political role that
can involve campaigning for a policy and forming alliances with politicians. It
can use different approaches in different settings at different times (and for a
list of the more specific skills of network management, see Chapter 5, this
volume).
So, the neutral, competent servants of the political executive must now
master the skills for managing the complex, non-routine issues, policies, and
relationships in networks; that is, metagoverning, boundary spanning, and
collaborative leadership. The task is to manage the mix of bureaucracy,
markets, and networks (Rhodes 1997b; and Chapter 11, this volume). The
public service needs these new skills, although it is a step too far to talk of these
new skills requiring ‘a full blown cultural transformation’ (Goldsmith and
Eggers 2004: 178; cf. Rhodes 2016).
For all the different emphases, the first two waves of governance share
common features. First, proponents of metagovernance take for granted the
characteristics of network governance. They agree networks are characterized
by trust and diplomacy. They accept that states are becoming increasingly
fragmented into networks based on several different stakeholders; and the
dividing line between the state and civil society is becoming more blurred
because the relevant stakeholders are private or voluntary sector organiza-
tions. So, Jessop (2000: 24) concedes, ‘the state is no longer the sovereign
authority . . . [it is] less hierarchical, less centralised, less dirigiste’. There is a
shared modernist-empiricist description of the characteristics of network
governance (see also Sørensen and Torfing 2007b).
Second, the analysis of metagovernance not only recognizes non-state
actors by granting them the power to self-regulate but also distinguishes
them from the state so creating the space for the state to exert macro-control
over their self-regulation. The state governs the other actors involved in
governance. In other words, metagovernance heralds the return of the state
by reinventing its governing role; it is ‘bringing the state back in (yet again)’
(Jessop 2007: 54). This return to the state opens opportunities for policy advice
on the practice of metagovernance. The two waves share a common concern
with providing advice on network governance. Both assume the role of the
state is to manage, directly and indirectly, the networks of service delivery. For
example, Part III of Sørensen and Torfing (2007b, chs 10–12) on ‘metagover-
nance’ is devoted to such topics as governing the performance of networks,
institutional design, and network management, and the possibilities for public
authorities to shape network outputs. They are not alone. Although it is not
rocket science, nonetheless the literature on network steering has proliferated
over the past decade (see Chapter 5, this volume). This work treats govern-
ment departments, local authorities, markets, and networks as fixed structures
that governments can manipulate using the right tools. It seeks to improve the
ability of the state to manage the mix of hierarchies, markets, and networks,
and of state managers to steer these structures.
Third, both narratives rely on a reified notion of structure. The proponents
of first-wave governance are self-confessed modernist-empiricists with a
reified notion of structure rooted in an explicit social science theory of
functional differentiation. The proponents of metagovernance also continue
to claim the state is a material object, a structure, or a social form. They draw
on critical realist epistemology and such notions as ‘emergence’ and ‘mech-
anisms’ ostensibly to guard against the charge of reification (see, for example,
Jessop 2007; and Rhodes 2017, Volume II, Chapter 12).
A DEC E N TR E D CR I T I Q U E
The State
Initially, I saw the state from an institutional, neo-Weberian perspective.

Following Nordlinger (1981: 9), I defined the state as ‘a complex set of
institutional arrangements for rule operating through continuous and regu-
lated activities of individuals acting as occupants of offices’ (Rhodes 1988:
97–8). It was a reified notion of the state that treated it as a structure.
A decentred approach rejects such essentialist definitions, arguing for a
more diverse view of state authority and its exercise. Decentred theory rejects
the notion of the state as a material object and governance as an emergent
structure. It is a ‘stateless’ theory in the sense that it rejects the idea of the
state as a pre-existing causal structure that can be understood as having an
autonomous existence and causal effects over and apart from people’s beliefs
and actions. The state is just an aggregate description for a vast array of
meaningful actions that coalesce into contingent, shifting, and contested
practices.
All patterns of rule arise as the contingent products of diverse actions and
political struggles informed by the varied beliefs of situated agents. So, the
notion of a monolithic state in control of itself and civil society was always a
myth. The myth obscured the reality of diverse state practices that escaped the
control of the centre because they arose from the contingent beliefs and
actions of diverse actors at the boundary of state and civil society. The state
is never monolithic and it always negotiates with others. Policy always arises
from interactions within networks of organizations and individuals. Patterns
of rule always traverse the public, private, and voluntary sectors. The bound-
aries between state and civil society are always blurred. Transnational and
international links and flows always disrupt national borders. In short, state
authority is constantly being remade, negotiated, and contested in widely
different ways within widely varying everyday practices.
Both network governance and metagovernance have an instrumental or
steering view of network. Networks are seen as structures to be managed by
the state and a tool of greater state control. It is the dominant mainstream
academic, and official, view of networks. This approach confronts the sour
laws of network governance; managing the mix, the problem of many hands,
the holy grail of coordination, and local ownership (see Chapter 5, this
volume). It treats government departments, local authorities, markets, and
networks as fixed structures that governments can manipulate using the right
tools (and for a survey of such tools, see Salamon 2002). It seeks to improve
the ability of the state to manage the mix of hierarchies, markets, and networks
that have flourished since the 1980s. It is a logical extension of the neo-
Weberian view of the state.
A decentred view of such steering challenges this instrumental approach.
Local networks cease to be local networks when they are centrally manipulated
or directed. In effect, when networks are centrally managed, horizontal rela-
tionships are transformed into vertical relationships. Such relationships are
better described as exercises in official consultation; at least this phrase does
not imply any local discretion or local ownership. But the effect is that central
management of local networks threatens their autonomy, distinctiveness,
and effectiveness. This threat arises because any pattern of governance is a
product of diverse practices that are themselves composed of multiple indi-
viduals acting on all sorts of conflicting beliefs that they have reached against
the background of many traditions and in response to varied dilemmas.
So, a decentred approach sees network governance arising from the bottom
up and suggests that central intervention will undermine the bottom-up
construction of governance, provoking resistance, and generating unintended
consequences.
Storytelling
A decentred approach undercuts the idea of network steering as a set of tools by

which we can manage governance. If governance is constructed differently,
contingently, and continuously, we cannot have a toolkit for managing it. This
line of reasoning challenges the idea of expertise as a basis for policy-making (see
also Rhodes 2017, Volume II, Chapter 10). Decentred narratives offer a different
approach to policy advice. Instead of revealing policy consequences through
insights into a social logic or law-like regularities, they enable policy makers to
see things differently; they exhibit new connections within governance and
new aspects of governance. In other words, a decentred approach treats policy
advice as stories that enable listeners to see governance afresh (Bevir 2011). An
interpretive approach encourages us to give up management techniques and
strategies for a practice of learning by telling stories and listening to them.
While statistics, models, and claims to expertise all have a place in such stories,
we should not become too preoccupied with them. On the contrary, we should
recognize that they too are narratives about how people have acted or will
react given their beliefs and desires. No matter what rigour or expertise we
bring to bear, all we can do is tell a story and conjecture what the future
might bring.
It is not uncommon for critics to talk of the ‘impossibility’ of a ‘positive
contribution’ to policy analysis from an interpretive approach because it is
‘descriptive rather than evaluative or critical’ (Bobrow and Dryzek 1987: 171).
I demur from this judgement, as does the organizational studies literature on
storytelling.3
My starting point is the idea that any organization ‘always hinges on the
creation of shared meaning and shared understandings’, with metaphors
exercising a ‘formative impact’ on the construction of meaning (Morgan
1993: 11 and 276–80; see also Weick 1995: ch. 8). Stories spell out the shared
meaning and shared understandings. Of course, stories come in many versions
and often have no clear beginning and no ending. They are provisional and
unfolding. In telling the stories, we freeze them at one point in time. They can
appear set in stone. So, they unfold constantly.
In a British government department, there is at least one departmental
philosophy and it is the storehouse of many stories. It is a form of folk
psychology. It provides the everyday theory and shared languages for story-
telling. It is the collective memory of the department. Institutional memory
resides in the stories people tell one another; ‘stories are to the storytelling
system what precedent cases are to the judicial system’. They were used to
‘formulate recognizable, cogent, defensible and seemingly rational collective
accounts that will serve as precedents for individual assumption, decision and
action’ (Boje 1991: 106).
Most if not all civil servants will accept that the art of storytelling is an
integral part of their work. Such utterances as: ‘Have we got our story straight?’,
‘Are we telling a consistent story?’, and ‘What is our story?’ abound. Civil
servants and ministers learn and filter current events through the stories they
3
See, for example, Czarniawska 1998, 2004; Gabriel 2000; and Denning 2004, 2007. There is
even a book on storytelling for business in the ‘For Dummies’ series (Dietz and Silverman 2013).
See also Chapter 3, this volume, for a more extended discussion.
hear and tell one another. It is an integral part of the everyday practice of civil
servants. Stories explain past practice and events and justify recommendations
for the future. It is an organized, selective, retelling of the past to make sense of
the present.
Public servants know they tell the minister stories. Stories come in many
forms. Some stories are short. They are told in a single sentence. When you
belong to the same organization, the listener can unpack these stories. They do
not need to be recounted in full. The shortest example is ‘you know’ as in you
know the story already. For example, one short story told to new recruits is
that ‘there is a bit of mystique around ministers and they make you feel
inferior’. It invokes the idea of hierarchy, the subordinate role of civil servants,
and the ceremonial side of being the Queen’s minister. Its meaning is clear:
‘you are a subordinate’. Gossip is another form of storytelling; personalized
with a variable regard for accuracy. Submissions and briefs are stories by
another name and recognized to be so by the civil servants who tell them.
When the minister resigned, the civil servants asked: ‘What is our story?’ They
wanted to find out what had happened. They talked of ‘getting the story
straight’, ‘getting it together’, ‘we’ve got the story’, ‘when you have the narra-
tive’, and ‘we’ve reached agreement on some of the main storylines’. Officials
were also explicitly invited to tell a story. So, the interpretive approach has a
technique for policy analysis—storytelling—which is both recognized by
managers and provides guides for managerial action.
Rein (1976: 74–5) suggests advice is based ‘on social understandings and
depends on the use of illustrative stories, or accounts from past experience’.
In his view, policy narratives present a chronology or sequence of linked
events, using a few major characters, and each step in the story ‘causes’ the
next step. There is a storyline or, if you will, a beginning, middle, and end
(although, of course, that ‘end’ is the start of the next story). The central
element in the story is the metaphor (or making the unfamiliar analogous
to familiar situations). ‘The simplest stories are proverbs and parables, used
to justify policy relevant stories’ (Rein 1976: 266) and so there is usually a
moral to the tale. The validity of stories is assessed by rules that are ‘partly
aesthetic and partly logical’. The story should be ‘the simplest, most com-
prehensive, internally consistent explanation we can offer’. We should also
ask if the explanation in the story could be generalized. The tasks of the civil
servant, therefore, are to invent stories, to design programmes of interven-
tion based on the stories, and to criticize the stories others commend (Rein
1976: 268).
Storytelling is closely linked to performance. In Rhodes (2011a) storytell-
ing had three characteristics: a language game, a performing game, and a
management game. The language game identified and constructed the story-
line, answering the questions of what happened and why. The performing
game told the story to a wider audience, inside and outside the department.
Officials tested the facts and rehearsed the storyline in official meetings to see
how their colleagues responded. They had to adapt the story to suit the
minister, and both ministers and officials had to judge how the story would
play publicly. They then performed that agreed story to the media, parlia-
ment, and the general public. Finally, there was the management game,
which both implemented any policy changes and, perhaps even more
important, let them get on with ‘business as usual’ as quickly as possible.
The resulting story had to be reliable, defensible, accurate, and consistent
with the department’s traditions. As Fawcett (2016: 52) argues, the analysis
of storytelling requires us to understand not only the construction and
performance of stories but also their reception; why do some stories
capture the imagination when others fail? If storytelling is an important
metagoverning tool we need to examine the successes and failures of different
types of stories and ways of telling them.
Crucially, stories or narratives are not just chronological accounts of
events or people but also they explain actions. I use the term ‘narrative’ to
refer to the form of explanation that disentangles beliefs and actions to
explain human life (see Rhodes 2017, Volume II, Chapter 2). Narratives
are the form theories take in the human sciences; the analytical tools
dissecting beliefs and practices.4 It is often claimed that positivist political
science provides causal explanations while interpretive approaches provide
understanding of beliefs, motives, and actions. Not so. Narratives explain
actions. Scholars from all sorts of disciplines use the word ‘cause’ to signal
there is a significant relationship between people and events. Narrative is a
form of explanation that works by relating actions to the beliefs and desires
that produce them. Narratives depend on conditional connections. When
individuals act on their beliefs and desires, there is a conditional connection.
Conditional connections are neither necessary nor arbitrary. Because they
are not necessary, political science differs from the natural sciences. Because
they are not arbitrary, we can use them to explain actions and practices. So,
narratives identify the conditional connections that link people, events, and
ideas to one another and explain actions and practices. Although these
narrative structures also appear in works of fiction, we need not equate
political science to fiction. Political scientists offer us narratives that strive,
to the best of the narrator’s ability, to capture the way in which events did
happen in the past or are happening today, whereas writers of fiction need
not do so. Political scientists cannot ignore the facts, although we must
accept that facts, agreed or otherwise, are never simply given to them.
4
There is a massive literature on narratives. I found the following helpful: Barthes 1993; Bevir
1999: 252–62 and 298–306; 2000; 2006; Ricoeur 1981, ch. 11; 1991, ch. 6; and White 1973, 1987.
CONCLUSIONS
This talk of networks, governance, and recovering meaning might seem otiose;
the navel gazing of academics. It is important for all of us because the current
map of—the everyday understandings about—British government is seriously
misleading. Government is complex and in constant flux. The beast feeds on
itself—policy is its own cause. Our political leaders are hemmed in by their
inherited beliefs and practices, which do not recognize there are many con-
tending traditions. For academics to talk of a British governmental tradition is
to reinforce this misperception. We need to acknowledge the diverse frames
brought to bear on understanding and constructing policy. Imposing a dom-
inant frame erodes trust, legitimacy, and effectiveness. Local knowledge vies
with modernist expertise and frustrates implementation. These two books
focus on providing a vocabulary for a more accurate description of the
world. Edification is a way of revising the map. We need new maps that
build in fluidity, contestation, and complexity. Government is no longer
about the dominant prime minister or the power of bureaucracy but about
the intersection of multiple frames of governance encapsulating but not
limited to markets, networks, and hierarchies.
The Differentiated Polity Revisited
The differentiated polity identifies important empirical gaps in the Westminster

model and key changes in British government. It focuses on the dilemma that
arise as the ideas and practices of the centralized Westminster government
conflict with the ideas of practices of the differentiated polity. It opens new
avenues of exploration on key issues confronting policy-making and policy
implementation in the 1980s and 1990s, including: the sectoral character of
policy-making; the mix of governing structures; the philosopher’s stone of
central coordination; devolution to the constituent territories of the UK; and
the relentless rise of intergovernmental diplomacy. The differentiated polity
narrative is best seen as a corrective to the traditional Westminster model.
I use it to develop a new way of seeing state authority in its relationship to
civil society.
This narrative is not just a story that academics tell to one another. It is
hard to draw a clear-cut distinction between academic commentators and
elite actors. They share ideas and assumptions about how the system works.
These images or organizing perspectives are common currency (see pp. 173–4
above). Such is the case for network governance; elite actors talk of holistic
governance and of joined-up government (see Cm 4310, 1999; and Bevir 2005:
29–30 and 48–51).
When we provide a definition or general account of governance, it should

be couched as a set of ‘family resemblances’. Wittgenstein (1972: 17–20)
famously suggested that general concepts such as ‘game’ should be defined
by various traits that overlapped and criss-crossed in much the same way as
do the resemblances between members of a family—their builds, eye colour,
gait, personalities. He considered various examples of games to challenge the
idea that they all possessed a given property or set of properties—skill,
enjoyment, victory, and defeat—by which we could define the concept.
Instead, he suggested the examples exhibited a network of similarities, at
various levels of detail, so they coalesced even though no one feature was
common to them all.
We do not master such family resemblances by discovering a theory or rule
that tells us precisely when we should and should not apply it. Our grasp of the
concept consists in our ability to explain why it should be applied in one case
but not another, our ability to draw analogies with other cases, and our ability
to point to the criss-crossing similarities. Our knowledge of ‘governance’ is
analogous to our knowledge of ‘game’ as described by Wittgenstein. It is
‘completely expressed’ by our describing various cases of governance, showing
how other cases can be considered as analogous to these, and suggesting that
we would be unlikely to describe yet other cases as ones of governance.
Some of the family resemblances that characterize governance derive from a
focus on meaning in action and apply to all patterns of rule. A decentred
approach highlights, first, a more differentiated view of state authority and its
exercise. All patterns of rule arise as the contingent products of diverse actions
and political struggles informed by the varied beliefs of situated agents. State
authority is constantly remade, negotiated, and contested in widely different
ways within widely varying everyday practices.
A decentred approach suggests, second, these everyday practices arise from
situated agents whose beliefs and actions are informed by traditions and
expressed in stories. In every government department, we can identify depart-
mental traditions, often embodied in rituals and routines. For example, British
civil servants are socialized into the broad notions of the Westminster model,
such as ministerial responsibility, as well as the specific ways of doing things
‘around here’; they are ‘socialised into the idea of a profession’ and learn ‘the
framework of the acceptable’ (Bevir and Rhodes 2006a: ch. 7). Governance is
not any given set of characteristics. It is the stories people use to construct,
convey, and explain traditions, dilemmas, beliefs, and practices.
A decentred approach also might help to highlight a third family resemblance
that characterizes British governance but might not be found in patterns of rule
in other times or places. In Britain, the neo-liberal reforms of the Conservatives
and New Labour have brought about a shift from hierarchy to markets to
networks. While this shift is widely recognized, a decentred approach suggests,

crucially, that it takes many diverse forms; it is a contingent mix.
A fourth family resemblance is that the central state has adopted a less
hands-on role. Its actors are less commonly found within various local and
sectoral bodies, and more commonly found in quangos concerned to steer,
coordinate, and regulate such bodies. Once again, a decentred approach sug-
gests, crucially, that such steering, coordination, and regulation take many
diverse hands-off forms and involve many non-state actors. Governance is
found in many and new forms.
A decentred approach highlights the resemblances that contribute to a
general characterization of governance and a more specific characterization
of governance in Britain. It highlights plurality, the constructed nature of the
state, the oscillating mix of governing structures, and self-organizing net-
works. Nonetheless, it disavows any logic to the specific forms that governance
takes in particular circumstances. So, a decentred approach resolves the
theoretical difficulties that beset earlier waves or narratives of the changing
state. It avoids the unacceptable suggestion that institutions fix the actions of
individuals in them rather than being products of those actions. It replaces
unhelpful phrases such as path-dependency with an analysis of change rooted
in the beliefs and practices of situated agents. Yet it allows political scientists to
offer aggregate studies by using the concept of tradition to explain how people
come to hold beliefs and perform practices.
A Decentred Approach: Summary
A decentred account of differentiated polity is distinctive in seven ways.

• It represents a shift of topos from institutions to meanings in action.
• Institutions, whether a policy network or a prime ministerial office, do
not have essentialist features, only family resemblances that are construct-
ed, contested, and contingent.
• Decentred analysis explains shifting patterns of governance by focusing
on the actors’ own interpretations of their beliefs and practices, not
external causes such as a global financial crisis.
• The everyday practices arise from agents whose beliefs and actions are
informed by traditions.
• It explores the diverse ways in which situated agents are changing the
boundaries of state and civil society by constantly remaking practices as
their beliefs change in response to dilemmas.
• It reveals the contingency and contestability of narratives. It highlights

both the importance of local knowledge and the diversity of state author-
ity and its exercise.
• It provides instrumental knowledge expressed in stories.5
I believe this approach casts a searchlight on government that finds new and
fruitful ways to talk about it. I also agree with Marinetto (2003: 605) that the
Anglo-Governance School will give ways to alternative ways of understanding
government. I have suggested that decentred theory is that new way.
Until recently, the interpretive approach was a fringe preoccupation in
British political science. The British Academy’s study of the British contribu-
tion to political science in the twentieth century (Hayward et al. 1999) illus-
trates the point. In a text of 511 pages, anti-foundational approaches in their
multifarious guises are conspicuous only by their absence. The index has a
mere three passing comments on postmodernism. There can be no defence
that the book is about British political science because there are many entries
for American political science. In short, the mainstream writes the interpretive
turn out of its story of British political science. Not so in other disciplines. For
example, in anthropology, Inglis (2000: 112) opines that the work of such
Anglo-Saxon philosophers as Charles Taylor, Peter Winch, and Alasdair
MacIntyre—to whom I would add Richard Bernstein and Richard Rorty—
means that using the methods of the natural sciences in the human sciences is
‘comically improper’. Political scientists remain improper. We are rarely
funny and the mainstream prevails (Bevir and Rhodes 2010: ch. 2). Yet the
horizons of British political science broaden, if slowly.6
As a ‘scriptor’, I am not surprised at egregious criticism. I know how to
criticize myself, and have done so at several points above. Not only do these
several chapters mean something different to each and every reader but, with
the benefit of 20/20 hindsight, they mean something different to me. How that
came about is the central theme of Rhodes (2017, Volume II) on an interpret-
ive political science.
5
On the debate about decentred theory see Bevir and Rhodes 2006a: ch. 3; Finlayson et al.
2004; McAnulla 2006a, 2006b; and Hay 2002. I will outline and discuss the interpretive approach
and its critics in Rhodes 2017, Volume II, Chapters 2 and 12.
6
There is now a growing literature on this ‘interpretive turn’ in policy analysis and public
administration. See Fischer 2003; Fischer and Forester 1993; Fischer et al. 2015; Hajer 2009;
Hajer and Wagenaar 2003; Roe 1994; Stone 2011; Wagenaar 2011; and Yanow 1996. See also the
bibliography in Bevir and Rhodes 2012.
APPENDIX
Bibliography on the Anglo-Governance Debate
Criticism and Commentary

6, Perri (2010) ‘When Forethought and Outturn Part: Types of Unanticipated and
Unintended Consequences’. In H. Margetts, Perri 6, and C. Hood (eds), Paradoxes
of Modernise. Unintended Consequences of Public Policy Reform. Oxford: Oxford
University Press, pp. 44–60.
6, Perri (2015) ‘Governance: If Governance Is Everything, Maybe It’s Nothing’. In
A. Massey and K. Johnston (eds), The International Handbook of Public Adminis-
tration and Governance. Cheltenham: Edward Elgar, pp. 56–80.
Andeweg, R. (1997) ‘Collegiality and Collectivity: Cabinets, Cabinet Committees and
Cabinet Ministers’. In P. Weller, H. Bakviss, and R. A. W. Rhodes (eds), The Hollow
Crown. Houndmills, Basingstoke: Macmillan, pp. 58–83.
Bellamy, C. (2011) ‘The Whitehall Programme and After: Researching Government in
Time of Governance’, Public Administration, 89: 78–92.
Börzel, T. A. (1998) ‘Organizing Babylon: On the Different Conceptions of Policy
Networks’, Public Administration, 76: 253–73.
Börzel, T. A. (2011) ‘Networks: Reified Metaphor or Governance Panacea?’, Public
Administration, 89: 49–63.
Clifton, J. (2014) ‘Beyond Hollowing Out: Straitjacketing the State’, The Political
Quarterly, 85: 437–44.
Colebatch, H. K. (2009) ‘Governance as a Conceptual Development in the Analysis of
Policy’, Critical Policy Studies, 3: 58–67.
Dowding, K. (1995) ‘Model or Metaphor? A Critical Review of the Policy Network
Approach’, Political Studies, 43, 136–58.
Dowding, K. (2001) ‘There Must Be an End to Confusion: Policy Networks, Intellec-
tual Fatigue, and the Need For Political Science Methods Courses in British
Universities’, Political Studies, 49: 89–105.
du Gay, P. (2007) Organizational Identity: Persons and Organizations ‘After Theory’.
London: Sage, ch. 7.
Elgie, R. (2011) ‘Core Executive Studies Two Decades On’, Public Administration,
89: 64–77.
Fawcett, P. (2016) ‘Critical Encounters with Decentred Theory: Tradition, Metago-
vernance and Parrhēsia as Storytelling’. In N. Turnbull (ed.), Interpreting Govern-
ance, High Politics and Public Policy: Essays Commemorating Interpreting British
Governance. Routledge Studies in Governance and Public Policy. Abingdon, Oxon:
Routledge, pp. 39–56.
Finlayson, A. (ed.) (2008) ‘Symposium on Governance Stories and Interpreting British
Governance by Mark Bevir and R. A. W. Rhodes’, Political Studies Review 6 (2)
2008: 143–77.
Griffin, L. (2012) ‘Where Is Power in Governance? Why Geography Matters in the
Theory of Governance’, Political Studies Review, 10: 208–20.
226 Appendix
Holliday, I. (2000) ‘Is the British State Hollowing Out?’, The Political Quarterly,
71: 167–76.
Hughes, Owen (2010) ‘Does Governance Exist?’ In S. Osborne (ed.), The New Public
Governance: Emerging Perspectives on the Theory and Practice of Public Governance.
London and New York: Routledge / Taylor & Francis, pp. 87–104.
Jessop, B. (1990) State Theory. Cambridge: Polity Press.
Jessop, B. (2000) ‘Governance Failure’. In G. Stoker (ed.), The New Politics of British
Local Governance. Houndmills, Basingstoke: Macmillan, pp. 11–32.
Jordan, A., Wurzel, R. K. W., and Zito A. (2005) ‘The Rise of “New” Policy Instru-
ments in Comparative Perspective: Has Governance Eclipsed Government?’, Polit-
ical Studies, 53: 477–96.
Kassim, H. (1993) ‘Policy Networks, Networks and European Union Policy Making:
A Sceptical View’, West European Politics, 17: 15–27.
Kjaer, A. M. (2004) Governance. Cambridge: Polity Press.
Kjær, A. M. (2011) ‘Rhodes’ Contribution to Governance Theory: Praise, Criticism
and the Future Governance Debate’, Public Administration, 89: 101–13.
LeGalès, P. and Thatcher, M. (eds) (1995) Les Reseaux de Politique Publique. Debat
Autour des Policy Networks. Paris: Editions L’Harmattan.
Marsh, D. and McCaffrie, B. (2015) ‘One Cheer for Jordan and Cairney: Taking the
Governance Literature Seriously’, British Politics, 10: 475–85.
Marsh, D. and Smith, M. (2000) ‘Understanding Policy Networks: Towards a Dialect-
ical Approach’, Political Studies, 48: 4–21.
Marsh, D. and Smith, M. (2001) ‘There Is More than One Way to Do Political Science:
On Different Ways to Study Policy Networks’, Political Studies, 49: 528–41.
Marsh, D., Richards, D., and Smith, M. J. (2003) ‘Unequal Plurality: Towards an Asym-
metric Power Model of British Politics’, Government and Opposition, 38: 306–32.
Mossberger, K. (2007) ‘The Many Meanings of Governance: How Should We Develop
Research and Theory?’ Workshop on ‘A Global Look at Urban and Regional
Governance: The State-Market-Civic Nexus’, Halle Program on Governance,
Emory University, Atlanta, GA, 18–19 January 2007.
Newman, J. (2005) ‘Introduction’. In J. Newman (ed.), Remaking Governance. Bristol:
The Policy Press, pp. 1–15.
Offe, C. (2009) ‘Governance: An “Empty Signifier”?’ Constellations 16: 550–62.
Peters, B. G. and Pierre, J. (1998) ‘Governing without Government: Rethinking Public
Administration’, Journal of Public Administration and Theory, 8: 223–42.
Peterson, J. (1995a) ‘Decision-making in the European Union: Towards a Framework
for Analysis’, Journal of European Public Policy, 2: 69–93.
Peterson, J. (1995b) ‘Policy Networks and European Policy Making: A Reply to
Kassim’, West European Politics, 18: 389–407.
Peterson, J. (2003) ‘Policy Networks’. In A. Wiener and T. Diez (eds), European
Integration Theory. Oxford: Oxford University Press, pp. 105–24.
Pierre, J. (ed.) (2000) Debating Governance. Oxford: Oxford University Press.
Pierre, J. and Peters, B. G. (2000) Governance, Politics and the State. Houndmills,
Basingstoke: Macmillan.
Saward, M. (1996) ‘In Search of the Hollow Crown’. In P. Weller, H. Bakviss, and
R. A. W. Rhodes (eds), The Hollow Crown. London: Macmillan, pp. 16–36.
Scharpf, F. W. (1994) ‘Games Real Actors Could Play: Positive and Negative Coord-
ination in Embedded Negotiations’, Journal of Theoretical Politics, 6: 27–53.
Appendix 227
Scharpf, F. W. (1997) Games Real Actors Play. Actor-Centred Institutionalism in Policy
Research. Boulder, CO: Westview Press.
Sharpe, L. J. (1985) ‘Central Co-ordination and the Policy Network’, Political Studies,
33: 361–81.
Skelcher, C. (2000) ‘Changing Images of the State: Overloaded, Hollowed Out, Congested’,
Public Policy and Administration, 15: 3–19.
Smith, M. J. (2008) ‘Re-centring British Government: Beliefs, Traditions and Dilem-
mas in Political Science’, Political Studies Review, 6: 143–54.
Stoker, G. (2011) ‘Was Local Governance Such a Good Idea? A Global Comparative
Perspective’, Public Administration, 89: 5–31.
Taylor, A. (1997) ‘ “Arm’s Length but Hands On”. Mapping the New Governance. The
Department of National Heritage and Cultural Policies in Britain’, Public Adminis-
tration, 75: 441–66.
Taylor, A. (2000) ‘Hollowing Out or Filling In? Task Forces and the Management of
Cross-Cutting Issues in British Government’, British Journal of Politics and Inter-
national Relations, 2: 46–71.
Thatcher, M. (1998) ‘The Development of Policy Network Analyses’, Journal of
Theoretical Politics, 10: 389–416.
Torfing, J., Peters, B. G., Pierre, J., and Sørensen, E. (2012) Interactive Governance:
Advancing the Paradigm. Oxford: Oxford University Press.
Turnbull, N. (ed.) (2016) Interpreting Governance, High Politics and Public Policy:
Essays Commemorating Interpreting British Governance. Routledge Studies in
Governance and Public Policy. Abingdon, Oxon: Routledge.
Walters, W. (2004) ‘Some Critical Notes on “Governance” ’, Studies in Political
Economy, 73 (Spring): 27–46.
Weller, P. (ed.) (2011) ‘From Local Government to Narratives: Essays in Honour of
R. A. W. Rhodes, Editor, Public Administration, 1986 to 2010’. Special issue of Public
Replies
Bevir, M. (2007) ‘New Labour in Time’, Parliamentary Affairs, 60: 332–40.
Bevir, M. and Richards, D. (2009a) ‘Decentring Policy Networks: A Theoretical
Agenda’, Public Administration, 87: 3–14.
Bevir, M. and Richards, D. (2009b) ‘Decentring Policy Networks: Lessons and Pros-
pects’, Public Administration, 87: 132–41.
Rhodes, R. A. W. (1986) ‘ “Power-dependence” Theories of Central–Local Relations:
A Critical Assessment’. In M. J. Goldsmith (ed.), New Research in Central–Local
Relations. Aldershot: Gower, pp. 1–33.
Rhodes, R. A. W. (2007) ‘Understanding Governance: Ten Years On’, Organization
Studies, 28: 1243–64.
Rhodes, R. A. W. (2011) ‘Thinking On: A Career in Public Administration’, Public
Rhodes, R. A. W. and Marsh, D. (1994) ‘Policy Networks: “Defensive Comments”,
Modest Claims and Plausible Research Strategies’. In P. Dunleavy and J. Stanyer
(eds), Contemporary Political Studies 1994. Paper to the PSA Annual Conference,
University of Swansea, 29–31 March.
References
6, Perri (1997) Holistic Governance. London: Demos.

6, Perri (2010) ‘When Forethought and Outturn Part: Types of Unanticipated
and Unintended Consequences’. In Helen Margetts, Perri 6, and Christopher
Hood (eds), Paradoxes of Modernization. Unintended Consequences of Public Policy
Reform. Oxford: Oxford University Press, pp. 44–60.
6, Perri, Leat, D., Seltzer, K., and Stoker, G. (2002) Towards Holistic Governance. The
New Reform Agenda. Houndmills, Basingstoke: Palgrave.
Abélès, M. (2004) Identity and Borders: An Anthropological Approach to EU Institu-
tions. University of Wisconsin, Milwaukee: Twenty-First Century Papers: On-line
Working Papers from the Center for 21st Century Studies. Available at http://
www4.uwm.edu/c21/pdfs/workingpapers/abeles.pdf. Last accessed 6 October 2015.
Agranoff, R. (1990) Frameworks for Comparative Analysis of Intergovernmental
Relations. Bloomington, Indiana: Indiana University, School of Public and Envir-
onmental Affairs, Occasional Paper No. 26, August.
Agranoff, R. (2003) Leveraging Networks. Arlington, VA: IBM Endowment for the
Business of Government.
Agranoff, R. (2007) Managing within Networks: Adding Value to Public Organizations.
Washington, DC: Georgetown University Press.
Alderman, R. K. and Carter, N. (1991) ‘A Very Tory Coup: The Ousting of
Mrs Thatcher’, Parliamentary Affairs, 44: 125–39.
Aldrich, H. (1979) Organisations and Environments. Englewood Cliffs, NJ: Prentice
Hall.
Alexander, E. R. (1995) How Organisations Act Together. Amsterdam: OPA/Gordan
and Breach.
Alford, R. R. and Friedland. R. (1985) Powers of Theory: Capitalism, the State, and
Democracy. Cambridge: Cambridge University Press.
Allison, G. (1971) The Essence of Decision: Explaining the Cuban Missile Crisis. Boston:
Little, Brown.
Anderson, J. J. (1990) ‘Sceptical Reflections on a Europe of Regions: Britain, Germany
and the ERDF’, Journal of Public Policy, 10: 417–47.
Anderson, J. J. (1996) ‘Germany and the Structural Funds: Unification leads to
Bifurcation’ In L. Hooghe (ed.), Cohesion Policy and European Integration: Building
Multi-Level Governance. Oxford: Oxford University Press, pp. 163–94.
Anderson, W. (1960) Intergovernmental Relations in Review. Minneapolis: University
of Minnesota Press.
Andeweg, R. (1997) ‘Collegiality and Collectivity: Cabinets, Cabinet Committees and
Cabinet Ministers’. In P. Weller, K. Bakviss, and R. A. W. Rhodes (eds), The Hollow
Crown. Houndmills, Basingstoke: Macmillan, pp. 58–83.
Ansell, C. (2000) ‘The Networked Polity: Regional Developments in Western Europe’,
Governance, 13: 303–33.
230 References
Ansell, C. and Gash, A. (2008) ‘Collaborative Governance in Theory and Practice’,
Journal of Public Administration Theory and Practice, 18: 543–71.
Ansell, C. and Torfing, J. (2016) ‘Epilogue: the Current Status and Future Develop-
ment of Governance Theories’. In C. Ansell and J. Torfing (eds), Handbook on
Theories of Governance. Cheltenham: Edward Elgar, pp. 551–9.
Atkinson, M. M. and Coleman, W. D. (1989) ‘Strong States and Weak States: Sectoral
Policy Networks in Advanced Capitalist Economies’, British Journal of Political
Science, 19: 47–67.
Atkinson, M. M. and Coleman, W. D. (1992) ‘Policy Networks, Policy Communities
and the Problem of Governance’, Governance, 5: 154–80.
Bache, I. (2008) Europeanization and Multi-level Governance: Cohesion Policy in the
European Union and Britain. Oxford and New York: Rowman & Littlefield.
Bache, I. and Flinders, M. (eds) (2004) Multi-level Governance. Oxford: Oxford
University Press.
Bache, I. and Flinders, M. (eds) (2015) Multi-level Governance: Essential Readings.
Cheltenham: Edward Elgar.
Bache, I., Bulmer, S., George S., and Parker, O. (2014) Politics in the European Union.
Fourth edition. Oxford: Oxford University Press.
Bale, T. (2014) ‘Inside the Tory Mind’, Progress, 3 February 2014. Available at: http://
www.progressonline.org.uk/2014/02/03/inside-the-tory-mind-2/. Last accessed 8
July 2015.
Balme, R. and Jouve, B. (1996) ‘Building the Regional State: Europe and Territorial Organ-
isation in France’. In L. Hooghe (ed.), Cohesion Policy and European Integration: Building
Multi-Level Governance. Oxford: Oxford University Press, pp. 219–55.
Bang, H. (2005) ‘Among Everyday Makers and Expert Citizens’. In J. Newman (ed.),
Remaking Governance. Bristol: The Policy Press, pp. 159–79.
Bang, H. and Sørensen, E. (1998) The Everyday Maker: A New Challenge to Democratic
Governance. Aalborg Department of Economics, Politics and Public Administra-
tion. Working Paper, ISSN 1396–3503.
Bang, Henrik P. and Sørensen, Eva (1999) ‘The Everyday Maker: A New Challenge to
Democratic Governance’, Administrative Theory and Praxis, 21: 325–42.
Bang, Henrik Paul and Sørensen, Eva (2001) ‘The Everyday Maker: Building Political
Rather than Social Capital’. In P. Dekker and E. M. Uslaner (eds), Social Capital and
Participation in Everyday Life. London: Routledge, pp. 148–61.
Barber, B. (1983) The Logic and Limits of Trust. New Brunswick, NJ: Rutgers Univer-
sity Press.
Barber, J. (1972) The Presidential Character. Englewood Cliffs, NJ: Prentice Hall.
Bardach, E. (1998) Getting Agencies to Work Together: The Theory and Practice of
Managerial Craftsmanship. Washington DC: Brookings Institution Press.
Barnes, J. A. (1954) ‘Class and Committees in a Norwegian Island Parish’, Human
Relations, February: 39–58.
Barnes, J. A. (1972) Social Networks. Reading, MA: Addison, Wesley.
Barnes, J. (1989) Roundtable Symposium on ‘The British Cabinet in the 1980s’, Annual
Conference of the Political Studies Association, 5 April, University of Warwick.
Barnett, M. (1969) The Politics of Legislation: The Rent Act 1957. London: Weidenfeld &
Nicolson.
References 231
Barrett, S. and Fudge, C. (eds) (1981) Policy and Action. London: Methuen.
Barthes, R. (1977) ‘Death of the Author’. In his Image-Music-Text. New York: Hill &
Wang, pp. 142–8.
Barthes, R. (1993) A Roland Barthes Reader. Edited with an introduction by Susan
Sontag. London: Vintage.
Barzelay, M. with Armajani, B. J. (1992) Breaking through Bureaucracy: A New Vision
for Managing Government. Berkeley: University of California Press.
Bate, S. P. (1997) ‘Whatever Happened to Organizational Anthropology? A Review of
the Field of Organizational Ethnography and Anthropological Studies’, Human
Relations, 50: 1147–75.
Batley, R. and Stoker, G. (eds) (1991) Local Government in Europe. London: Macmillan.
Battista, R. (1994) Servicing a Stigma: The Implementation of AIDS Policy. Unpub-
lished MA thesis, University of York.
Baumgarten, F. R. and Leech, B. L. (1998) Basic Interests. The Importance of Groups in
Politics and in Political Science. Princeton, NJ: Princeton University Press.
Beer, S. H. (1973) ‘The Modernisation of American Federalism’, Publius, 3 (2): 49–95.
Beer, S. H. (1976) ‘The Adoption of General Revenue Sharing: A Case Study in Public
Sector Politics’, Public Policy, 27: 127–95.
Beer, S. H. (1978) ‘Federalism, Nationalism and Democracy in America’, American
Political Science Review, 72: 9–21.
Behn, R. D. (2001) Rethinking Democratic Accountability. Washington, DC: The
Brookings Institution.
Bell, S. and Hindmoor, A. (2009) Rethinking Governance: The Centrality of the State in
Modern Society. Port Melbourne, Vic.: Cambridge University Press.
Bellamy, C. (2011) ‘The Whitehall Programme and After: Researching Government in
Time of Governance’, Public Administration, 89: 78–92.
Benn, Tony (1980) ‘The Case for a Constitutional Premiership’, Parliamentary Affairs,
3: 7–22.
Benn, Tony (1981) Arguments for Democracy. London: Cape.
Bennington, J. and Harvey, J. (1994) ‘Spheres or Tiers? The Significance of Trans-
national Local Authority Networks’. In P. Dunleavy and J. Stanyer (eds), Contem-
porary Political Studies 1994. PSA Annual Conference, University of Swansea, 29–31
March, pp. 943–61.
Bennister, M. (2007) ‘Tony Blair and John Howard: Comparative Predominance and
Institution Stretch in the UK and Australia’, British Journal of Politics and Inter-
national Relations, 9: 327–45.
Bennister, M. (2012) Prime Ministers in Power: Blair and Howard Compared. Hound-
mills, Basingstoke: Palgrave.
Bennister, M. and Heffernan, R. (2011) ‘Cameron as Prime Minister: The Intra-
Executive Politics of Britain’s Coalition Government’, Parliamentary Affairs, 65:
778–801.
Benson, J. K. (1975) ‘The Inter-organisational Network as a Political Economy’,
Administrative Science Quarterly, 20: 229–49.
Benson, J. K. (1982) ‘A Framework for Policy Analysis’. In D. Rogers, D. Whitten, and
Associates (eds), Inter-organisational Coordination. Ames: Iowa State University
Press, pp. 137–76.
232 References
Berger, P. and Luckman, T. (1971) The Social Construction of Reality: A Treatise in the
Sociology of Knowledge. Harmondsworth: Penguin.
Berman, M. (1983) All That Is Solid Melts into Air. London: Verso.
Berrill, K. (1985) ‘Strength at the Centre: The Case for a Prime Minister’s Department’.
In A. King (ed.) The British Prime Minister. Second edition. London: Macmillan,
pp. 242–57.
Berrington, H. (1974) ‘Review Article—The Fiery Chariot: British Prime Ministers and
the Search for Love’, British Journal of Political Science, 4: 345–69.
Berry, J. M. (1997) The Interest Group Society. Third edition. New York:
HarperCollins.
Bevir, M. (1999) The Logic of the History of Ideas. Cambridge: Cambridge University
Press.
Bevir, M. (2000) ‘Historical Explanation, Folk Psychology, and Narrative’, Philosoph-
ical Explorations, 3: 152–68.
Bevir, M. (2001) ‘Prisoners of Professionalism: On the Construction and Responsibil-
ity of Political Studies’, Public Administration, 79: 469–89.
Bevir, M. (2005) New Labour: A Critique. Abingdon, Oxon: Routledge.
Bevir, M. (2006) ‘How Narratives Explain’. In D. Yanow and P. Schwartz-Shea (eds),
Interpretation and Method: Empirical Research Methods and the Interpretive Turn.
Armonk, NY: M.E. Sharpe, pp. 281–90.
Bevir, M. (2011) ‘Public Administration as Storytelling’, Public Administration, 89:
183–95.
Bevir, M. and Rhodes R. A. W. (2003) Interpreting British Governance. London:
Routledge.
Bevir, M. and Rhodes, R. A. W. (2006a) Governance Stories. Abingdon, Oxon:
Routledge.
Bevir, M. and Rhodes, R. A. W. (2006b) ‘Prime Ministers, Presidentialism and
Westminster Smokescreens’, Political Studies, 54: 671–90.
Bevir, M. and Rhodes, R. A. W. (2010) The State as Cultural Practice. Oxford: Oxford
University Press.
Bevir, M. and Rhodes, R. A. W. (2011) ‘The Stateless State’. In Mark Bevir (ed.), The
Sage Handbook of Governance. London: Sage, pp. 203–17.
Bevir, M. and Rhodes, R. A. W. (2012) ‘Interpretivism and the Analysis of Traditions
and Practices: A Reply to Wagenaar’, Critical Policy Studies, 6: 201–8.
Bevir, M. and Rhodes, R. A. W. (eds) (2016) Rethinking Governance: Ruling, Ration-
alities and Resistance. Abingdon, Oxon: Routledge.
Bevir, M., Rhodes, R. A. W., and Weller, P. (2003a) ‘Comparative Governance:
Prospects and Lessons’, Public Administration, 81: 191–210.
Bevir, M., Rhodes, R. A. W., and Weller, P. (eds) (2003b) Special Issue on Traditions of
Governance: History and Diversity. Public Administration, 81 (1): 1–210.
Birch, A. H. (1964) Representative and Responsible Government. London: Allen &
Unwin.
Blackstone, T. and Plowden, W. (1988) Inside the Think Tank: Advising the Cabinet
1971–1983. London: Heinemann.
Blair, Tony (1999) Speech by the Prime Minister, Rt. Hon Tony Blair, MP, at the
British Venture Capital Association Conference at the Intercontinental Hotel,
References 233
London, on Tuesday, 6 July. Available at: http://www.parliament.uk/business/
publications / business-papers/commons /deposited-papers / ?max=10&y=1999&sea
rch_term=Prime+Minister&itemId=120543.
Blau, P. M. (1964) Exchange and Power in Social Life. New York: Wiley.
Blick, A. (2004) People Who Live in the Dark: The Special Adviser in British Politics.
London: Politico’s.
Blick, A. and Jones, G. (2010) Premiership. Exeter: Imprint Academic.
Blyth, M. (2013) Austerity: The History of a Dangerous Idea. Oxford: Oxford Univer-
sity Press.
Bobrow, D. B. and Dryzek, J. S. (1987) Policy Analysis by Design. Pittsburgh, PA:
Pittsburgh University Press.
Bogason, P. (1995) ‘The Fragmented Locality’. In P. Bogason (ed.), New Modes of Local
Political Organizing. Local Government Fragmentation in Scandinavia. New York:
NOVA Science Publishers, pp. 169–89.
Bogdanor, Vernon (ed.) (2005) Joined-Up Government. Oxford: Oxford University
Press for the British Academy.
Boje, D. (1991) ‘The Storytelling Organization: A Story of Story Performance in an
Office-Supply Firm’, Administrative Science Quarterly, 36: 106–26.
Börzel, T. A. (1998) ‘Organizing Babylon: On the Different Conceptions of Policy
Networks’, Public Administration, 76: 253–73.
Börzel, T. A. (2011) ‘Networks: Reified Metaphor or Governance Panacea?’ Public
Boudon, R. (1993) ‘Towards a Synthetic Theory of Rationality’, International Studies
in the Philosophy of Science, 7: 5–19.
Bovens, M. (1990) ‘The Social Steering of Complex Organisations’, British Journal of
Political Science, 20: 91–117.
Bovens, M. (1998) The Quest for Responsibility: Accountability and Citizenship in
Complex Organizations. Cambridge: Cambridge University Press.
Bowman, A., Ertürk, I., Folkman, P., Froud, J., Haslam, C., Johal, S., Leaver, A., Moran, M.,
Tsitsianis, N., and Williams, K. (2015) What a Waste: Outsourcing and How It Goes
Wrong. Manchester: Manchester University Press.
Bradbury, J. (2016) ‘The Governance of Social Care for the Elderly in England’. In
M. Bevir and R. A. W. Rhodes (eds), Rethinking Governance: Ruling Rationalities
and Resistance. Abingdon, Oxon: Routledge, pp. 105–22.
Braithwaite, J. (2003) ‘On Speaking Softly and Carrying Big Sticks: Neglected Dimensions
of a Republican Separation of Powers’, University of Toronto Law Review, 47: 305–61.
Brown, R. G. S. (1971) ‘Public Administration and the Study of Administrative
Organizations’, PAC Bulletin, 11, December: 16–27.
Bruce-Gardyne, J. and Lawson, N. (1976) The Power Game. London: Macmillan.
Bulmer, S. (1983) ‘Domestic Politics and European Community Policy Making’,
Journal of Common Market Studies, 21: 349–63.
Bulmer, S. (1994) ‘Institutions, Governance Regimes and the Single European Market:
Analysing the Governance of the European Union’. Conference of Europeanists,
Chicago, 31 March–2 April.
Burch, M. (1983) ‘Mrs Thatcher’s Approach to Leadership in Government,
1979–1983’, Parliamentary Affairs, 36: 399–416.
234 References
Burch, M. (1989) Contribution to a Roundtable Symposium on ‘The British Cabinet in
the 1980s’, Annual Conference of the Political Studies Association, 5 April, Univer-
sity of Warwick.
Burch, M. and Holliday, I. (1996) The British Cabinet System, Hemel Hempstead,
Herts: Prentice Hall/Wheatsheaf.
Burch, M. and Holliday, I. (2004) ‘The Blair Government and the Core Executive’,
Government and Opposition, 39: 1–21.
Butler, D., Adonis, A., and Travers, T. (1994) Failure in British Government: The
Politics of the Poll Tax. Oxford: Oxford University Press.
Butler, Sir Robin (1992) ‘The New Public Management: The Contribution of Whitehall
and Academia’, Public Policy and Administration, 7 (3): 4–14.
Butler, Sir Robin (1993) ‘The Evolution of the Civil Service—a Progress Report’, Public
Cabinet Office (2000) Wiring It Up. London: Cabinet Office.
Cabinet Office (2012) Open Public Service Report 2012. London: Cabinet Office.
Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/255417/Open_Public_Services_2012_full_report.pdf. Last accessed 14 July
2015.
Cadbury Report (1992) The Report of the Committee on the Financial Aspects of
Corporate Governance. London: Gee & Co. Ltd.
Callon, M., Law, J. and Rip, A. (eds) (1986) Mapping the Dynamics of Science and
Technology. London: Macmillan Press.
Carpenter, D. P., Esterling, K. M., and Lazer, D. M. J. (2004) ‘Friends, Brokers, and
Transitivity: Who Informs Whom in Washington Politics?’ The Journal of Politics,
66: 224–46.
Cartwright, D. and Zander, A. (eds) (1968) Group Dynamics. Third edition. New York:
Harper.
Castells, M. (2000) ‘Materials for an Exploratory Theory of the Network Society’,
British Journal of Sociology, 51: 5–24.
Castle, B. (1973) ‘Mandarin Power’, The Sunday Times, 10 June.
Cawson, A. (1978) ‘Pluralism, Corporatism and the Role of the State’, Government and
Opposition, 13: 178–98.
Cawson, A. (1982) Corporatism and Welfare. London: Heinemann.
Cawson, A. (1986a) Corporatism and Political Theory. Oxford: Blackwell.
Cawson, A. (ed.) (1986b) Organised Interests and the State. London: Sage.
Checkel, J. T. (2007) ‘Constructivism and EU Politics’. In K. E. Jørgensen, M. Pollack,
and B. Rosamond (eds), The Sage Handbook of European Union Politics. London:
Sage, pp. 57–76.
Chester, Sir Norman (1975) The Nationalisation of British Industry, 1945–51. London:
HMSO.
Chester, Sir Norman (1981) The English Administrative System, 1780–1870. Oxford:
Clarendon Press.
Child, J. and Faulkner, D. (1998) Strategies of Co-operation. Managing Alliances,
Networks and Joint Ventures. Oxford: Oxford University Press.
Chisholm, D. (1989) Coordination without Hierarchy: Informal Structures in Multi-
organizational Systems. Berkeley: University of California Press.
References 235
Christiansen, T., Jorgensen, K. E., and Wiener, A. (eds) (2001) The Social Construction
of Europe. London: Sage.
CIPFA (The Chartered Institute of Public Finance and Accountancy) (1994) Corpor-
ate Governance in the Public Services. London: CIPFA.
Clarke, M. and Stewart, J. (1991) Choices for Local Government for the 1990s and
Beyond. Longman: Harlow.
Clifton, J. (2014) ‘Beyond Hollowing Out: Straitjacketing the State’, The Political
Quarterly, 85: 437–44.
Cm 2211 (1993) Department of National Heritage, Annual Report 1993. London: HMSO.
Cm 2430 (1993) Next Steps Review 1993. London: HMSO.
Cm 2511 (1994) Department of National Heritage, Annual Report 1994. London:
HMSO.
Cm 2811 (1995) Department of National Heritage, Annual Report 1995. London:
HMSO.
Cm 3270 (1996) Second Report of the Committee on Standards in Public Life, Volume 1,
Report, Local Public Spending Bodies. London: HMSO.
Cm 4310 (1999) Modernization Government. London: HMSO.
Coates, D. (1984) ‘Food Law: Brussels, Whitehall and Town Hall’. In D. Lewis and
H. Wallace (eds), Policies into Practice. National and International Studies in
Implementation. London: Heinemann, pp. 144–60.
Cockerell, M. (1988) Live From Number 10. London: Faber.
Cockerell, M., Hennessy, P. and Walker, D. (1985) Sources Close to the Prime Minister:
Inside the Hidden World of the News Manipulators. London: Macmillan.
Colebatch, H. K. (2009) ‘Governance as a Conceptual Development in the Analysis of
Policy’, Critical Policy Studies, 3: 58–67.
Coleman, W. D. and Skogstad, G. (1990) Policy Communities and Public Policy in
Canada. Toronto: Copp Clark Pitman.
Collingwood, R. G. (1965) Essays in the Philosophy of History. Austin, Texas: Univer-
sity of Texas Press.
Collingwood, R. G. (1978) [1939] An Autobiography. Oxford: Oxford University Press.
Collingwood, R. G. (1993) [1946] The Idea of History. Revised edition. Oxford: Oxford
University Press.
Committee of Public Accounts (1994) The Proper Conduct of Public Business, HC 154
Session 1993–94, London: HMSO.
Considine, M. (1994) Public Policy: A Critical Approach. Melbourne: Macmillan.
Considine, M. (2002) ‘The End of the Line? Accountable Governance in the Age of
Networks, Partnerships, and Joined-Up Services’, Governance, 15: 21–40.
Considine, M. and Lewis, J. (1999) ‘Governance at Ground Level: The Frontline
Bureaucrat in the Age of Markets and Networks’, Public Administration Review,
59: 467–80.
Craig, G. A. and George, A. L. (1983) Force and Statecraft, Diplomatic Problems of Our
Time. Oxford: Oxford University Press.
Cram, L. (2009) ‘From Integration by Stealth to “Good Governance”. EU Social Policy
in Historical Perspective’. In A. Verdun and I. Toemme (eds), Innovative Govern-
ance in the European Union: The Politics of Multilevel Policy Making. Boulder, CO:
Lynne Reiner, pp. 87–99.
236 References
Cribb, J., Disney, R., and Sibieta, L. (2014) The Public Sector Workforce: Past, Present
and Future. IFS Briefing Note BN145. London: Institute for Fiscal Studies. Available
at: file:///C:/RAWR/Continuing%20research/Selected%20essays/Reading/Cribb%
20et%20al%202014%20Public%20Employment.pdf. Last accessed 8 October 2015.
Crosbie, P. V. (ed.) (1975) Interaction in Small Groups. London: Macmillan.
Crossman, R. H. S. (1963) ‘Introduction’ to W. Bagehot, The English Constitution.
London: Fontana.
Crossman, R. H. S. (1975) The Diaries of a Cabinet Minister, Volume 1, Minister of
Housing. London: Cape and Hamilton.
Crozier, M. (1964) The Bureaucratic Phenomenon. Chicago: Chicago University Press.
Crozier, M. and Friedberg, E. (1980) Actors and Systems. Chicago: University of
Chicago Press.
Crozier, M. and Thoenig, J–C. (1976) ‘The Regulation of Complex Organised Systems’,
Administrative Science Quarterly, 21: 547–70.
Czarniawska, B. (1998) A Narrative Approach to Organization Studies. Thousand
Oaks, CA: Sage.
Czarniawska, B. (2004) Narratives in Social Science Research. London: Sage.
Davis, G. and Rhodes, R. A. W. (2000) ‘From Hierarchy to Contracts and Back Again:
Reforming the Australian Public Service’. In M. Keating, J. Wanna, and P. Weller (eds),
Institutions on the Edge? Capacity for Governance. Sydney: Allen & Unwin, pp. 74–98.
Davis, G., Wanna, J., Warhurst, J., and Weller, P. (1993) Public Policy in Australia.
Second edition. St Leonards, NSW: Allen & Unwin.
de Callières, F. (1963) [1716] On the Manner of Negotiating with Princes. Washington:
University of America Press.
De Rynck, S. (1996) ‘Europe and Cohesion Policy Making in the Flemish Region’. In
L. Hooghe (ed.), Cohesion Policy and European Integration: Building Multi-Level
Governance. Oxford: Oxford University Press, pp. 129–62.
Dearlove, J. and Saunders, P. (1984) Introduction to British Politics. Cambridge: Polity
Press.
Denhardt, J. V. and Denhardt, R. B. (2000) ‘The New Public Service: Serving Rather
than Steering’, Public Administration Review, 60: 549–59.
Denning, S. (2004) ‘Telling Tales’, Harvard Business Review, 8 (5): 1–8.
Denning, S. (2007) The Secret Language of Leadership: How Leaders Inspire Action
through Narrative. San Francisco: Jossey–Bass.
Diamond, P. (2014) Governing Britain: Power, Politics and the Prime Minister.
London: I. B. Taurus.
Dietz, K. and Silverman L. L. (2013) Business Storytelling for Dummies. New York:
Wiley.
Diez, T. (2015) ‘European Politics’. In M. Bevir and R. A. W. Rhodes (eds), The Routledge
Handbook of Interpretive Political Science. Abingdon, Oxon: Routledge, pp. 268–81.
Donoughue, B. (1987) Prime Minister: the Conduct of Policy under Harold Wilson and
James Callaghan. London: Cape.
Dorey, P. (2005) Policy Making in Britain: An Introduction. London: Sage.
Dowding, K. (1994) ‘Policy Networks: Don’t Stretch a Good Idea Too Far’. In
P. Dunleavy and J. Stanyer (eds), Contemporary Political Studies 1994. PSA Annual
Conference, University of Swansea, 29–31 March, pp. 59–78.
References 237
Dowding, K. (1995) ‘Model or Metaphor? A Critical Review of the Policy Network
Approach’, Political Studies, 43: 136–58.
Dowding, K. (2001) ‘There Must Be an End to Confusion’, Political Studies, 49:
89–105.
Dowding, K. (2013) ‘The Prime Ministerialization of the British Prime Minister’,
Parliamentary Affairs, 66: 617–35.
Dowding, K., Berlinski, S., and Torun, D. (2012) Accounting for Ministers: Scandal and
Survival in British Government, 1945–2007. Cambridge: Cambridge University Press.
Dryzek, J. (1993) ‘Policy Analysis and Planning: From Science to Argument’. In
F. Fischer and J. Forester (eds), The Argumentative Turn in Policy Analysis and
Planning. Durham: Duke University, pp. 213–30.
Dryzek, J. (2006) ‘Policy Analysis as Critique’. In M. Moran, M. Rein, and R. E. Goodin
(eds), The Oxford Handbook of Public Policy. Oxford: Oxford University Press,
pp. 190–203.
Dunleavy, P. (1980) ‘Social and Political Theory and the Issues in Central–Local
Relations’. In G. W. Jones (ed.), New Approaches to the Study of Central–Local
Relationships. Aldershot: Gower, pp. 116–36.
Dunleavy, P. (1984) ‘The Limits to Local Government’. In M. Boddy and C. Fudge
(eds), Local Socialism: Labour Councils and New Left Alternatives. London:
Macmillan, pp. 49–81.
Dunleavy, P. (1989a) ‘The Architecture of the British Central State, Part 1: Framework
for Analysis’, Public Administration, 67: 249–75.
Dunleavy, P. (1989b) ‘The Architecture of the British Central State, Part II: Empirical
Findings’, Public Administration, 67: 391–417.
Dunleavy, P. (1991) Democracy, Bureaucracy and Public Choice: Economic Approaches
in Political Science. Hemel Hempstead: Harvester/Wheatsheaf.
Dunleavy, P. (1994) ‘The Globalization of Public Services Production: Can Govern-
ment be “Best in World”?’ Public Policy and Administration, 9 (2): 36–64.
Dunleavy, P. (1995a) ‘Reinterpreting the Westland Affair: Theories of the State and
Core Executive Decision Making’. In R. A. W. Rhodes and P. Dunleavy (eds), Prime
Minister, Cabinet and Core Executive. London: Macmillan, pp. 181–218.
Dunleavy, P. (1995b) ‘Estimating the Distribution of Positional Influence in Cabinet
Committees under Major’. In R. A. W. Rhodes and P. Dunleavy (eds), Prime
Minister, Cabinet and Core Executive. London: Macmillan, pp. 298–321.
Dunleavy, P. and O’Leary, B. (1987) Theories of the State. London: Macmillan.
Dunleavy, P. and Rhodes, R. A. W. (1990) ‘Core Executive Studies in Britain’, Public
Dunleavy, P., Jones, G. W., and O’Leary, B. (1990) ‘Prime Ministers and the Com-
mons: Patterns of Behaviour, 1979–87’, Public Administration, 68: 123–40.
Dunleavy, P. and Jones, G. W. with Burnham, J., Elgie, R., and Fysh, P. (1995) ‘Leaders,
Politics and Institutional Change: The Decline of Prime Ministerial Accountability
to the House of Commons, 1868–1990’. In R. A. W. Rhodes and P. Dunleavy (eds),
Prime Minister, Cabinet and Core Executive. London: Macmillan, pp. 275–97.
Dunsire, A. (1986) ‘A Cybernetic View of Guidance, Control and Evaluation in the
Public Sector’. In F.-X. Kaufman, G. Majone, and V. Ostrom (eds), Guidance,
Control, and Evaluation in the Public Sector. Berlin: de Gruyter, pp. 325–46.
238 References
Dunsire, A. (1990) ‘Holistic Governance’, Public Policy and Administration, 5 (1): 4–19.
Dunsire, A. (1993) ‘Manipulating Social Tensions: An Alternative Mode of Govern-
ment Intervention’. Discussion Paper 93/7, August, Koln: Max Planck Institut fur
Gesellschaftsforschung.
Dunsire, A. (1995) ‘Administrative Theory in the 1980s: A Viewpoint’, Public Admin-
istration, 73: 17–40.
Dunsire, A. and C. Hood (1989) Cutback Management in Public Bureaucracies:
Popular Themes and Observed Outcomes in Whitehall. Cambridge: Cambridge
University Press.
Dyson, K. (1980) The State Tradition in Western Europe. Oxford: Martin Robertson.
Efficiency Unit (1988) Improving Management in Government: The Next Steps.
London: HMSO.
Elgie, R. (1997) ‘Models of Executive Politics: A Framework for the Study of Executive
Power Relations in Parliamentary and Semi-Presidential Regimes’, Political Studies,
45: 217–31.
Elgie, R. (2011) ‘Core Executive Studies Two Decades On’, Public Administration,
89: 64–77.
Elgin, D. S. and Bushnell, R. A. (1977) ‘The Limits to Complexity: Are Bureaucracies
Becoming Unmanageable?’ The Futurist, December: 327–49.
Elkin, S. L. (1975) ‘Comparative Urban Politics and Inter-organizational Behaviour’.
In K. Young (ed.), Essays on the Study of Urban Politics,. London: Macmillan,
pp. 158–84.
Emerson, R. E. (1962) ‘Power-Dependence Relations’, American Sociological Review,
27: 31–41.
Evan, W. M. (1976a) ‘An Organisation Set Model of Inter-organizational Relations’. In
W. M. Evan (ed.), Inter-organizational Relations. Harmondsworth: Penguin Books,
pp. 78–90.
Evan, W. M. (ed.) (1976b) Inter-organizational Relations Harmondsworth: Penguin
Books.
Eymeri-Douzans, Jean-Michel, Bioy, X., and Mouton, S. (eds) (2015) Le règne des
entourages: Cabinets and conseillers de l’exécutif. Paris: Presses de Sciences Po.
Fawcett, P. (2016) ‘Critical Encounters with Decentred Theory: Tradition, Metago-
vernance and Parrhēsia as Storytelling’. In N. Turnbull (ed.), Interpreting Govern-
ance, High Politics and Public Policy: Essays Commemorating Interpreting British
Governance. Routledge Studies in Governance and Public Policy. Abingdon, Oxon:
Routledge, pp. 39–56.
Fenno, R. E. (1990) Watching Politicians: Essays on Participant Observation. Berkeley:
Institute of Governmental Studies, University of California.
Ferlie, E. and Pettigrew, A. (1996) ‘Managing through Networks: Some Issues and
Implications for the NHS’, British Journal of Management, 7: 81–99.
Fernandez, R. M. and Gould, R. V. (1994) ‘A Dilemma of State Power: Brokerage and
Influence in the National Health Policy Domain’, American Journal of Sociology,
99: 1455–91.
Fine, G. A., Morrill, C., and Surianarain, S. (2009) ‘Ethnography in Organizational
Settings’. In D. Buchanan and A. Bryman (eds), Handbook of Organizational
Research Methods. London: Sage, pp. 602–19.
References 239
Finer, S. E. (1970) Comparative Government. London: Allen Lane, The Penguin Press.
Finer, S. E. (1987) ‘Thatcherism and British Political History’. In K. Minogue and
M. Biddiss (eds), Thatcherism: Personality and Politics. London: Macmillan, pp. 127–40.
Finlayson, A. (ed.) (2004) ‘The Interpretive Approach in Political Science: A Symposium’,
British Journal of Politics and International Relations, 6: 1–36.
Fischer, F. (2003) Reframing Policy Analysis. Oxford: Oxford University Press.
Fischer, F. and Forester, J. (eds) (1993) The Argumentative Turn in Policy Analysis and
Planning. Durham: Duke University Press.
Fischer, F., Torgerson, D., Durnová, A., and Orsini, M. (eds) (2015) Handbook of
Critical Policy Studies. Aldershot: Edward Elgar Publishing.
Fish, S. (1989) Doing What Comes Naturally. Oxford: Clarendon Press.
Fleming, J. and Rhodes, R. A. W. (2005) ‘Bureaucracy, Contracts and Networks: The
Unholy Trinity and the Police’, Australian and New Zealand Journal of Criminology,
38: 192–205.
Flynn, R., Williams, G., and Pickard, S. (1996) Markets and Networks: Contracting in
Community Health Services. Buckingham: Open University Press.
Foley, M. (1993) The Rise of the British Presidency. Manchester: Manchester University Press.
Foley, M. (2000) The British Presidency. Manchester: Manchester University Press.
Ford, D., Gadde, L-E., Håkansson, H., and Snehota, I. (2003) Managing Business
Relationships. Second edition. Chichester: Wiley.
Foster, C. D., Jackman, R. A., Perlman, M., and Lynch, B. (1980) Local Government
Finance in a Unitary State. London: Allen & Unwin.
Fox, A. (1974) Beyond Contract: Work, Power and Trust Relations. London: Faber.
Fox, C. J. and Miller, H. T. (1995) Postmodern Public Administration: Toward Dis-
course. London: Sage.
Frances, J., Levacic, R., Mitchell, J., and Thompson, G. (1991) ‘Introduction’. In
G. Thompson, J. Frances, R. Levacic, and J. Mitchell (eds), Markets Hierarchies
and Networks: The Co-Ordination of Social Life. London: Sage, pp. 1–19.
Frederickson, H. G. (1996) The Spirit of Public Administration. San Francisco: Jossey-Bass.
Freeman, J. L. (1955) The Political Process. New York: Doubleday.
Freeman, J. L. and Stevens, J. P. (1987) ‘A Theoretical and Conceptual Re-examination
of Sub-system Politics’, Public Policy and Administration, 2 (1): 1–25.
Friend, J. K., Power, J., and Yewlett, C. J. L. (1974) Public Planning: The Intercorporate
Dimension. London: Tavistock.
Gabriel, Y. (2000) Storytelling in Organizations: Facts, Fictions and Fantasies. Oxford:
Oxford University Press.
Galligan, B. (1995) A Federal Republic. Cambridge: Cambridge University Press.
Gamble, A. (1990) ‘Theories of British Politics’, Political Studies, 38: 404–20.
Gardner, A. and Lowndes, V. (2016) ‘Negotiating Austerity and Local Traditions’. In
Bevir, M. and Rhodes, R. A. W (eds.) Rethinking Governance: Ruling Rationalities
and Resistance. Abingdon, Oxon: Routledge, pp. 125–43.
Gash, T., Panchamia, N., Sims, S., and Hotson, L. (2013) Making Public Service
Markets Work. London: Institute for Government.
Geertz, C. (1973) The Interpretation of Cultures. New York: Basic Books.
Gergen, K. J. (1986) ‘Correspondence versus Autonomy in the Language of Under-
standing Human Action’. In D. W. Fiske and R. A. Shweder (eds), Metatheory in
240 References
Social Science: Pluralisms and Subjectivities. Chicago: Chicago University Press,
pp. 136–62.
Giddens, A. (1990) The Consequences of Modernity. Cambridge: Polity Press.
Goldsmith, M. J. (ed.) (1986) New Research in Central–Local Relations. Aldershot:
Gower.
Goldsmith, M. J. and Rhodes, R. A. W. (1986) Register of Research and Research Digest
on Central–Local Government Relations in Britain. London: ESRC.
Goldsmith, S. and Eggers, W. D. (2004) Governing by Networks. Washington, DC:
Brookings Institution Press.
Goldsmith, S. and Kettl, Donald (2009) Unlocking the Power of Networks: Keys to
High-Performance Government. Washington, DC: Brookings Institution Press.
Goodsell, C. T. (1985) The Case for Bureaucracy. Second edition. Chatham, NJ:
Chatham House.
Goodsell, C. T. (1994) The Case for Bureaucracy. Third edition. Chatham, NJ: Chatham
House.
Goodsell, C. T. (2015) The New Case for Bureaucracy. London: Sage.
Goss, S. (2001) Making Local Government Work. Houndmills, Basingstoke: Palgrave.
Granovetter, M. (1985) ‘Economic Action and Social Structure: The Problem of
Embeddedness’, American Journal of Sociology, 91: 481–51.
Grant, W. P., Paterson, W., and Whitson, C. (1988) Government and the Chemical
Industry: A Comparative Study of Britain and West Germany. Oxford: Clarendon
Press.
Gray, A. and Jenkins, W. I. et al. (1991) ‘The Management of Change in Whitehall: The
Experience of FMI’, Public Administration, 69: 41–59.
Gray, C. (1982) ‘The Regional Water Authorities’. In B. Hogwood and M. Keating
(eds), Regional Government in England. Oxford: The Clarendon Press, pp. 143–67.
Greenleaf, W. H. (1983) The British Political Tradition. Volume 1: The Rise of
Collectivism. London: Methuen.
Greenstein, F. I. (1975) Personality and Politics. New York: Norton.
Gretschmann, K. (1986) ‘Solidarity and Markets’. In F. X. Kaufman, G. Majone, and
V. Ostrom (eds) (1986) Guidance, Control, and Evaluation in the Public Sector.
Berlin: de Gruyter, pp. 47–67.
Gyford, J. and James, M. (1983) National Parties and Local Politics. London: Allen &
Unwin.
Haas, P. M. (1992) ‘Epistemic Communities and International Policy Coordination’,
International Organisation, 46: 1–35.
Haines, J. (1977) The Politics of Power. London: Cape.
Hajer, M. A. (2003) ‘Policy without a Polity? Policy Analysis and the Institutional
Void’, Policy Sciences, 36: 175–95.
Hajer, M. A. (2009) Authoritative Governance. Oxford: Oxford University Press.
Hajer, M. A. and Wagenaar, H. (eds) (2003) Deliberative Policy Analysis. Under-
standing Governance in a Networked Society. Cambridge: Cambridge University
Press.
Ham, C. (1981) Policy Making in the National Health Service. London: Macmillan.
Hammersley, M. (1990) Reading Ethnographic Research. A Critical Guide. Harlow,
Essex: Longman.
References 241
Hammersley, M. and Atkinson, P. (2007) [1983] Ethnography: Principles in Practice.
Third edition. London: Routledge.
Hancher, L. and Moran, M. (1989) ‘Organising Regulatory Space’. In L. Hancher and
M. Moran (eds), Capitalism, Culture and Economic Regulation. Oxford: Clarendon
Press, pp. 271–95.
Handy, C. B. (1985) Understanding Organisations, Third edition. Harmondsworth:
Penguin Books.
Hanf, K. (1978) ‘Introduction’. In K. Hanf and F. W. Scharpf (eds), Inter-
organisational Policy Making. London: Sage, pp. 1–15.
Hanf, K. and Scharpf, F. F. (eds) (1978) Inter-organizational Policy Making. London: Sage.
Harrison, S. (1994) ‘The Wrong Kind of Chaos? The Early Days of an NHS Trust’,
Public Money and Management, 14 (1): 1–8.
Harrison, S. and Pollitt, C. (1994) Controlling Health Professionals. Buckingham: Open
University Press.
Hay, C. (2002) Political Analysis. Houndmills, Basingstoke: Palgrave.
Hay, C. and Richards, D. (2000) ‘The Tangled Webs of Westminster and Whitehall’,
Public Administration, 78: 1–28.
Hayward, J., Barry, B., and Brown, B. (eds) (1999) The British Study of Politics in the
Twentieth Century. Oxford: Oxford University Press for the British Academy.
Haywood, S. and Hunter, D. J. (1982) ‘Consultative Processes in Health Policy in the
United Kingdom: A View from the Centre’, Public Administration, 60: 143–62.
Heady, B. (1974) British Cabinet Ministers. London: Allen & Unwin.
Heclo, H. (1978) ‘Issue Networks and the Executive Establishment’. In A. King (ed.),
The New American Political System., Washington, DC: American Enterprise Inc.
pp. 87–124.
Heclo, H. and Wildavsky, A. (1977) [1974] The Private Government of Public Money.
Second edition. London: Macmillan.
Heffernan, R. (2003) ‘Prime Ministerial Predominance? Core Executive Politics in the
UK’, British Journal of Politics and International Relations, 5: 347–72.
Heffernan, R. (2005) ‘Exploring (and Explaining) the British Prime Minister’, The
British Journal of Politics and International Relations, 7: 605–20.
Heffernan, R. (2013) ‘There’s No Need for the “-isation”: The Prime Minister Is Merely
Prime Ministerial’, Parliamentary Affairs, 66: 636–45.
Heilman, S. and Stepan, M. (eds) (2016) China’s Core Executive: Leadership Styles,
Structures and Processes under Xi Jinping. Berlin: MERICS.
Heimeriks, K. H., Klijn, E., and Reuer, J. J. (2009) ‘Building Capabilities for Alliance
Portfolios’, Long Range Planning, 42: 96–114.
Heisler, M. (1979) ‘Corporate Pluralism Revisited: Where Is the Theory?’ Scandi-
navian Political Studies, 2: 277–97.
Hennessy, P. (1986) Cabinet. Oxford: Blackwell.
Hennessy, P. (1989) Whitehall. London: Secker & Warburg.
Hennessy, P. (1992) Never Again: Britain 1945–51. London: Jonathan Cape.
Hennessy, P. (1995) The Hidden Wiring. London: Gollancz.
Hennessy, P. (2000a) The Prime Ministers. London: Allen Lane, The Penguin Press.
Hennessy, P. (2000b) The Blair Revolution in Government. University of Leeds,
Institute for Politics and International Studies.
242 References
Hennessy, P. (2002) ‘The Blair Government in Historical Perspective: An Analysis of
the Power Relationships within New Labour’, History Today, 52 (1): 21–3.
Henney, A. (1984) Inside Local Government. London: Sinclair Browne.
Hesse, J. J. (1991) ‘Local Government in International Perspective’. In J. J. Hesse (ed.),
Local Government and Urban Affairs in International Perspective. Baden-Baden:
Nomos Verlagsgesellschaft., pp. 603–21.
Hirst, P. (1994) Associative Democracy. Cambridge: Polity Press.
Hoffman, S. (1966) ‘The Fate of the Nation State’, Daedalus, 95: 862–915.
Hogwood, B. W., Judge, D., and McVicar, M. (2000) ‘Agencies and Accountability’. In
R. A. W. Rhodes (ed.), Transforming British Government. Volume 1. Changing
Institutions. London: Macmillan, pp. 195–222.
Holden, H. (2010) ‘The Commission on Scottish Devolution—the Calman Commission’,
House of Commons Library Research Briefing, Standard Note: SN/PC/04744. Last
updated 4 June 2010. Available at: file://soton.ac.uk/ude/personalfiles/users/rar1C11/
mydocuments/Downloads/SN04744%20(4).pdf. Last accessed 8 January 2016.
Holliday, I. (2000) ‘Is the British State Hollowing Out?’, Political Quarterly, 71: 167–76.
Hood, C. (1983) The Tools of Government. London: Macmillan.
Hood, C. (1990) ‘Public Administration: Lost an Empire, Not Yet Found a Role?’ In
A. Leftwich (ed.), New Developments in Political Science. Aldershot: Edward Elgar,
pp. 107–25.
Hood, C. (1991) ‘A Public Management for All Seasons?’ Public Administration,
69 (1): 3–19.
Hood, C. (1999) ‘British Public Administration: Dodo, Phoenix or Chameleon?’ In
J. Hayward, B. Barry, and A. Brown (eds), The British Study of Politics in the
Twentieth Century. Oxford: Oxford University Press for the British Academy,
pp. 287–311.
Hood, C. (2011) ‘It’s Public Administration, Rod, but Maybe Not as We Know It.
British Public Administration in the 2000s’, Public Administration, 89: 128–39.
Hood, C. and Jackson, M. W. (1991) ‘The New Public Management: A Recipe for
Disaster’, Canberra Bulletin of Public Administration, 64, May: 16–24.
Hood, C. and Lodge, M. (2007) ‘Civil Service Reform Syndrome: Are We Heading for a
Cure?’ Transformation, Spring, 58–9. Available at: http://www.christopherhood.net/
pdfs/15576-T5%20PROOF2%20(1).pdf.
Hooghe, L. (ed.) (1996) Cohesion Policy and European Integration: Building Multi-
Level Governance. Oxford: Oxford University Press.
Hooghe, L. and Marks, G. (1996) ‘Europe with the Regions: Channels of Regional
Representation in the European Union’, Publius: The Journal of Federalism, 26:
73–92.
Hooghe, L. and Marks, G. (2003) ‘Unravelling the Central State, but How? Types of
Multi-Level Governance’, American Political Science Review, 97: 233–43.
Houlihan, B. (1988) Housing Policy and Central–Local Government Relations.
Aldershot: Avebury.
Hudson, J. (1984) ‘Prime Ministerial Popularity in the UK, 1960–81’, Political Studies,
32: 86–97.
Hughes, Owen (2010) ‘Does Governance Exist?’ In S. Osborne (ed.), The New Public
Governance: Emerging Perspectives on the Theory and Practice of Public Governance.
London and New York: Routledge / Taylor & Francis, pp. 87–104.
References 243
Hummel, R. P. (1991) ‘Stories Managers Tell: Why They Are as Valid as Science’,
Public Administration Review, 51 (1): 31–41.
Hunter, D. J. and Wistow, G. (1987) ‘The Paradox of Policy Diversity in a Unitary
State: Community Care in Britain’, Public Administration, 65: 3–23.
Huxham, C. and Vangen, C. (2005) Managing to Collaborate: The Theory and Practice
of Collaborative Advantage. London: Routledge.
Inglis, F. (2000) Clifford Geertz. Culture, Custom and Ethics. Oxford: Blackwell.
Iremonger, L. (1970) The Fiery Chariot: British Prime Ministers and the Search for
Love. London: Secker & Warburg.
Jachtenfuchs, M. (2001) ‘The Governance Approach to European Integration’, Journal
of Common Market Studies, 39: 245–64.
Jackson, P. M. (1985) Implementing Government Policy Initiatives. The Thatcher
Administration 1979–83, London: RIPA, pp. 11–31.
James, S. (1992) British Cabinet Government. London: Routledge.
Janis, I. (1972) Victims of Groupthink. Boston: Houghton Mifflin.
Jeffery, C. (ed.) (2015) The Regional Dimension of the European Union: Towards a
Third Level in Europe. Routledge Studies in Federalism and Decentralisation.
Abingdon, Oxon: Routledge.
Jenkins, K. (1995) On ’What is History?’ London: Routledge.
Jensen, L. (1998) ‘Cultural Theory and Democratizing Functional Domains: The Case
of Danish Housing’, Public Administration, 76: 117–39.
Jessop, B. (1990) State Theory. Cambridge: Polity Press.
Jessop, B. (1997) ‘The Governance of Complexity and the Complexity of Governance:
Preliminary Remarks on Some Problems and Limits of Economic Guidance’. In
A. Amin and J. Hausner (eds), Beyond Market and Hierarchy: Interactive Govern-
ance and Social Complexity. Cheltenham: Edward Elgar, pp. 111–47.
Jessop, B. (2000) ‘Governance Failure’. In G. Stoker (ed.), The New Politics of British
Local Governance. Houndmills, Basingstoke: Macmillan, pp. 11–32.
Jessop, B. (2003) ‘Governance and Metagovernance: On Reflexivity, Requisite Variety,
and Requisite Irony’. In H. P. Bang (ed.), Governance as Social and Political
Communication. Manchester: Manchester University Press, pp. 101–16.
Jessop, B. (2004) ‘Hollowing Out the “Nation-State” and Multi-Level Governance’. In
P. Kennett (ed.), A Handbook Of Comparative Social Policy. Cheltenham: Edward
Elgar Publishing, pp. 11–25.
Jessop, B. (2007) State Power. Cambridge: Polity Press.
Jessop, B. (1978) ‘Corporatism, Parliamentarism and Social Democracy’. In
P. C. Schmitter and G. Lehmbruch (eds), Trend Towards Corporatist Intermedi-
ation. London: Sage, pp. 185–212.
Johnson, N. (1975) ‘The Place of Institutions in the Study of Politics’, Political Studies,
23: 271–83.
Johnson, P. and Chandler, D. (2015) ‘The Coalition and the Economy’. In A. Seldon
and M. Finn (eds), The Coalition Effect, 2010–2015. Cambridge: Cambridge
Jones, G. W. (1975) ‘Development of the Cabinet’. In W. Thornhill (ed.), The Mod-
ernization of British Government. London: Pitman, pp. 31–62.
Jones, G. W. (1980) ‘The Prime Minister’s Aides’. Hull: Hull University Department of
Politics, Hull Papers in Politics No. 6.
244 References
Jones, G. W. (1983) ‘Prime Ministers’ Departments Really Create Problems:
A Rejoinder to Patrick Weller’, Public Administration, 61: 79–84.
Jones, G. W. (1985) ‘The Prime Minister’s Power’. In King, A. (ed.), The British Prime
Minister. Second edition. London: Macmillan, pp. 195–220.
Jones, G. W. (1986) ‘Preface’. In C. E. Hill, A Bibliography of the Writings of
W. A. Robson. London School of Economics and Political Science, Greater London
Paper No. 17, pp. 6–12.
Jones, G. W. (1989) ‘A Revolution in Whitehall? Changes in British Central Govern-
ment since 1979’, West European Politics, 12: 238–61.
Jones, G. W. (1995) ‘The Downfall of Margaret Thatcher’. In R. A. W. Rhodes and
Dunleavy P. (eds) (1995) Prime Minister, Cabinet and Core Executive. London:
Jordan, A. and Schout, A. (2006) The Coordination of the European Union: Exploring
the Capacities for Networked Governance. Oxford: Oxford University Press.
Jordan, A., Wurzel, R. K. W., and Zito, A. (2005) ‘The Rise of “New” Policy Instru-
ments in Comparative Perspective: Has Governance Eclipsed Government?’ Polit-
ical Studies, 53: 477–96.
Jordan, G. (1981) ‘Iron Triangles, Woolly Corporatism and Elastic Nets: Images of the
Policy Process’, Journal of Public Policy, 1: 95–123.
Jordan, G. (1984) ‘Pluralistic Corporatism and Corporate Pluralism’, Scandinavian
Political Studies, 7: 137–53.
Jordan, G. (1990) ‘Sub-Governments, Policy Communities and Networks: Refilling the
Old Bottles?’ Journal of Theoretical Politics, 2: 317–37.
Jordan, G. (1992) ‘Next Steps Agencies: From Managing by Command to Managing by
Contract?’ Aberdeen Papers in Accountancy, Finance and Management, Working
Paper W6, University of Aberdeen.
Jordan, G. and Cairney, P. (2013) ‘What is the Dominant Model of British Policy
Making? Comparing Majoritarian and Policy Community Ideas’, British Politics,
8: 233–59.
Jordan, G. and Richardson, J. J. (1982) ‘The British Policy Style or the Logic of
Negotiation’. In J. J. Richardson (ed.), Policy Styles in Western Europe. London:
Allen & Unwin, pp. 80–110.
Jordan, G. and Richardson, J. J. (1987a) British Politics and the Policy Process. London:
Allen & Unwin.
Jordan, G. and Richardson, J. J. (1987b) Government and Pressure Groups in Britain.
Oxford: The Clarendon Press.
Josselin, D. (1994) Domestic Policy Networks and the Making of EC Policy: The Case of
Financial Services in France and the UK, 1987–1992. Unpublished PhD thesis,
University of London.
Josselin, D. (1997) Money Politics in the New Europe. Houndmills, Basingstoke:
Macmillan.
Karpik, L. (1978) ‘Organisations, Institutions and History’. In L. Karpik (ed.), Organ-
isation and Environment. London: Sage, pp. 15–68.
Kassim, H. (1993) ‘Policy Networks, Networks and European Policy making:
A Sceptical View’, West European Politics, 17: 15–27.
References 245
Kassim, H., Peterson, J., Bauer, M. W., Connolly, S., Dehousse, R., Hooghe, L., and
Thompson, A. (2013) The European Commission of the Twenty-First Century.
Oxford: Oxford University Press.
Katsaris, A. (2014) Europeanization and Policy Networks in the EU’s Southern Neigh-
bourhood: The European Commission’s Relations with Morocco and Algeria on
Climate Change Policy. PhD thesis, University of Southampton.
Katz, D. and Kahn, R. L. (1978) The Social Psychology of Organisations. Second edition.
New York: Wiley.
Kaufman, F. X., Majone, G., and Ostrom, V. (eds) (1986) Guidance, Control, and
Evaluation in the Public Sector. Berlin: de Gruyter.
Kaufman, G. (1992) ‘Privatising the Ministers’, The Guardian, 7 December: 21.
Kavanagh, D. (1990) Thatcherism and British Politics: The End of Consensus? Oxford:
Kavanagh, D. (1991) ‘Why Political Science Needs History’, Political Studies,
39: 479–95.
Keating, M. and Jones, B. (eds) (1985) Regions in the European Community. Oxford:
Clarendon Press.
Keck, M. E. and Sikkink, K. (1998) Activists beyond Borders. Advocacy Networks in
International Politics. Ithaca: Cornell University Press.
Keeling, D. (1972) Management in Government. London: Allen & Unwin.
Keliher, L. (1995) ‘Core Executive Decision Making on High Technology’. In
R. A. W. Rhodes and P. Dunleavy (eds), Prime Minister, Cabinet and Core Executive.
London: Macmillan, pp. 219–47.
Kellner, P. and Crowther-Hunt, Lord (1980) The Civil Servants: An Inquiry Into
Britain’s Ruling Class. London: Macdonald.
Kenealy, D., Peterson, J., and Corbett, R. (2015) The European Union: How Does It
Work? Fourth edition. Oxford: Oxford University Press.
Kenis, P. and Schneider, V. (1991) ‘Policy Networks and Policy Analysis: Scrutinizing
a New Analytical Toolbox’. In B. Marin and R. Mayntz (eds), Policy Networks:
Empirical Evidence and Theoretical Considerations. Frankfurt am Main: Campus
Verlag, pp. 25–59.
Keohane, R. O. (2002) Power and Governance in a Partially Globalized World. London:
Routledge.
Keohane, R. O. (2003) ‘Global Governance and Democratic Accountability’. In
D. Held and M. Koenig-Archibugi (eds), Taming Globalization: Frontiers of Gov-
ernance. Cambridge: Polity Press, pp. 130–59.
Keohane, R. O. and Hoffman, S. (eds) (1991) The New European Community: Decision
Making and Institutional Change. Boulder, CO: Westview Press.
Keohane, R. O. and Nye, J. S. (1977) Power and Interdependence. Boston: Little Brown.
Keohane, R. O. and Nye, J. S. (1987) ‘Power and Independence Revisited’, Internation-
al Organization, 41: 725–53.
Kettl, D. F. (1993) Sharing Power: Public Governance and Private Markets. Washing-
ton, DC: The Brookings Institution.
Keynes, S. and Tetlow, G. (2014) Survey of Public Spending in the UK. IFS Briefing
Note BN43. London: Institute for Fiscal Studies. Available at: http://www.ifs.org.uk/
246 References
uploads/publications/bns/BN43%20Public%20Spending%202014.pdf. Last accessed
8 October 2015.
Kickert, W. (1993a) ‘Autopoiesis and the Science of (Public) Administration: Essence,
Sense and Nonsense’, Organisation Studies, 14: 261–78.
Kickert, W. (1993b) ‘Complexity, Governance and Dynamics: Conceptual Explor-
ations of Public Network Management’. In J. Kooiman (ed.), Modern Governance.
London: Sage pp. 191–204.
Kickert, W. J. M., Klijn, E-H., and Koppenjan, J. F. M. (eds) (1997 Managing Complex
Networks: Strategies for the Public Sector. London: Sage.
King, A. (1985a) ‘Introduction: The Textbook Prime Ministership’. In A. King (ed.),
The British Prime Minister: A Reader. Second edition. London: Macmillan, pp. 1–11.
King, A. (1985b) ‘Margaret Thatcher: The Style of a Prime Minister’. In A. King (ed.),
The British Prime Minister: A Reader. Second edition. London: Macmillan, pp. 96–140.
King, A. (2001) Does the United Kingdom Still Have a Constitution? London: Sweet &
Maxwell.
King, A. and Crewe, I. (2013) The Blunders of our Government London: Oneworld.
Kingdon, J. W. (1984) Agendas, Alternatives and Public Policies. Boston: Little, Brown.
Kingdom, J. (1991) Government and Politics in Britain. Cambridge: Polity Press.
Kjær, A. M. (2004) Governance. Cambridge: Polity Press.
Klijn, E-H. (2002) ‘Governing Networks in the Hollow State: Contracting Out, Process
Management or a Combination of the Two?’ Public Management Review, 4: 149–65.
Klijn, E-H. (2008) ‘Governance and Governance Networks in Europe: An Assessment
of 10 Years of Research on the Theme’, Public Management Review, 10: 505–25.
Klijn, E-H. and Koppenjan, J. (2015) Governance Networks in the Public Sector.
A Network Approach to Public Problem Solving, Policy Making and Service Delivery.
Abingdon, Oxon: Routledge.
Klijn, E. H., Koopenjan, J., and Termeer, K. (1995) ‘Managing Networks in the Public
Sector’, Public Administration, 73: 437–54.
Knoke, D. (1990) Political Networks: The Structural Perspective. New York: Cambridge
University Press.
Knoke, D., Pappi, F. U., Broadbent, J., and Tsujinaka, Y. (1996) Comparing Policy
Networks. Labor Politics in the U.S., Germany and Japan. New York: Cambridge
University Press.
Kogan, M. (1975) Educational Policy Making. London: Allen & Unwin.
Kohler-Koch, B. and Eising, R. (eds) (1999) The Transformation of Governance in the
European Union. London: Routledge/ECPR Studies in European Political Science.
Koliba, C., Meek, J. W., and Zia, A. (2011) Governance Networks in Public Adminis-
tration and Public Policy. Boca Raton, FL: CRC Press.
Kooiman, J. (1993a) ‘Social-Political Governance: Introduction’. In J. Kooiman (ed.),
Modern Governance. London: Sage, pp. 1–6.
Kooiman, J. (1993b) ‘Findings Speculations and Recommendations’. In J. Kooiman
(ed.), Modern Governance. London: Sage, pp. 249–62.
Kooiman, J. (ed.) (1993c) Modern Governance. London: Sage.
Kramer, R. M. and Tyler, T. (eds) (1996) Trust in Organizations: Frontiers of Theory
and Research. London: Sage.
References 247
Krasner, S. D. (ed.) (1982) ‘International Regimes’; special issue of International
Organisation, 36 (2).
La Porte, T. R. (ed.) (1975) Organized Social Complexity. Princeton, NJ: Princeton University
Press.
Laffan, B. (1996) ‘Ireland: a Region without Regions: The Odd Man Out?’ In L. Hooghe
(ed.) (1996) Cohesion Policy and European Integration: Building Multi-Level Gov-
ernance. Oxford: Oxford University Press, pp. 320–41.
Laffin, M. (1986) Professionalism and Policy: The Role of the Professions in the Central-
Local Government Relationship. Aldershot: Gower.
Landau, M. (1969) ‘Redundancy, Rationality and the Problem of Duplication and
Overlap’, Public Administration Review, 29: 346–58.
Larson, A. (1992) ‘Network Dyads in Entrepreneurial Settings: A Study of Governance
Exchange Relationships’, Administrative Science Quarterly, 37: 76–104.
Laumann, E. and Knoke, D. (1987) The Organizational State. Wisconsin: University of
Wisconsin Press.
Laumann, E. and Pappi, F. U. (1976) Networks of Collective Action: A Perspective on
Community Influence Systems. New York: Academic Press.
Lawton, A. and Rose, A. (1991) Organisation and Management in the Public Sector.
London: Pitman.
Lee, M. (1974) ‘ “Central Capacity” and Established Practice: The Changing Character
of the “Centre of the Machine” in British Cabinet Government’. In B. Chapman and
A. Potter (eds), WJMM: Political Questions—Essays in Honour of W. J. M. McKenzie.
Manchester: Manchester University Press, pp. 162–89.
Leftwich, A. (1993) ‘Governance, Democracy and Development in the Third World’,
Third World Quarterly, 14: 605–24.
LeGalès, P. and Thatcher, M. (eds) (1995) Les Reseaux de Politique Publique. Debat
Autour des Policy Networks. Paris: Editions L’Harmattan.
Lewis, D. and Wallace, H. (eds) (1984) Policies into Practice: National and Internation-
al Case Studies in Implementation. London: Heinemann.
Leonardi, R. (1992) The Regions and the European Community. The Regional Response
to the Single Market in the Underdeveloped Areas, Special Issue of Regional Politics
and Policy, 1–2: 1–274.
Lindblom, C. E. (1965) The Intelligence of Democracy. New York: The Free Press.
Lindquist, E. A. (1996) ‘New Agendas for Research on Policy Communities: Policy Analysis,
Administration and Governance’. In L. Dobuzinskis, M. Howlett, and D. Laycock (eds),
Policy Studies in Canada: The State of the Art. Toronto: University of Toronto Press, 219–41.
Ling, T. (2002) ‘Delivering Joined-Up Government in the UK: Dimensions, Issues and
Problems’, Public Administration, 80: 615–42.
Linklater, M. and Leigh, D. (1986) Not with Honour: the Inside Story of the Westland
Scandal. London: The Observer.
Linn, N. (1999) ‘Social Networks and Status Attainment’, Annual Review of Sociology,
61: 900–7.
Linstead, S. (1997) ‘The Social Anthropology of Management’, British Journal of
Management, 8: 85–98.
Lipsky, M. (1980) Street Level Bureaucracy. New York: Russell Sage Foundation.
248 References
Loader, I. (2000) ‘Plural Policing and Democratic Governance’, Social and Legal
Studies, 9: 323–45.
Lorsch, J. W. (ed.) (1987) Handbook of Organisational Behaviour. Englewood Cliffs,
NJ: Prentice Hall.
Loughlin, M. (1992) Public Law and Political Theory. Oxford: Clarendon Press.
Loughlin, J. and Peters, B. G. (1997) ‘State Traditions, Administrative Reform and
Regionalization’. In M. Keating and J. Loughlin (eds), The Political Economy of
Regionalism. London: Frank Cass, pp. 41–62.
Low, S. (1904) The Governance of England. London, Fisher Unwin.
Lowi, T. (1969) The End of Liberalism. New York: Norton.
LSE Public Policy Group (2011) Maximizing the Impacts of Your Research:
A Handbook for Social Scientist. London: LSE. Available at: http://www.lse.ac.uk/
government/research/resgroups/LSEPublicPolicy/Docs/LSE_Impact_Handbook_
April_2011.pdf. Last accessed 5 September 2016.
Luhmann, N. (1982) The Differentiation of Society. New York: Columbia University Press.
Luhmann, N. (1986) ‘The Autopoiesis of Social Systems’. In F. Geyer and J. van der
Zouwen (eds), Sociocybernetic Paradoxes. London: Sage, pp. 172–92.
MAC (Management Advisory Committee) (2004) Connecting Government. Whole of
Government Response to Australia’s Priority Challenges. Canberra: Australian Public
Service Commission.
MacIntyre, A. (1972) ‘Is A Science of Comparative Politics Possible?’ In P. Laslett,
W. G. Runciman, and Q. Skinner (eds), Philosophy, Politics and Society. Fourth
series. Oxford: Basil Blackwell, pp. 8–26.
MacIntyre, A. (1983) ‘The Indispensability of Political Theory’. In D. Miller and
L. Siedentop (eds), The Nature of Political Theory. Oxford: Clarendon Press, pp. 17–33.
Mackenzie, W. J. M. (1975) ‘Public Administration in the Universities’. In his Explor-
ations in Government: Collected Papers 1951–1968. London: Macmillan, pp. 4–16.
Mackie, T. T. and Hogwood, B. W. (eds) (1985) Unlocking the Cabinet: Cabinet
Structures in Comparative Perspective. London: Sage.
Mackintosh, J. P. (1968) [1962] The British Cabinet. Second edition. London: Stevens & Co.
Madgwick, P. (1986) ‘Prime Ministerial Power Revisited’, Social Studies Review, 1: 28–35.
Madgwick, P. (1991) British Government: The Central Executive Territory. London:
Philip Allan.
Mandell, M. P. (ed.) (2001) Getting Results through Collaboration. Networks and
Network Structures for Public Policy and Management. Westport, Connecticut:
Quorum Books.
Manzer, R. A. (1970) Teachers and Politics. Manchester: Manchester University Press.
Marando, V. L. and Florestano, P. S. (1990) ‘Intergovernmental Management: the State
of the Discipline’. In Lynn, N. and Wildavsky, A. (eds.) Public Administration: the
state of the discipline. Chatham, N. J., Chatham House, pp. 287–317.
March, J. G. and Olsen, J. P. (1989) Rediscovering Institutions. New York: The Free
Press.
Margach, J. (1978) The Abuse of Power: The War between Downing Street and the
Media from Lloyd George to James Callaghan. London: W. H. Allen.
Marin, B. and Mayntz, R. (eds) (1991) Policy Networks: Empirical Evidence and
Theoretical Considerations. Frankfurt am Main: Campus Verlag.
References 249
Marinetto, M. (2003) ‘Governing beyond the Centre: A Critique of the Anglo-
Governance School’, Political Studies, 51: 592–608.
Marks, G. (1992) ‘Structural Policy in the European Community’. In A. Sbragia (ed.),
Euro-Politics: Institutions and Policy Making in the ‘New’ European Community.
Washington DC: Brooking Institution, pp. 191–224.
Marks, G. (1993) ‘Structural Policy and Multi-level Governance in the EC’. In
A. W. Cafrany and G. G. Rosenthal (eds), The State of the European Community.
Volume 2. The Maastricht Debate and Beyond. Harlow, Essex: Longman, pp. 391–410.
Marks, G. (1996) ‘Exploring and Explaining Variations in EU Cohesion Policy’. In
L. Hooghe (ed.) (1996) Cohesion Policy and European Integration: Building Multi-
Level Governance. Oxford: Oxford University Press, pp. 388–422.
Marsh, D. (1983) ‘Introduction’. In D. Marsh (ed.), Pressure Politics. London: Junction
Books, pp. 1–19.
Marsh, D. (ed.) (1998) Comparing Policy Networks. Buckingham: Open University Press.
Marsh, D. (2008a) ‘Understanding British Government: Analysing Competing
Models’, British Journal of Politics and International Relations, 10: 251–68.
Marsh, D. (2008b) ‘What Is at Stake? A Response to Bevir and Rhodes’, British Journal
of Politics and International Relations, 10: 735–9.
Marsh, D. (2009) ‘Keeping Ideas in their Place: In Praise of Thin Constructivism’,
Australian Journal of Political Science, 44: 679–96.
Marsh, D. (2011) ‘The New Orthodoxy: The Differentiated Polity Model’, Public
Marsh, D. and McCaffrie, B. (2015) ‘One Cheer for Jordan and Cairney: Taking the
Governance Literature Seriously’, British Politics, 10: 475–85.
Marsh, D. and Rhodes, R. A. W. (1992a) Policy Networks in British Government.
Oxford: Clarendon Press.
Marsh, D. and Rhodes, R. A. W. (eds) (1992b) Implementing Thatcherite Policies.
Buckingham: Open University Press.
Marsh, D. and Smith, M. (2000) ‘Understanding Policy Networks: Towards a Dialect-
ical Approach’, Political Studies, 48: 4–21.
Marsh, D. and Smith, M. (2001) ‘There Is More than One Way to Do Political Science:
On Different Ways to Study Policy Networks’, Political Studies, 49: 528–41.
Marsh, D., Richards, D. and Smith, M. J. (2001) Changing Patterns of Governance in
the United Kingdom. Houndmills, Basingstoke: Palgrave.
Marsh, D., Richards, D., and Smith, M. J. (2003) ‘Unequal Plurality: Towards an
Asymmetric Power Model of British Politics’, Government and Opposition, 38:
306–32.
Marshall, G. (1986) Constitutional Conventions: the Rules and Forms of Political
Accountability. Oxford: Clarendon Press, paperback edition with additions.
Marshall, G. and Moodie, G. C. (1967) Some Problems of the Constitution. London:
Hutchinson.
Marwick, A., (1995) ‘Two Approaches to Historical Study: The Metaphysical (includ-
ing “Postmodernism”) and the Historical’, Journal of Contemporary History, 30:
5–35.
May, A. and Rowan, K. (1982) Inside Information: British Government and the Media.
London: Constable.
250 References
Maynard-Moody, S. and Musheno, M. (2003) Cops, Teachers, Counsellors: Stories from
the Front Lines of Public Service. Ann Arbor, Michigan: The University of Michigan
Press.
Mazey, S. and Richardson, J. J. (eds) (1993) Lobbying in the European Union. Oxford:
McAnulla, S. (2006a) ‘Challenging The New Interpretivist Approach: Towards
A Critical Realist Alternative’, British Politics, 1: 113–38.
McAnulla, S. (2006b) ‘Critical Realism, Social Structure and Political Analysis: A Reply
to Bevir and Rhodes’, British Politics, 1: 404–12.
McCall, M. W. (1977) ‘Leaders and Leadership: Of Substance and Shadow’. In
J. R. Hackman, E. Lawler III, and L. W. Porter (eds), Perspectives on Behaviour in
Organisations. New York: McGraw Hill, pp. 375–86.
McFarland, A. (1987) ‘Interest Groups and Theories of Power in America’, British
Journal of Political Science, 17: 129–47.
McGuire, M. (2002) ‘Managing Networks: Propositions on What Managers Do and
Why They Do It’, Public Administrative Review, 62: 599–609.
McPherson, A. and Raab, C. (1988) Governing Education. Edinburgh: Edinburgh
University Press.
Meier, K. J. and O’Toole, L. J. (2005) ‘Managerial Networking: Issues of Measurement
and Research Design’, Administration & Society, 37: 523–41.
Metcalfe, L. and Richards, S. (1991) Improving Public Management. Second edition.
London: Sage.
Mill, J. S. (1969) [1840] ‘Coleridge’. In J. M. Robson (ed.), Essays on Ethics, Religion
and Society. The Collected Works of John Stuart Mill. Volume 10. Toronto: Univer-
sity of Toronto Press, pp. 117–64. Available at: http://oll.libertyfund.org/titles/241.
Last accessed 26 February 2016.
Mills, M. and Saward, M. (1994) ‘All Very Well in Practice, But What about the
Theory? A Critique of the British Idea of Policy Networks’. In P. Dunleavy and
J. Stanyer (eds), Contemporary Political Studies 1994. PSA Annual Conference,
University of Swansea, 29–31 March, pp. 79–92.
Milward, H. B. and Provan, K. G. (2000) ‘Governing the Hollow State’, Journal of
Public Administration Research & Theory, 10: 359–80.
Minogue, K. and Biddiss, M. (eds) (1988) Thatcherism: Personality and Politics.
London: Macmillan.
Mitchell, J. (1994) ‘The Articulation of Regional Interests in the European Union’. In
P. Dunleavy and J. Stanyer (eds), Contemporary Political Studies 1994. PSA Annual
Conference, University of Swansea, 29–31 March.
Mitchell, J. (2009) Devolution in the United Kingdom. Manchester: Manchester
University Press.
Mitchell, J. C. (1969) ‘The Concept and Use of Social Networks’. In J. C. Mitchell (ed.),
Social Networks in Urban Situations. Manchester: Manchester University Press,
pp. 1–50.
Morgan, G. (1986) Images of Organisation. London: Sage.
Morgan, G. (1993) Imaginization. London: Sage.
Morphet, J. (2013) How Europe Shapes British Public Policy. Bristol: The Policy
Press.
References 251
Mottram, R. (1993) ‘Developments in the Public Sector’. PAC Annual Conference,
University of York, 7 September.
Mulgan, R. (2003) Holding Power to Account. Accountability in Modern Democracies.
Houndmills, Basingstoke: Palgrave Macmillan.
Nanetti, R. Y. (1996) ‘EU Cohesion and Territorial Restructuring in Member States’. In
L. Hooghe (ed.) (1996) Cohesion Policy and European Integration: Building Multi-
Level Governance. Oxford: Oxford University Press, pp. 59–88.
National Audit Office (2014) Adult Social Care in England: An Overview. HC1102.
Session 2013–14. London: HMSO.
Naughtie, J. (2002) The Rivals. The Intimate Story of a Political Marriage. Revised
edition. London: Fourth Estate.
Newman, J. (2005) ‘Introduction’. In J. Newman (ed.), Remaking Governance. Bristol:
The Policy Press, pp. 1–15.
Nicholson, H. (1950) Diplomacy. Oxford: Oxford University Press.
Nielsen, K. and Pedersen, O. K. (1988) ‘The Negotiated Economy: Ideal and History’,
Scandinavian Political Studies, 11: 79–101.
Nordlinger, L. (1981) On the Autonomy of the Democratic State. Cambridge, MA:
Harvard.
Norton, A. (1991) ‘Western European Local Government in Comparative Perspective’.
In Batley, R. and Stoker, G. (eds), Local Government in Europe. London: Macmillan,
pp. 21–40.
Norton, P. (1983) ‘The Norton Model’. In D. Judge (ed.), The Politics of Parliamentary
Reform. London: Heinemann, pp. 54–69.
Norton, P. (1984) The British Polity. New York: Longman.
Norton, P. (1988) ‘Prime Ministerial Power’, Social Studies Review, 3: 108–15.
Norton, P. (1996) ‘Constitutional Change’, Talking Politics, 9: 17–22.
Norton, P. (2000) ‘Barons in a Shrinking Kingdom: Senior Ministers in British
Government’. In R. A. W. Rhodes (ed.), Transforming British Government. Volume 2.
Changing Roles and Relationships. London: Macmillan, pp. 101–24.
Nugent, N. (1994) The Government and Politics of the European Union. London:
Macmillan.
Nugent, N. (2010) The Government and Politics of the European Union. Seventh
revised edition. London: Palgrave Macmillan.
O’Toole, L. (1997) ‘Treating Networks Seriously: Practical and Research Based Agendas
in Public Administration’, Public Administration Review, 57: 45–52.
Oakeshott, M. (1983) On History and Other Essays. Oxford: Oxford University Press.
Oakeshott, M. (2004) What is History? And Other Essays. Edited by Luke O’Sullivan.
Exeter: Imprint Academic.
Offe, C. (1984) Contradictions of the Welfare State. London: Hutchinson.
Olsen, J. P. (1983) Organised Democracy. Oslo: Universitetsforlaget.
Osborne, D. and Gaebler, T. (1992) Reinventing Government. Reading, MA: Addison-
Wesley.
Osborne, G. (2015) Speech by the Chancellor of the Exchequer on ‘Building a Northern
Powerhouse’, Victoria Warehouse, Manchester, 14 May 2015. Available at: https://
www.gov.uk/government/speeches/chancellor-on-building-a-northern-powerhouse/.
Last accessed 7 January 2016.
252 References
Osborne, S. P. (2010) ‘Introduction. The New Public Governance: A Suitable Case for
Treatment’. In S. P. Osborne (ed.), The New Public Governance: Emerging Perspec-
tives on the Theory and Practice of Public Governance. London and New York:
Routledge and Taylor & Francis, pp. 1–16.
Ostrom, E. (1986) ‘Method of Institutional Analysis’. In F. X. Kaufman, G. Majone,
and V. Ostrom (eds), Guidance, Control, and Evaluation in the Public Sector. Berlin:
de Gruyter, pp. 459–75.
Page, E. C. and Goldsmith, M. J. (eds) (1987) Central and Local Government Relations:
A Comparative Analysis of Western European Unitary States. London: Sage.
Painter, C., Rouse, J., and Isaac-Henry, K. (1997) ‘Local Authorities and Non-elected
Agencies: Strategic Responses and Organizational Networks’, Public Administra-
tion, 77: 225–45.
Parkin, F. (1979) Marxism and Class Theory: A Bourgeois Critique. London: Tavistock.
Parsons, W. (1995) Public Policy. Aldershot: Edward Elgar.
Perrow, C. (1986) [1972] Complex Organisations, Third edition. Glencoe, IL: Scott,
Foresman.
Peston, R. (2005) Brown’s Britain. London: Short Books.
Peters, B. G. (1992) ‘Bureaucratic Politics and the Institutions of the European Com-
munity’. In S. Sbragia (ed.), Euro-Politics: Institutions and Policy Making in the ‘New’
European Community. Washington: The Brookings Institution, pp. 75–122.
Peters, B. G. (1994) ‘Managing the Hollow State’, International Journal of Public
Peters, B. G. (1996) The Future of Governing: Four Emerging Models. Lawrence,
Kansas: University of Kansas Press.
Peters, B. G. (1998) ‘Managing Horizontal Government: The Politics of Coordination’,
Peters, B. G. and Pierre, J. (1998) ‘Governing without Government: Rethinking Public
Administration’, Journal of Public Administration and Theory, 8: 223–42.
Peters, B. G. and Pierre, J. (2009) ‘Governance Approaches’. In A. Wiener and T. Diez
(eds), European Integration Theory. Second edition. Oxford: Oxford University
Press, pp. 91–104.
Peters, B. G., Rhodes, R. A. W. and Wright, V. (eds) (2000) Administering the Summit.
London: Macmillan.
Peterson, J. (1992) ‘The European Technology Community: Policy Networks in a
Supranational Setting’. In D. Marsh and R. A. W. Rhodes (eds), Policy Networks
in British Government. Oxford: Clarendon Press, pp. 226–48.
Peterson, J. (1994) Understanding Decision-Making in the European Union. York:
Department of Politics, University of York, Working Paper No. 4.
Peterson, J. (1995a) ‘Decision-making in the European Union: Towards a Framework
for Analysis’, European Journal of Public Policy, 2: 69–93.
Peterson, J. (1995b) ‘Policy Networks and European Policy Making: A Reply to
Kassim’, West European Politics, 18: 389–407.
Peterson, J. (2009) ‘Policy Networks’. In A. Wiener and T. Diez (eds), European
Integration Theory. Second edition. Oxford: Oxford University Press, pp. 105–24.
Peterson, J. and Bomberg, E. (1999) Decision-Making in the European Union. Hound-
mills, Basingstoke: Palgrave Macmillan.
References 253
Pfeffer, J. and Salancik, G. R. (1978) The External Control of Organizations: A Resource
Dependence Perspective. New York: Harper & Row.
Piattoni, S. (2010) The Theory of Multi-level Governance. Oxford: Oxford University
Press.
Pierre, J. (ed.) (2000) Debating Governance. Oxford: Oxford University Press.
Pierre, J. and Peters, B. G. (2000) Governance, Politics and the State. Houndmills,
Poguntke, T. and Webb, P. (eds) (2005) The Presidentialization of Politics:
A Comparative Study of Modern Democracies. Oxford: Oxford University Press.
Pollitt, C. (1993) Managerialism and the Public Services. Second edition. Oxford:
Blackwell.
Pollitt, C. (2003) ‘Joined-up Government: A Survey’, Political Studies Review, 1:
34–49.
Pollitt, C. and Bouckaert, G. (2011) Public Management Reform. A Comparative
Analysis: New Public Management, Governance and the Neo-Weberian State.
Third edition. Oxford: Oxford University Press.
Powell, W. (1991) ‘Neither Market nor Hierarchy: Network Forms of Organisation’. In
G. Thompson et al. (eds), Markets, Hierarchies & Networks: the Co-ordination of
Social Life. London: Sage, pp. 265–76.
Powell, W. (1996) ‘Trust-based Forms of Governance’. In R. M. Kramer and T. Tyler
(eds), Trust in Organisations: Frontiers of Theory and Research. London: Sage,
pp. 51–67.
Pressman, J. and Wildavsky, A. (1984) [1973] Implementation. How Great Expectations
in Washington Are Dashed in Oakland. Third edition. Berkeley: University of
California Press.
Preston, C. (1984) The Politics of Implementation: The European Regional Develop-
ment Fund and EC Regional Aid to the UK (1975–81). Unpublished PhD thesis,
University of Essex, Colchester.
Putnam, R. A. (1988) ‘Diplomacy and Domestic Politics: The Logic of Two-Level
Games’, International Organization, 42: 427–60.
Raab, J. and Milward, H. B. (2003) ‘Dark Networks as Problems’, Journal of Public
Administration Theory and Research, 13: 413–39.
Raco, M. (2016) ‘Mass-Privatisation and the Changing Nature of Governance in the
UK’. In M. Bevir and R. A. W. Rhodes (eds), Rethinking Governance: Ruling
Rationalities and Resistance. Abingdon, Oxon: Routledge, pp. 50–69.
Ranson, S. and Stewart, J. (1989) ‘Citizenship and Government: The Challenge for
Management in the Public Domain’, Political Studies, 37: 5–24.
Ranson, S., Jones, G. W., and Walsh, K. (eds) (1985) Between Centre and Locality.
London: Allen & Unwin.
Rawnsley, A. (2001) Servants of the People. The Inside Story of New Labour. Revised
edition. London: Penguin Books.
Redford, E. S. (1969) Democracy in the Administrative State. London: Oxford Univer-
sity Press.
Rein, M. (1976) Social Science and Public Policy. Harmondsworth: Penguin Books.
Rhodes, R. A. W. (1973) ‘The European Community and British Public Administration:
The Case of Local Government’, Journal of Common Market Studies, 11: 263–75.
254 References
Rhodes, R. A. W. (1976) ‘Centre-Local Relations’. In Committee of Inquiry into Local
Government Finance (Layfield), Appendix 6, The Relationship Between Central and
Local Government. London: HMSO, pp. 174–202.
Rhodes, R. A. W. (1977) Future Research into Central–Local Relations. Report to the
SSRC Panel on Research into Local Government, April.
Rhodes, R. A. W. (1978a) Central–Local Relations: The Search for Positive Theory.
Report to the SSRC Panel on Central-Local Government Relationships for the
Public Administration Committee Working Party on Research into Central–Local
Relations. London: SSRC, mimeo, January.
Rhodes, R. A. W. (1978b) The Rationality of Ambiguous Confusion: Rules, Strategies
and Prizes in British Central–Local Relations. Report to the SSRC Panel on Central–
Local Government Relationships, London: SSRC, mimeo, May.
Rhodes, R. A. W. (1979a) Public Administration and Policy Analysis: Recent Develop-
ment in Britain and America. Farnborough: Saxon House.
Rhodes, R. A. W. (1979b) ‘Research into Central-Local Relations in Britain: a Frame-
work for Analysis’. In Social Science Research Council, Central–Local Government
Relationships. London: SSRC: Appendix A.
Rhodes, R. A. W. (1980) ‘Analysing Intergovernmental Relations’, European Journal of
Political Research, 8: 289–322.
Rhodes, R. A. W. (1981) Control and Power in Central–Local Government Relation-
ships. Farnborough: Gower.
Rhodes, R. A. W. (1985) ‘Power-Dependence, Policy Communities and Intergovern-
mental Relations’, Public Administration Bulletin, No. 49, December: 4–31.
Rhodes, R A. W. (1986a) The National World of Local Government. London: Allen &
Unwin.
Rhodes, R. A. W. (1986b) ‘ “Power-dependence” Theories of Central–Local Relations:
A Critical Assessment’. In M. J. Goldsmith (ed.), New Research in Central–Local
Relations. Aldershot: Gower, pp. 1–33.
Rhodes, R. A. W. (1986c) European Policy Making, Implementation and Sub-central
Governments. Maastricht: European Institute of Public Administration.
Rhodes, R. A. W. (1988) Beyond Westminster and Whitehall: Sub-central Governments
in Britain. London: Allen & Unwin.
Rhodes, R. A. W. (1990) ‘Policy Networks: A British Perspective’, Journal of Theoret-
ical Politics, 2: 293–317.
Rhodes, R. A. W. (1991a) ‘Theory and Methods in British Public Administration: The
View from Political Science’, Political Studies, 39 (3): 533–54.
Rhodes, R. A. W. (1991b) Local Governance. Swindon: Report to the ESRC’s Society
and Politics Research Development Group, November.
Rhodes, R. A. W. (1992a) ‘Beyond Whitehall: Researching Local Governance’, Social
Sciences, 13, February: 2.
Rhodes, R. A. W. (1992b) ‘Management in Local Government: Twenty Years On’. In
S. Leach (ed.), Strengthening Local Government in the 1990s. Harlow: Longman,
pp. 14–48.
Rhodes, R. A. W. (1993) The Changing Nature of the British Executive: A Research
Proposal. Swindon, Report to the ESRC’s Research Programmes Board, May.
References 255
Rhodes, R. A. W. (1994) ‘The Hollowing-Out of the State’, Political Quarterly, 65:
138–51.
Rhodes, R. A. W. (1995a) ‘From Prime Ministerial Power to Core Executive’. In
R. A. W. Rhodes and P. Dunleavy (eds) (1995) Prime Minister, Cabinet and Core
Executive. London: Macmillan, pp. 11–37.
Rhodes, R. A. W. (1995b) ‘The Institutional Approach’. In G. Stoker and D. Marsh
(eds), Theories and Methods in Political Science. London: Macmillan, pp. 42–57.
Rhodes, R. A. W. (1995c) The New Governance: Governing without Government.
London: RSA and ESRC.
Rhodes, R. A. W. (1996a) ‘Looking beyond Managerialism’, Australian Journal of
Rhodes, R. A. W. (1996b) ‘The New Governance: Governing without Government’,
Rhodes, R. A. W. (1997a) Understanding Governance. Buckingham: Open University
Press.
Rhodes, R. A. W. (1997b) ‘From Marketization to Diplomacy: It’s the Mix that
Matters’, Australian Journal of Public Administration, 56: 40–53.
Rhodes, R. A. W. (1997c) ‘Reinventing Whitehall, 1979–95’. In W. Kickert (ed.), Public
Management and Administrative Reform in Western Europe. Aldershot: Edward
Elgar: 43–60.
Rhodes, R. A. W. (1998) ‘Different Roads to Unfamiliar Places: UK Experience in
Comparative Perspective’, Australian Journal of Public Administration, 57 (4): 19–31.
Rhodes, R. A. W. (1999a) [1981] Control and Power in Central–Local Government
Relationships. Aldershot: Ashgate. Revised edition.
Rhodes, R. A. W. (1999b) ‘Foreword: Governance and Networks’. In G. Stoker
(ed.), The New Management of British Local Governance. London: Macmillan,
pp. xii–xxvi.
Rhodes, R. A. W. (ed.) (2000a) Transforming British Government. Volume 1. Changing
Institutions. Volume 2. Changing Roles and Relationships. London: Macmillan.
Rhodes, R. A. W. (2000b) ‘Introduction: The ESRC Whitehall Programme: A Guide to
Institutional Change’. In R. A. W. Rhodes (ed.), Transforming British Government.
Volume 1. Changing Institutions. London: Macmillan, pp. 1–22.
Rhodes, R. A. W. (2000c) ‘New Labour’s Civil Service: Summing Up Joining-Up’,
Political Quarterly, 71: 151–66.
Rhodes, R. A. W. (2002) ‘Putting the People Back into Networks’, Australian Journal of
Political Science, 37: 399–415.
Rhodes, R. A. W. (2003a) ‘What Is Governance and Why Does It Matter?’ In
J. Hayward and A. Menon (eds), Governing Europe. Oxford: Oxford University
Press, pp. 61–73.
Rhodes, R. A. W. (2003b) ‘The Unholy Trinity: Hierarchy, Contracts and Networks’.
Commonwealth of Australia’s Public Service Commission’s Planning Retreat, 19
June 2003, Motel Royal Tara, Binalong, New South Wales, Australia; the ANZSOG
Executive Fellows Program, Macquarie Graduate School of Management, 1–2
September 2003; the Leadership Development Network Breakfast Meeting on
‘The Network State’, University House, ANU, 24 February 2004; and the Blake
256 References
Dawson and Waldron/Australian National University Public Lecture series, Na-
tional Museum of Australia, Canberra, 23 August 2005.
Rhodes, R. A. W. (2005) ‘Everyday Life in a Ministry: Public Administration as
Anthropology’, American Review of Public Administration, 35: 3–26.
Rhodes, R. A. W. (2006a) ‘Policy Network Analysis’. In M. Moran, M. Rein, and
R. E. Goodin (eds), The Oxford Handbook of Public Policy. Oxford: Oxford Univer-
sity Press, pp. 423–45.
Rhodes, R. A. W. (2006b) ‘The Sour Laws of Network Governance’. In J. Fleming and
J. Wood (eds), Fighting Crime Together. The Challenges of Policing and Security
Networks. Sydney: UNSW Press, pp. 15–34.
Rhodes, R. A. W. (2007a) ‘The Everyday Life of a Minister: A Confessional and
Impressionist Tale’. In R. A. W. Rhodes, P. ‘t Hart, and M. Noordegraaf (eds),
Observing Government Elites: Up Close and Personal. Houndmills, Basingstoke:
Palgrave-Macmillan, pp. 21–50.
Rhodes, R. A. W. (2007b) ‘Understanding Governance: Ten Years On’, Organization
Studies, 28: 1243–64.
Rhodes, R. A. W. (2011a) Everyday Life in British Government. Oxford: Oxford
University Press.
Rhodes, R. A. W. (2011b) ‘Dead End Street: the American Influence on British Public
Administration’, Public Administration Review, 71: 559–71.
Rhodes, R. A. W. (2011c) ‘Thinking On: I Was So Much Older Then’, Public Admin-
istration, 89 (1): 196–212.
Rhodes, R. A. W. (2014) ‘Core Executives, Prime Ministers, Statecraft and Court
Politics: Towards Convergence’. In Glyn Davis and R. A. W. Rhodes (eds), The
Craft of Governing: Essays in Honour of Professor Patrick Weller. Crows Nest, NSW:
Allen & Unwin, pp. 53–72.
Rhodes, R. A. W. (2016) ‘Recovering the Craft of Public Administration’, Public
Administration Review, 76 (4): 638–47.
Rhodes, R. A. W. (2017) Interpretive Political Science: Selected Essays Volume II.
Rhodes, R. A. W. and Dunleavy, P. (eds) (1995) Prime Minister, Cabinet and Core
Executive. London: Macmillan.
Rhodes, R. A. W. and Marsh, D. (1994) ‘Policy Networks: “Defensive Comments”,
Modest Claims and Plausible Research Strategies’. In P. Dunleavy and J. Stanyer
(eds), Contemporary Political Studies 1994. PSA Annual Conference, University of
Swansea, 29–31 March.
Rhodes, R. A. W. and Weller, P. (eds) (2001) The Changing World of Top Officials:
Mandarins or Valets? Buckingham: Open University Press.
Rhodes, R. A. W. and Weller, P. (2005) ‘Westminster Transplanted and Westminster
Implanted: Explanations for Political Change’. In H. Patapan, J. Wanna, and P. Weller
(eds), Westminster Legacies: Democracy and Responsible Government in Asia, Austra-
lasia and the Pacific. Sydney: University of New South Wales Press, pp. 1–12.
Rhodes, R. A. W. and Wright, V. (eds) (1987) Tensions in Western Europe. London:
Frank Cass.
Rhodes, R. A. W., Hardy, B., and Pudney, K. (1982) Patterns of Resource Exchange
within the National Community of Local Government. Colchester: Department of
Government, University of Essex: Discussion Paper No. 3.
References 257
Rhodes, R. A. W., Bache, I., and George, S. (1996) ‘Policy Networks and Policy Making
in the European Union: A Critical Appraisal’. In L. Hooghe (ed.), Cohesion Policy
and European Integration. Oxford: Clarendon Press, pp. 367–87.
Rhodes, R. A. W., Carmichael, P., McMillan, J., and Massey, A. (2003) Decentralizing
the Civil Service: From Unitary State to Differentiated Policy in the United Kingdom.
Buckingham: Open University Press.
Richards, D. and Smith, M. J. (2002) Governance and Public Policy in the UK. Oxford:
Richardson, J. J. (ed.) (1982) Policy Styles in Western Europe. London: Allen & Unwin.
Richardson, J. J. (1999) ‘Pressure Groups and Parties and Social Movements: A “Haze
of Common Knowledge” or the Empirical Advance of a Discipline?’ In J. Hayward,
B. Barry, and A. Brown (eds), The British Study of Politics in the Twentieth Century.
Oxford: Oxford University Press for the British Academy, pp. 181–222.
Richardson, J. J. (2000) ‘Government, Interest Groups and Policy Change’, Political
Studies, 48: 1006–25.
Richardson, J. J. (2012) ‘Supranational State Building in the European Union’. In
Jeremy Richardson (ed.), Constructing a Policy Making State? Policy Dynamics in
the EU. Oxford: Oxford University Press, pp. 3–28.
Richardson, J. J. and Jordan, G. (1979) Governing under Pressure. The Policy Process in
a Post-Parliamentary Democracy. Oxford: Martin Robertson.
Richelson, J. T. and Ball, D. (1985) The Ties that Bind. Boston: Allen & Unwin.
Ricoeur, P. (1981) Hermeneutics and the Human Sciences: Essays On Language, Action
and Interpretation. Edited, translated and introduced by J. B. Thompson.
Cambridge: Cambridge University Press.
Ricoeur, P. (1991) From Text to Action: Essays in Hermeneutics, II. Translated by
K. Blamey and J. B. Thompson. Evanston, Illinois: Northwestern University Press.
Ridley, F. F. (1975) The Study of Government. London: Allen & Unwin.
Ripley, R. and Franklin, G. (1980) Congress, the Bureaucracy and Public Policy. Home-
wood, IL: Dorsey Press.
Risse, T. (2009) ‘The Social Construction of Europe’. In A. Wiener and T. Diez (eds),
European Integration Theory. Second edition. Oxford: Oxford University Press,
pp. 144–60.
Risse, T., Ropp, S. C., and Sikkink, K. (eds) (1999) The Power of Human Rights:
International Norms and Domestic Change. Cambridge: Cambridge University
Press.
Robson, W. A. (1962) Nationalised Industries and Public Ownership. Second edition.
London: Allen & Unwin.
Robson, W. A. (1975) ‘The Study of Public Administration Then and Now’, Political
Studies, 23 (1/2): 193–201.
Roe, E. (1994) Narrative Policy Analysis: Theory and Practice. Durham, NC: Duke
University Press.
Rokkan, S. (1966) ‘Norway: Numerical Democracy and Corporate Pluralism’. In
R. A. Dahl (ed.), Political Oppositions in Western Democracies. New Haven: Yale
University Press, 70–115.
Rorty, R. (1980) Philosophy and the Mirror of Nature. Oxford: Blackwell.
Rosamond, B. (2000) Theories of European Integration. Houndmills, Basingstoke:
Palgrave Macmillan.
258 References
Rose, R. (1987) Ministers and Ministries: A Functional Analysis. Oxford: Clarendon
Press.
Rose, R. (2001) The Prime Minister in a Shrinking World. Cambridge: Polity Press.
Rosenau, J. (1992a) ‘Governance, Order, and Change in World Politics’. In
J. N. Rosenau and E-O. Czempiel (eds), Governance without Government: Order
and Change in World Politics. Cambridge: Cambridge University Press, pp. 1–29.
Rosenau, J. (1992b) ‘Citizenship in a Changing Global Order’. In J. N. Rosenau and
E-O. Czempiel (eds), Governance without Government: Order and Change in World
Politics. Cambridge: Cambridge University Press, pp. 272–94.
Rosenau, J. (2000) ‘Governance in Globalizing Space’. In J. Pierre (ed.), Debating
Governance. Oxford: Oxford University Press, pp. 167–200.
Rosenau, P. M. (1992) Postmodernism and the Social Sciences: Insights, Inroads
Intrusions. Princeton: Princeton University Press.
Ryan, M., Savage, S. P., and Wall, D. S. (eds) (2001) Policy Networks in Criminal
Justice. Houndmills, Basingstoke: Palgrave.
Sabatier, P. (1986) ‘Top-Down and Bottom-Up Models of Policy Implementation:
A Critical Analysis and Suggested Synthesis’, Journal of Public Policy, 6: 21–48.
Sabatier, P. (1993) ‘Policy Change over a Decade or More’. In P. A. Sabatier H. C. and
Jenkins-Smith (eds), Policy Change and Learning: An Advocacy Coalition Approach.
Boulder: Westview Press, pp. 13–39.
Sabatier, P. and Jenkins-Smith, H. C. (1993) Policy Change and Learning: An Advocacy
Coalition Approach. Boulder, CO: Westview Press.
Salamon, L. M. (ed.) (2002) The Tools of Government: A Guide to the New Governance.
Sartori, G. (1997) Comparative Constitutional Engineering. Second edition. New York:
New York University Press.
Saunders, P. (1983) ‘The “Regional State”: A Review of Literature and Agenda for
Research’. SSRC Conference on Political Theory and Intergovernmental Relations,
Nuffield College, Oxford, September.
Saward, M. (1990) ‘The Civil Nuclear Network in Britain’. In D. Marsh and
R. A. W. Rhodes (eds), Policy Networks in British Government. Oxford: Clarendon
Press, pp. 75–99.
Saward, M. (1997) ‘In Search of the Hollow Crown’. In P. Weller, H. Bakvis, and
R. A. W. Rhodes (eds), The Hollow Crown. London: Macmillan, pp. 16–36.
Sbragia, A. (2000) ‘The European Union as Coxswain: Governance by Steering’. In
J. Pierre (ed.), Debating Governance: Authority, Democracy and Steering. Oxford:
Oxford University Press, pp. 219–40.
Scharpf, F. W. (1988) ‘The Joint Decision Trap: Lessons from German Federalism and
European Integration’, Public Administration, 66: 239–78.
Scharpf, F. W. (1997) Games Real Actors Play. Actor-Centred Institutionalism in Policy
Research. Boulder, CO: Westview Press.
Scharpf, F. W., Reissert, B., and Schnabel, F. (1976) Politikverflechtung. Kronberg:
Scriptor.
Schein, E. H. (1980) Organisational Psychology. Third edition. Englewood Cliffs, NJ:
Prentice Hall.
References 259
Schimmelfenning, F. (2012) ‘Constructivist Perspectives’. In E. Jones, A. Menon, and
S. Weatherill (eds), The Oxford Handbook of the European Union. Oxford: Oxford
Schmitter, P. C. (1979) ‘Still the Century of Corporatism?’ In P. C. Schmitter and
G. Lehmbruch (eds), Trends Towards Corporatist Intermediation. London: Sage, pp. 7–52.
Schmitter, P. C. and Lehmbruch, G. (eds) (1979) Trends Towards Corporatist Inter-
mediation. London: Sage.
Scott, J. (2012) Social Network Analysis: A Handbook. Third edition. London: Sage.
Sedgemoor, B. (1980) The Secret Constitution. London: Hodder & Stoughton.
Seidman, H. (1975) Politics, Position and Power. Second edition. Oxford: Oxford
University Press.
Self, P. J. O. (1971) ‘Tests of Efficiency: Public and Business Administration’, PAC
Bulletin, No. 11, December: 28–41.
Selznick, P. (1966) [1949]. TVA and the Grass Roots, Berkeley: Los Angeles, University
of California.
Seymour-Ure, C. (1984) ‘British “War Cabinets” in Limited Wars: Korea, Suez and the
Falklands’, Public Administration, 62: 181–200.
Shamir, B., Dayan-Horesh, H., and Adler, D. (2005) ‘Leading by Biography: Towards a
Life-Story Approach to the Study of Leadership’, Leadership, 1: 13–29.
Sharpe, L. J. (1985) ‘Central Co-ordination and the Policy Network’, Political Studies,
33: 361–81.
Sharpe, L. J. (ed.) (1993) The Rise of Meso-Government in Europe. London: Sage.
Shore, C. (2000) Building Europe: The Cultural Politics of European Integration.
London: Routledge.
Sikkink, K. (1993) ‘Human Rights, Principled Issue-Networks and Sovereignty in Latin
America’, International Organization, 47: 411–41.
Simon, H. A. (1981) [1969]. The Sciences of the Artificial. Second edition. Cambridge,
MA: MIT Press.
Simon, H. A. (2000) ‘Public Administration in Today’s World of Organizations and
Markets’, PS: Political Science & Politics, 33: 749–56.
Skelcher, C. (2000) ‘Changing Images of the State: Overloaded, Hollowed Out, Con-
gested’, Public Policy and Administration, 15: 3–19.
Skinner, Q. (2002) [1997] ‘The Practice of History and the Cult of the Fact’. In his
Visions of Politics. Volume 1. Regarding Method. Cambridge: Cambridge University
Press pp. 8–26.
Slaughter, Anne-Marie (2003) ‘Everyday Global Governance’, Daedalus, 132: 83–91.
Smircich, L. and Morgan, G. (1982) ‘Leadership: The Management of Meaning’, The
Journal of Applied Behavioral Science, 18: 257–73.
Smith, M. J. (1990) The Politics of Agricultural Support in Britain. The Development of
the Agricultural Policy Community. Aldershot: Dartmouth.
Smith, M. J. (1993) Pressure Power & Policy. State Autonomy and Policy Networks in
Britain and the United States. Hemel Hempstead: Harvester Wheatsheaf.
Smith, M. J. (1995) ‘Interpreting the Rise and Fall of Margaret Thatcher: Power-
Dependence and the Core Executive’. In R. A. W. Rhodes and P. Dunleavy (eds)
Prime Minister, Cabinet and Core Executive. London: Macmillan, pp. 108–24.
260 References
Smith, M. J. (1996) ‘Theoretical and Empirical Challenges to British Central Govern-
ment’. Sheffield: University of Sheffield. Department of Politics, mimeo.
Smith, M. J. (1999) The Core Executive in Britain. London: Macmillan.
Sørensen, E. and Torfing, J. (2007a) ‘Theoretical Approaches to Metagovernance’. In
E. Sørsensen and J. Torfing (eds), Theories of Democratic Network Governance.
Houndmills, Basingstoke: Palgrave Macmillan, pp. 169–82.
Sørensen, E., and Torfing, J. (eds) (2007b) Theories of Democratic Network Govern-
ance. Houndmills, Basingstoke: Palgrave Macmillan.
SSRC (Social Science Research Council) (1979) Central–Local Government Relation-
ships. London: SSRC.
SSRC (Social Science Research Council) (1983) Government and Industry Relation-
ships: A Framework for Analysis. London: SSRC. Government and Industry Sub-
committee, mimeo, December.
Stanworth, P. and Giddens, A. (eds) (1974) Elites and Power in British Society. London:
Cambridge University Press.
Stewart, J. (1971) ‘The Study of Public Policy making’, PAC Bulletin, No. 11, December:
42–56.
Stewart, J. and Walsh, K. (1992) ‘Change in the Management of Public Services’, Public
Stoker, G. (1994) ‘Local Governance in Britain’, Glasgow: Department of Government,
University of Strathclyde, mimeo, November.
Stoker, G. (1997) ‘Governance: Five Propositions’, International Social Science Journal,
50: 17–28.
Stoker, G. (1998) ‘Theory and Urban Politics’, International Political Science Review,
19: 119–29.
Stoker, G. (1999a) The New Management of British Local Governance. Houndmills,
Stoker, G. (1999b) ‘Urban Political Science and the Challenge of Urban Governance’. In
Pierre, J. (ed.), Debating Governance. Oxford: Oxford University Press, pp. 91–109.
Stoker, G. (1999c) ‘Introduction: the Unintended Costs and Benefits of New Management
Reform for British Local Government’. In Stoker, G. (ed.), The New Management of
Local Governance: Audit of an Era of Change in Britain. London: Macmillan, pp. 1–21.
Stoker, G. (2004) Transforming Local Governance. Houndmills, Basingstoke: Palgrave.
Stone, D. (2011) Policy Paradox: The Art of Political Decision Making. Third Edition.
New York: W. W. Norton & Company.
Stones, R. (1992) ‘Labour and International Finance, 1964–67’. In D. Marsh and
R. A. W. Rhodes (eds), Policy Networks in British Government. Oxford: Clarendon
Press, pp. 200–25.
Strange, S. (1982) ‘ “Cave! Hic Dragones”: A Critique of Regime Analysis’, Internation-
al Organisation, 36: 479–96.
Tarrow, S. (1978) ‘Introduction’. In S. Tarrow, P. J. Katzenstein, and L. Graziano (eds),
Territorial Politics in Industrial Nations. London: Praeger, pp. 1–27.
Taylor, A. (1997) ‘ “Arm’s Length but Hands On”. Mapping the New Governance. The
Department of National Heritage and Cultural Policies in Britain’, Public Adminis-
tration, 75: 441–66.
References 261
Taylor, A. (2000) ‘Hollowing Out or Filling In? Task Forces and the Management
on Cross-cutting Issues in British Government’, British Journal of Politics and
International Relations, 2: 46–71.
Thain, C. and Wright, M. (1990) ‘Running Costs Control in UK Central Government’,
Financial Accounting and Management, 6: 115–31.
Thatcher, Margaret (1993) The Downing Street Years. London: HarperCollins.
Thatcher, M. (1998) ‘The Development of Policy Network Analyses’, Journal of
Theoretical Politics, 10: 389–416.
Theakston, K. (1987) Junior Ministers in British Government. Oxford: Oxford Univer-
sity Press.
Thoenig, J-C. and Friedberg, E. (1976) ‘The Power of the Field Staff ’. In A. F. Leemans
(ed.), The Management of Change in Government. The Hague: Martinus Nijhoff,
pp. 31–7.
Thompson, G. (1993) ‘Network Co-ordination’. In R. Maidment and G. Thompson
(eds), Managing the United Kingdom. London: Sage, pp. 55–74.
Thompson, G., Frances, J., Levacic, R., and Mitchell, J. (eds) (1991) Markets Hierarch-
ies and Networks: The Co-ordination of Social Life. London: Sage.
Thompson, J. D. (1967) Organisations in Action. New York: McGraw-Hill.
Timmins, N. (2012) Never Again? The Story of the Health and Social Care Act 2012.
London: Institute of Government.
Tivey, L. (1988) Interpretations of British Politics. Hemel Hempstead: Harvester
Wheatsheaf.
Torfing, J., Peters, B. G., Pierre, J., and Sørsensen, E. (2012) Interactive Governance:
Advancing the Paradigm. Oxford: Oxford University Press.
Toynbee, P. and Walker, D. (2015) Cameron’s Coup: How the Tories Took Britain to
the Brink. London: Guardian Faber Publishing.
Travers, T. (1986) ‘More Accountability or Just More Paperwork?’ Local Government
Chronicle, 18 April: 428.
Travers, T. (2007) ‘Local Government’. In A. Seldom (ed.), Blair’s Britain 1997–2007.
Cambridge: Cambridge University Press, pp. 54–78.
Travers, T. (2015) ‘The Coalition and Society (1): Home Affairs and Local Govern-
ment’. In A. Seldon and M. Finn (eds), The Coalition Effect 2010–2015. Cambridge:
Cambridge University Press, pp. 228–56.
Tricker, R. I. (1984) International Corporate Governance. Englewood Cliffs, NJ: Pren-
tice Hall.
Truman, D. B. (1951) The Governmental Process. New York: Knopf.
Tyler, T. I. and Degoey, P. (1996) ‘Trust in Organisational Authorities’. In
R. M. Kramer and T. Tyler (eds), Trust in Organisations: Frontiers of Theory and
Research. London: Sage, pp. 331–56.
van Eeten, M. J. G., van Twist, M. J. W., and Kalders, P. R. (1996) ‘Van een Narratieve
Bestuurskunde naar een I?’ Bestuurskunde, 5: 168–89. English translation supplied
by Mark van Twist.
Van Gunsteren, H. (1976) The Quest for Control. London: Wiley.
Van Maanen, J. (1988) Tales of the Field: On Writing Ethnography. Chicago: University
of Chicago Press.
262 References
Van Waarden, F. (1992) ‘Dimensions and Types of Policy Networks’, European
Journal of Political Research, 21: 29–52.
Vare, D. (1938) Laughing Diplomat. London: John Murray.
Vickers, Sir Geoffrey (1968) The Art of Judgement. London: Methuen University
Paperback.
Vinzant, J. and Crothers, L. (1998) Street Level Leadership. Washington, DC: George-
town University Press.
Wagenaar, H. (2011) Meaning in Action: Interpretation and Dialogue in Policy Analysis.
Armonk, NY: M. E. Sharpe.
Wagenaar, H. (2012) ‘Dwellers on the Threshold of Practice: The Interpretivism of
Bevir and Rhodes’, Critical Policy Studies, 6: 85–99.
Waldegrave, W. (1990) Trafford Memorial Lecture, 12 December, transcript.
Walker, P. G. (1970) The Cabinet. London: Cape.
Wallace, H. (1984) ‘Implementation across National Boundaries’. In D. Lewis and
H. Wallace (eds), Policies and Practice. London: Heinemann, pp. 129–43.
Wallace, W. (1986) ‘What Price Independence? Sovereignty and Interdependence in
British Politics’, International Affairs, 62: 367–89.
Wallace, H., Wallace, W., and Webb, C. (eds) (1977) Policy Making in the European
Community. Chichester: John Wiley.
Wallace, H., Pollack, M. A., and Young, A. R. (eds) (2014) Policy Making in the
European Union. Seventh edition. Oxford: Oxford University Press.
Wallace, W. (1982) ‘Europe as a Confederation: The Community and the Nation-
State’, Journal of Common Market Studies, 20: 57–68.
Wallace, W. (1983) ‘Less than a Federation, More than a Regime: The Community as a
Political System’. In H. Wallace, W. Wallace, and C. Webb (eds), Policy Making in
the European Community. Chichester: Wiley, pp. 403–36.
Walsh, K. (1991) Competitive Tendering for Local Authority Services: Initial Experi-
ences. London: HMSO.
Walters, W. (2004) ‘Some Critical Notes on “Governance” ’, Studies in Political
Economy, 73: 27–46.
Wamsley, G., Goodsell, C. T., Rohr, J. A., White, O. F., and Wolf, J. (1984) ‘The Public
Administration and the Governance Process: Refocusing the American Dialogue’,
Dialogue 6, 2: 2–18.
Wanna, J. (2005) ‘“Connecting Government” and “Whole of Government”: Meanings,
Claims and Critique’, 17th Annual Government Business Conference, Australian
Government Agencies, Queensland Regional Heads Forum, Gold Coast, 19–20 May.
Wapshot, N. and G. Brock (1983) Thatcher. London: Fontana.
Ward, H. with Samways, D. and Benton, T. (1990) ‘Environmental Politics and Policy’.
In P. Dunleavy, A. Gamble, and G. Peele (eds), Developments in British Politics 3.
London: Macmillan, pp. 221–45.
Wass, D. (1984) Government and the Governed. London: Routledge and Kegan Paul.
Wasserman, S. and Faust, K. (1994) Social Network Analysis. Methods and Applica-
tions. Cambridge: Cambridge University Press.
Wassmer, U. (2010) ‘Alliance Portfolios: A Review and Research Agenda’, Journal of
Management, 36: 141–71.
Watson, A. (1982) Diplomacy. London: Eyre Methuen.
References 263
Weick, K. E. (1995) Sensemaking in Organizations. London: Sage.
Weller, P. (1983) ‘Do Prime Ministers’ Departments Really Create Problems?’ Public
Weller, P. (1989) Malcolm Fraser, the Westminster System and the Separation of
Powers. Brisbane: Griffith University, Research Lecture Series.
Weller, P., Bakviss, H., and Rhodes, R. A. W. (eds) (1997) The Hollow Crown. Houndmills,
Wettenhall, R. (2003) ‘The Rhetoric and Reality of Public-Private Partnerships’, Public
Organization Review, 3: 77–107.
White, H. (1973) Metahistory. Baltimore: Johns Hopkins University Press.
White, H. (1978) Tropics of Discourse. Baltimore: Johns Hopkins University Press.
White, H. (1987) The Content of the Form: Narrative Discourse and Historical Repre-
sentation. Baltimore: Johns Hopkins University Press.
White, H. (1995) ‘Response to Arthur Marwick’, Journal of Contemporary History,
30: 233–46.
White, L. G. (1989) ‘Public Management in a Pluralistic Arena’, Public Administration
Review, 49: 522–32.
Whitten, N. E. and Wolfe, A. W. (1972) ‘Network Analysis’. In J. Honingmann (ed.),
Handbook of Social and Cultural Anthropology. Chicago: Rand McNally, pp. 247–70.
Wildavsky, A. (1980) The Art and Craft of Policy Analysis. London: Macmillan.
Wilks, S. (1989) ‘Government–Industry Relations’, Public Administration 67: 329–39.
Wilks, S. and Wright, M. (1987) ‘Conclusion: Comparing Government–Industry
Relations: States, Sectors, and Networks’. In S. Wilks and M. Wright (eds), Com-
parative, Government–Industry Relations: Western Europe, the United States, and
Japan. Oxford: Clarendon Press, pp. 274–313.
Williams, D. and Young, T. (1994) ‘Governance, the World Bank and Liberal Theory’,
Williamson, O. (1993) ‘Calculativeness, Trust, and Economic Organisation’, Journal of
Law and Economics, 36: 453–86.
Wilson, D. J. and Game, C. (2011) Local Government in the United Kingdom. Fifth
edition. Houndsmill, Basingstoke: Palgrave Macmillan.
Wistow, G., Knapp, M. R. J., Hardy, B. and Allen, C. (1994) Social Care in the Mixed
Economy. Buckingham: Open University Press.
Wistow, G., Knapp, M. R., Hardy, B., Forder, J., Kendall, J., and Manning, R. (1996)
Social Care Markets: Progress and Prospects, Buckingham: Open University Press.
Wittgenstein, L. (1972) Philosophical Investigations. Translated by G. Anscombe.
Oxford: Basil Blackwell.
Wodak, R. (2011) The Discourse of Politics in Action: Politics as Usual. Second revised
edition. Basingstoke: Palgrave Macmillan.
World Bank (1992) Governance and Development. Washington, DC: The World Bank.
Wright, D. S. (1974) ‘Intergovernmental Relations: An Analytical Overview’, The
Annals of the American Academy of Political and Social Science, 416: 1–16.
Wright, D. S. (1978) Understanding Intergovernmental Relations. North Scituate, MA:
Duxbury Press.
Wright, D. S. (1983) ‘Managing the Intergovernmental Scene: The Changing Dramas
of Federalism, Intergovernmental Relations and Intergovernmental Management’.
264 References
In W. B. Eddy (ed.), Handbook of Organisation Management. Berlin: de Gruyter,
pp. 417–54.
Wright, M. (1988a) ‘Policy Community, Policy Network and Comparative Industrial
Policies’, Political Studies, 36: 593–612.
Wright, M. (1988b) ‘The City Rules, OK’, Public Administration, 66: 389–410.
Wright, M. (1991) ‘The Comparative Analysis of Industrial Policies: Policy Networks
and Sectoral Governance Structures in Britain and France’, Staatswissenschaften
und Staatspraxis, 4: 503–33.
Wright, V. (1994) ‘Reshaping the State: Implications for Public Administration’, West
European Politics, 17: 103–34.
Wright, V. and Hayward, J. E. (2000) ‘Governing from the Centre: Policy Coordin-
ation in Six European Core Executives’. In R. A. W. Rhodes (ed.), Transforming
British Government. Volume 2. Changing Roles and Relationships. London:
Yanow, Dvora (1996) How Does a Policy Mean? Interpreting Policy and Organizational
Actions. Washington DC: Georgetown University Press.
Yeatman, A. (1994) ‘The Reform of Public Management: An Overview’, Australian
Journal of Public Administration, 53: 287–95.
Yin, R. K. (1984) Case Study Research: Design and Methods. London: Sage.
Young, H. (1989) One of Us. London: Macmillan.
Young, H. and Sloman, A. (1986) The Thatcher Phenomenon. London: BBC.
Ysa, Tamyko and Esteve, M. (2013) ‘Networks Never Walk Alone: The Governance of
Network Portfolios’. Paper to the Global Governance Club, Netherlands Institute
for Advanced Study in the Humanities and Social Science (NIAS), Wassenaar, The
Netherlands, 29 May to 1 June.
Author Index
6, Perri 47, 111–12, 192 Berman, M. 56

Bernstein, R. 224
Abélès, M. 72 Berrill, K. 143
Agranoff, R. 43, 47, 74, 82, 164, 169, 184 Berrington, H. 140
Alderman, R. K. 139, 145 Berry, J. M. 40
Aldrich, H. 22 Bevir, M. 5, 10, 42n4, 49, 51, 55n13, 71,
Alexander, E. R. 184 101n6, 105, 108, 111, 115, 115n7, 134, 184,
Alford, R. R. 23 199n1, 205, 206, 208, 218, 220n4, 221, 222,
Allison, G. 140 224, 224n5, 224n6
Anderson, J. J. 48n10, 61–2 Bichard, Michael 128
Anderson, W. 19–20 Biddiss, M. 142
Andeweg, R. 207 Birch, A. H. 174
Ansell, C. 42n4, 46, 48n10, 82, 84, 213 Blackstone, Tessa 143
Armajani, B. J. 161n4 Blair, Tony 132, 135, 208–11
Armstrong, William 143 Blau, P. M. 44n7
Atkinson, M. M. 42, 58 Blick, A. 137n2
Atkinson, P. 100 Blunkett, David 201
Blyth, M. 133
Bache, I. 4, 57n1, 69n13, 70–1 Bobrow, D. B. 218
Bagehot, W. 200 Bogason, P. 169
Bakviss, H. 10 Bogdanor, V. 80n2
Bale, T. 133 Boje, D. 218
Ball, D. 146 Bomberg, E. 69n14, 71
Balme, R. 61 Börzel, T. A. 39, 42n5, 43
Bang, H. 51–2, 108–10, 110n5, Bouckaert, G. 135
112, 192 Boudon, R. 85
Barber, B. 186 Bovens, M. 42, 53, 78, 168
Barber, J. 140 Bowman, A. 133, 134, 191
Bardach, E. 47n9 Bradbury, J. 191
Barnes, J. 16–17, 144 Braithwaite, J. 53
Barnett, M. 154 Brock, G. 142
Barrett, S. 24 Brown, Gordon 79, 132, 208–9
Barthes, R. 205, 220n4 Brown, R. G. S. 5
Barzelay, M. 161n4 Bruce-Gardyne, Jock 154
Bate, S. P. 206 Bryce, J. 200
Batley, R. 60n9 Bulmer, S. 58, 64, 67
Battista, R. 176 Burch, M. 142, 146, 156–7, 207,
Baumgarten, F. R. 40 210–11
Bealey, F. 4 Bushnell, R. A. 128
Beer, S. H. 20 Butler, D. 135, 177
Behn, R. D. 54n12 Butler, Sir Robin 8, 126, 129, 155
Bell, S. 212n2
Bellamy, C. 156 Cairney, P. 35
Benn, Tony 139, 149 Callaghan, James 144, 146, 148
Bennington, J. 63, 64, 65, 203 Callon, M. 37
Bennister, M. 209, 210–11 Cameron, David 132, 133, 136
Benson, J. K. 5, 21–2, 25, 41n3 Carmichael, P. 200
Bentham, J. 103 Carpenter, D. P. 47n9
Berger, P. 104 Carter, N. 139, 145
266 Author Index
Cartwright, D. 16 Dylan, Bob 3, 202
Castells, M. 37 Dyson, K. 23, 113
Castle, Barbara 149
Cawson, A. 23, 25, 42 Eggers, W. D. 74, 82, 215
Chamberlain, Neville 143 Eising, R. 70
Chandler, D. 133 Elgie, R. 10, 156, 207–8, 210
Chapman, R. 4 Elgin, D. S. 128
Checkel, J. T. 72 Elkin, S. L. 18, 41
Cherwell, Lord 143 Emerson, R. E. 44n7
Chester, N. 3–4, 5 Esteve, M. 81
Child, J. 45n8 Evan, W. M. 16, 18, 65
Chisholm, D. 54 Eymeri-Douzans, J-M. 10, 156
Christiansen, T. 72
Chubb, B. 4 Faulkner, D. 45n8
Churchill, Winston 143 Faust, K. 41
Clarke, M. 122 Fawcett, P. 220
Clifton, J. 134 Fenno, R. E. 52, 100, 101–2
Cockerell, M. 143 Ferlie, E. 45
Colebatch, H. K. 211 Fernandez, R. M. 47n9
Coleman, W. D. 39n2, 42, 58 Finer, S. E. 142, 158
Coleridge, Samuel Taylor 103 Finlayson, A. 224n5
Collingwood, R. G. 104, 104n2, 199 Fischer, F. 224n6
Considine, M. 39n2, 49, 53n11 Fleming, J. 173n1, 190, 193–5
Craig, G. A. 188 Flinders, M. 70
Cram, L. 72 Florestano, P. S. 169
Crewe, I. 79, 135, 191 Flynn, R. 53, 78, 178–80, 184n5, 185, 186–7
Cribb, J. 132 Foley, M. 137n2, 140, 142
Crosbie, P. V. 16 Ford, D. 45n8
Crossman, R. H. S. 139, 149 Forester, J. 224n6
Crothers, L. 192 Foster, C. D. 25
Crowther-Hunt, Norman 149 Fox, A. 75, 187
Crozier, M. 5, 18, 22, 41n3 Fox, C. J. 171
Czarniawska, B. 218n3 Frances, J. 165, 186
Franklin, G. 19, 34, 37
Davis, G. 39n2, 212 Frederickson, H. G. 43, 132n4
Davis, Miles 202 Freeman, J. L. 19, 37, 38
Dearlove, J. 22 Friedberg, E. 20, 22
de Callierès, F. 76, 86, 188 Friedland, R. 23
Degoey, P. 186 Friend, J. K. 164, 169
Denhardt, J. V. 82 Fudge, C. 24
Denhardt, R. B. 82
Denning, S. 84, 218n3 Gabriel, Y. 55, 55n13, 218n3
De Rynck, S. 61 Gaebler, T. 42, 121, 130, 161
Diamond, P. 56 Galligan, B. 48
Dicey, A. V. 200 Gamble, A. 64, 173, 174, 175
Dietz, K. 218n3 Game, C. 190
Diez, T. 72 Gardner, A. 190
Donoughue, Bernard 146, 148 Gash, A. 46, 82, 84
Dorey, P. 56 Gash, T. 133
Dowding, K. 35, 39, 41, 45, 47, 48, 51, 63, Geertz, C. 51, 86, 87, 109, 113, 202, 206
63n11, 156, 202, 203, 204, 211 George, A. L. 188
Dryzek, J. S. 218 George, S. 4, 57n1, 69n13
Dunleavy, P. 21, 23, 31, 137n1, 145, 151–3, Gergen, K. J. 104
153n5, 154, 161n4, 206–7 Giddens, A. 17, 192n10
Dunsire, A. 7, 9, 147, 160n3, 161n5, 162n7, Gillard, Julia 210
166n9, 170 Goldsmith, M. J. 20, 24, 33
Author Index 267
Goldsmith, S. 74, 82, 215 Hummel, R. P. 55, 55n13
Goodin, R. E. 37n1, 43n6 Hunter, D. J. 26, 32
Goodsell, C. T. 130 Huxham, C. 46
Goss, S. 82
Gould, R. V. 47n9 Ingham, Bernard 143
Granovetter, M. 186 Inglis, F. 224
Grant, W. P. 27, 28–9, 31, 32, 33, 58n5, 65–6 Iremonger, L. 140
Gray, A. 147
Gray, C. 26 Jachtenfuchs, M. 70
Greenleaf, W. H. 173 Jackson, M. W. 129
Greenstein, F. I. 140 Jackson, P. M. 24
Gretschmann, K. 172n11 James, M. 24
Gunn, L. 5 James, S. 143, 155
Gyford, J. 24 Janis, I. 144
Jeffery, C. 70
Haas, P. M. 37, 48, 58n4, 70 Jenkins, W. I. 147
Haines, J. 143 Jenkins-Smith, H. C. 50
Hajer, M. A. 52, 224n6 Jensen, L. 109
Ham, C. 26 Jessop, B. 25, 53, 110, 132n4, 165n8, 212n2,
Hammersley, M. 51, 100, 109, 206 214, 215, 216
Hancher, L. 27, 29 Johnson, N. 175
Handy, C. B. 16 Johnson, P. 133
Hanf, K. 18, 20, 22, 41n3, 164 Jones, B. 60n9
Hardy, B. 100n5 Jones, G. W. 3, 137n2, 139, 143, 145, 147
Harrison, S. 123 Jordan, A. 70, 72, 212n2
Harvey, J. 63, 64, 65, 203 Jordan, G. 17, 19, 23–4, 25, 26, 31, 32, 33, 35,
Hay, C. 51 37, 39n2, 42, 128–9
Hayward, J. E. S. 54, 224, 199n1 Josselin, D. 48n10, 58n5, 65
Haywood, S. 32 Jouve, B. 61
Heady, B. 137 Judge, D. 53
Heath, Edward 143, 150
Heclo, H. 6, 8, 17–18, 24, 26, 32, 37, 40, 51, Kahn, R. L. 16
109n4, 137, 145 Karpik, L. 18
Heffernan, R. 209–11 Kassim, H. 48n10, 64, 65, 71, 72, 203
Heilman, S. 10, 156 Katsaris, A. 72
Heimeriks, K. H. 82 Katz, D. 16
Heisler, M. 19, 42 Kaufman, F. X. 20, 23, 42, 170
Hennessy, P. 8–9, 76, 125, 125n3, 135, 137, Kaufman, Gerald 128
137n2, 141, 143, 173, 208 Kavanagh, D. 105, 121n2, 160
Henney, A. 30, 111 Keating, M. 60n9
Heseltine, Michael 125n3, 142 Keck, M. E. 48
Hesse, J. J. 127 Keeling, D. 187
Hindmoor, A. 212n2 Keliher, L. 148
Hirst, P. 172n11 Kellner, P. 149
Hoffman, S. 58n4, 62 Kenis, P. 44
Hogwood, B. W. 53, 144 Keohane, R. O. 34, 44n7, 48, 54n12, 58n4
Holden, H. 190 Kettl, D. F. 43, 74, 82, 167
Holliday, I. 49, 156–7, 207, 211 Keynes, S. 132
Hood, C. 5, 6, 10, 11, 124, 129, 135, 147, Kickert, W. 43, 43n6, 45, 46, 74, 82, 164, 165,
152, 161 165n8, 181, 183n4, 184
Hooghe, L. 57n3, 61, 70 King, A. 79, 135, 140, 142, 176n2, 191
Houlihan, B. 24 Kingdom, J. 137
Howard, John 209–10 Kingdon, J. W. 52
Howe, Geoffrey 140n3, 142 Kjær, A. M. 42, 212n2
Hudson, J. 154 Klijn, E-H. 11, 35, 42n5, 46, 74, 82, 132n4,
Hughes, O. 211 169, 181, 183n4
268 Author Index
Knoke, D. 41 March, J. G. 185
Kogan, M. 19 Margach, J. 143
Kohler-Koch, B. 70 Marin, B. 39n2
Koliba, C. 55, 214 Marinetto, M. 9, 49, 199, 224
Kooiman, J. 162–3, 162n7, 181, 183n4 Marks, G. 59–60, 67, 68, 70
Koppenjan, J. 35, 42n5, 46, 74, 82 Marsh, D. 23, 24, 25, 33, 35, 39, 45, 49, 50–1,
Kramer, R. M. 75, 184n5 63n11, 64, 65, 66, 106, 151, 156, 164, 171,
Krasner, S. D. 34 180, 203–4, 205, 212n2
Marshall, G. 146
Laffan, B. 61 Marwick, A. 104n2
Laffin, M. 24 May, A. 143
Landau, M. 127 Maynard-Moody, S. 192
La Porte, T. R. 43n6 Mayntz, R. 11, 39n2, 44
Larson, A. 164, 184n5, 185, 186 Mazey, S. 48n10, 59, 68
Laumann, E. 41, 65 McAnulla, S. 224n5
Law, J. 37 McCaffrie, B. 35
Lawson, Nigel 142, 148, 154 McCall, M. W. 140
Lawton, A. 140 McFarland, A. 23, 40
Lee, M. 143 McGuire, M. 82
Leech, B. L. 40 McPherson, A. 51, 109n4
Leftwich, A. 162, 162n6 McVicar, M. 53
LeGales, P. 39, 39n2 Meier, K. J. 43
Lehmbruch, G. 42 Menon, A. 199n1
Leigh, D. 142, 154 Metcalfe, L. 125, 164
Leonardi, R. 60n9 Miliband, Ed 134
Lewis, D. 58 Mill, J. S. 103
Lewis, J. 53n11 Miller, H. T. 171
Lilley, Peter 128 Mills, M. 63n11, 64
Lindblom, C. E. 54 Milward, H. B. 56n14, 132n4
Lindquist, E. A. 39n2 Minogue, K. 142
Ling, T. 80n2 Mitchell, J. 60n9, 190
Linklater, M. 142, 154 Mitchell, J. C. 17
Linn, N. 37 Moodie, G. C. 146
Linstead, S. 206 Moran, M. 27, 29, 37n1
Lipsky, M. 83, 192 Morgan, G. 84, 165n8, 218
Loader, I. 48 Morphet, J. 135
Lodge, M. 135 Mottram, Sir Richard 122
Lorsch, J. W. 16 Mugabe, Robert 148
Loughlin, J. 112, 113 Mulgan, R. 53, 54, 79
Loughlin, M. 174 Musheno, M. 192
Low, S. 158n2
Lowi, T. 19, 38, 171 Nanetti, R. Y. 61
Lowndes, V. 190 Naughtie, J. 208
Luckman, T. 104 Newman, J. 199, 212n2
Luhmann, N. 42, 43n6, 163, 165n8, 173 Nicholson, H. 76, 187–8
Nielsen, K. 42
MacIntyre, A. 49, 174, 224 Nordlinger, L. 23, 31, 216
Mackenzie, W. J. M. 3–4, 5 Norton, A. 124
Mackie, T. T. 144 Norton, P. 137, 139, 175, 207
Mackintosh, J. P. 137, 139 Nugent, N. 57
Macmillan, Harold 207 Nye, J. S. 34, 44n7
Madgwick, P. 139
Major, John 42 Oakeshott, M. 104, 157
Mandell, M. P. 39n2 Offe, C. 170
Manzer, R. A. 19 O’Leary, B. 23
Marando, V. L. 169 Olsen, J. P. 11, 19, 42, 185
Author Index 269
Osborne, D. 42, 121, 130, 161 Rokkan, S. 34, 42
Osborne, George 191 Rorty, R. 202, 224
Ostrom, E. 30 Rosamond, B. 69n14
O’Toole, L. 39n2, 43 Rose, A. 140
Rose, R. 137n2, 145, 147
Page, E. C. 20, 33 Rosenau, J. 48, 163–4, 171
Painter, C. 46 Rosenau, P. M. 104
Pappi, F. U. 41, 65 Rowan, K. 143
Parkin, F. 31 Rudd, Kevin 209–10
Parsons, W. 50, 55 Ryan, M. 48
Paterson, W. 58n5
Pedersen, O. K. 42 Sabatier, P. 50, 52, 63
Perrow, C. 22 Salamon, L. M. 43, 46, 54, 217
Peston, R. 208 Salancik, G. R. 44n7, 45n8
Peters, B. G. 9, 43, 54, 58–9, 60, 64, 112, 113, Sartori, G. 209
119n1, 132n4, 172, 212, 212n2, 213 Saunders, P. 22, 26
Peterson, J. 48, 61, 62, 64, 65, 67, 68, 69–70, Saward, M. 33, 63n11, 64, 212n2
69n13–14, 71, 72 Sbragia, A. 72
Pettigrew, A. 45 Scharpf, F. W. 11, 18, 20, 41n3, 44–5, 58n5,
Pfeffer, J. 44n7, 45n8 123, 164, 203, 212
Piattoni, S. 69n14, 70 Schein, E. H. 16
Pierre, J. 42, 43, 172, 212, 212n2, 213 Schimmelfenning, F. 72
Plowden, W. 143 Schmitter, P. C. 24, 42
Poguntke, T. 208 Schneider, V. 44
Pollack, M. A. 203 Schout, A. 70, 72
Pollitt, C. 6, 11, 80n2, 123, 126, 135, 161 Scott, J. 17, 37, 41
Polsby, N. 101n7 Sedgemoor, B. 149
Powell, W. 53n11, 75, 76, 164, 184n5, 185, 186 Seidman, H. 80
Prescott, John 79 Self, P. J. O. 5
Pressman, J. 83 Selznick, P. 5
Preston, C. 58n5 Seymour-Ure, C. 141
Provan, K. G. 132n4 Shamir, B. 84
Putnam, R. A. 60n10 Sharpe, L. J. 24, 60n9
Shore, C. 72, 206
Raab, C. 51, 109n4 Sikkink, K. 48
Raab, J. 56n14 Silverman, L. L. 218n3
Raco, M. 133, 191 Simon, H. A. 43n6, 53n11
Ranson, S. 24, 125 Skelcher, C. 212n2
Rawnsley, A. 208 Skinner, Q. 104n2
Redford, E. S. 19 Skogstad, G. 39n2
Rein, M. 37n1, 55n13, 219 Slaughter, A-M. 48
Richards, D. 51, 55, 56, 212n2 Sloman, A. 142
Richards, S. 125, 164 Smircich, L. 84
Richardson, J. J. 19, 23–4, 26, 31, 32, 33, 35, Smith, M. J. 45, 50–1, 55, 56, 58n5, 63n11,
37, 39, 39n2, 47, 48n10, 49, 52, 59, 68, 139, 175, 203–4, 207, 212n2
135, 205 Sørensen, E. 51–2, 108–10, 110n5, 112,
Richelson, J. T. 146 192, 214
Ricoeur, P. 220n4 Stanworth, P. 17
Riddell, Peter 200 Stepan, M. 10, 156
Ridley, F. F. 7 Stevens, J. P. 19, 37, 38
Rip, A. 37 Stewart, J. 4, 5, 122, 125
Ripley, R. 19, 34, 37, 38 Stoker, G. 7, 42n4, 60n9, 84, 110,
Risse, T. 48, 72 113, 114, 159, 176, 190
Robinson, Geoffrey 79 Stone, D. 224n6
Robson, W. A. 3–4, 5 Stones, R. 65
Roe, E. 224n6 Strange, S. 34
270 Author Index
Tarrow, S. 21, 31 Wamsley, G. 130
Taylor, A. 47n9, 182–3, 212n2 Wanna, J. 80
Taylor, C. 224 Wapshott, N. 142
Tetlow, G. 132 Ward, H. 149
Thain, C. 147 Wass, Douglas 76, 145
’t Hart, P. 11 Wasserman, S. 41
Thatcher, M. 39, 39n2, 42, 48 Wassmer, U. 81
Thatcher, Margaret 6, 9, 42, 121n2, 122, 133, Watson, A. 76, 188, 188n7
139–40, 141–2, 143, 144, 145, 146–7, 148, Webb, C. 58
150, 210 Webb, P. 208
Theakston, K. 146 Weick, K. E. 55n13, 218
Thoenig, J-C. 5, 11, 18, 20, 41n3 Weller, P. 10, 143, 174, 199
Thompson, G. 53n11, 76, 164, 184, Wettenhall, R. 47
184n5, 185 White, H. 104, 104n2, 220n4
Thompson, J. D. 5, 18, 41n3 White, L. G. 169
Timmins, N. 135 Whitson, C. 58n5
Tivey, L. 173 Whitten, N. E. 16–17, 65
Torfing, J. 35, 54, 71, 172, 212, 212n2, Wildavsky, A. 6, 8, 17, 24, 32, 51, 83, 109n4,
213, 214 126, 131–2, 137, 145
Toynbee, P. 133 Wilks, S. 27–8, 29, 30, 32, 33
Travers, T. 128, 190 Williams, D. 162
Tricker, R. I. 160 Williams, Marcia 143
Trollope, Anthony 4 Williamson, O. 185, 186
Truman, D. B. 17 Wilson, D. J. 190, 191
Tyler, T. 75, 184n5, 186 Wilson, Harold 121, 141, 143
Wilson, Horace 143
Vadera, Shriti 79 Wilson, Sir Richard 135
Van Eeten, M. J. G. 55n13 Winch, P. 224
Vangen, C. 46 Wistow, G. 26, 32, 123, 180
Van Gunsteren, H. 42, 53 Wittgenstein, L. 222
Van Maanen, J. 206 Wodak, R. 72
Van Waarden, F. 40 Wolfe, A. W. 16–17, 65
Vare, D. 187 Wright, D. S. 20, 48, 169
Vickers, Sir Geoffrey 107n3, 162n7, 168 Wright, M. 27, 28, 29, 32–3, 67, 147
Vinzant, J. 192 Wright, V. 9, 20, 33, 54, 167
Wagenaar, H. 110n5, 224n6 Yanow, D. 224n6

Waldegrave, William 125 Yeatman, A. 161n4
Walker, D. 133 Yin, R. K. 154
Walker, P. G. 143 Young, A. R. 203
Wallace, H. 34, 58, 69n14, 203 Young, H. 141, 142
Wallace, W. 34, 58, 61, 149 Young, T. 162
Walsh, K. 122, 125 Ysa, T. 81
Walters, Alan 142, 143
Walters, W. 212n2 Zander, A. 16
Subject Index
accountability obstruction by 149–50

bureaucracy and 172 politicization of 125–6
hollow state and 120, 128–9, 131 reform of 119, 147
ministerial accountability 154 storytelling in 218–19
in the National Health Service 180 civil society 42, 84, 112, 170, 212, 215,
new public management and 114, 125, 160 216–17, 221
policy networks and 53–4, 78–9, 128, collaboration 43, 84, 179–80. See also
171–2, 180 cooperation
prime ministerial cliques and 144 Committee of Public Accounts
actions 103, 105, 106, 107, 108, 113, 115, 205, (CPA) 180
211, 216, 220, 222, 223 communitarianism 115
actor-centred institutionalism 44 Community Health Services (CHS) 178–80,
advocacy coalition framework (ACF) 50 184, 189
agency 206. See also situated agency Community Support Framework (CSF) 61
AIDS policy 176, 184 competition 78, 114, 159, 168, 179–80, 183,
American political science 6, 29, 30 201. See also marketization; new public
Anglo-Governance School 9, 49, 199, 224. See management (NPM)
also differentiated polity complexity 43n6
Anglo-Saxon tradition 112 constructivism 72–3
anti-foundationalism 103–16 contracting out 114, 115, 121–2, 133–4, 176,
Australia, whole-of-government approach 181, 191
in 79–80 erosion of trust by 189
in the National Health Service 78, 122,
beliefs 101, 103, 105, 106, 107, 108, 109, 178–80
110, 113, 115, 205–6, 211, 216, 220, in the police 193–4
221, 222, 223 control 42
Benefits Agency 159, 183 cooperation 53, 78, 165, 179, 201. See also
bottom-up initiatives 83 collaboration
Brexit 134, 135, 201 coordination 42, 54, 62, 79–80, 127, 129, 135,
bureaucracy 130–1, 159, 161, 170, 172, 183, 138, 207
193, 212. See also hierarchies core executive 10, 49, 137–8, 150–7, 199–200,
bureaucratic coordination model 149–50, 156 206–11
bureau-shaping model 151–3 corporate governance 160
corporate pluralism 42
Cabinet government 137, 138, 139, 144–5, corporatism 24–5, 42
150, 156 court politics 211
central capability 120, 129, 131, 135–6,
156, 157 dark networks 56n14
centralization 81–6, 112, 190, 200, 201 decentred approach 51–2, 87, 103–16, 205–6,
central–local government relations 23, 211, 216–24. See also interpretive
24–7, 33 approach
change 49–52, 106, 113–15, 205–6 Denmark 51–2, 108–10, 112
Chartered Institute of Public Finance and Department of National Heritage
Accountancy (CIPFA) 160 (DNH) 182–3, 184–5
Citizen’s Charter 126 dependence 177
civil service differentiated polity 7, 70, 150, 173–4,
employment cuts in 121 175–6, 199–216, 221–4. See also core
interview with permanent secretary 96–9, executive; hollowing out; network
100–1 governance; policy networks
272 Subject Index
dilemmas 105, 106, 107, 113–15, 176–7, Fulton Committee 119
178–80, 181–3, 205, 221, 223 functional differentiation 5
diplomacy 76, 173, 187–9, 215
game playing 178–80
Economic and Social Research Council game theory 44, 45
(ESRC). See also Social Science Research Germanic tradition 112
Council (SSRC) GIR. See government–industry relations
Programme on Central and Local (GIR)
Government Relationships 7 global governance 48
Whitehall Programme 7–10, 156, good governance 160, 162
166n10 gossip 219
Efficiency Unit 122, 123, 143 governance. See network governance
entrepreneurial government 161 governance narrative 199. See also
epistemic communities 37, 48 differentiated polity
ERDF. See European Regional Development government–industry relations (GIR) 23,
Fund (ERDF) 27–31, 32, 33
ESRC. See Economic and Social Research Greater London Council (GLC) 128
Council (ESRC) group dynamics 15–16
ethnography guidance 42
guidelines for 101–2
interview with chair of a primary care hierarchies 42, 43, 159, 190, 212. See also
trust 93–6, 100–1 bureaucracy
interview with permanent secretary 96–9, holistic government 111, 221
100–1 hollowing out 10, 119–36, 156, 167,
social workers’ cases 87–92, 100, 101 199–200, 212
usefulness of 100, 206
European Exchange Rate Mechanism 132 IGM. See intergovernmental management
European Regional Development Fund (IGM)
(ERDF) 59–60, 61–2 IGR. See intergovernmental relations (IGR)
European Union (EU) 57–73 implementation 62, 63
accountability in 128 inflation 113–14
Brexit 134, 135 Institute of Local Government Studies
Community Support Framework 61 (INLOGOV) 4, 5
impact on British government 120, 123–4, institutional approach to network
136, 148–9 management 46
intergovernmental relations in 59, 59n7, institutional complexity 176–7
60–1 institutional void 52–5, 77
multi-level governance in 59, 59n7, 60–1, institutions 5, 64–5, 103, 205
63, 67, 70–1 instrumental approach to network
policy communities in 58 management 45, 46
sovereignty and 123, 134–5 interaction approach to network
subsidiarity and 124 management 45–6
transnational networks in 48 interdependence 153–4, 184, 187–8,
Everyday Maker 51–2, 108–10, 112 199, 201
exchange theory 5 interest groups 59
executive 137–8. See also core executive interest intermediation 38–40
expertise 217 interests 106, 107
intergovernmental management
Financial Management Initiative 147 (IGM) 168–70, 183, 190
flexibility 43 intergovernmental networks 26
fragmentation 43, 49, 78, 79, 111, 114, 120, intergovernmental relations (IGR) 19–21, 22,
126–7, 129, 135, 170, 176, 181–2, 191, 59, 59n7, 60–1
201, 215 interorganizational analysis 18, 40–2
franchise state 133–4, 191. See also interpretive approach 35, 72, 103,
contracting out; hollowing out 105, 202, 218, 224. See also decentred
French tradition 112 approach
Subject Index 273
iron triangles 17, 18, 19, 37, 38. See also as professionalized policy network 26, 32
subgovernments purchaser–provider split 114, 122–3,
issue networks 17–18, 26, 37, 39, 40, 48 178–80
reform of 119
joined-up government 42, 45, 99, 115, 135, trust in 53
156, 221 negotiated economy 42
neo-liberalism 36, 114–15, 133–4, 136, 184,
liberal tradition 111 191, 201, 213–14, 222
local government neo-pluralism 22–3, 30
central–local government relations 23, network governance. See also differentiated
24–7, 33 polity; policy networks; steering
devolution to local councils 191 characteristics of 165–6, 176–89
employment in 160 criticisms of 211–16
public spending by 177, 184 decentred approach to 103–16, 205–6,
reform of 119 222–3
service delivery in 122 meanings of 9, 158–72, 200, 201, 222–3
local networks 217 self-organizing networks 164–72, 180, 183,
London Underground 79 213, 223
LSE Public Policy Group 172 sour laws of 52–5, 77–81, 85–6
ten commandments of 82
macro-level analysis of policy networks 21–3, network management 43, 45–7, 71–2, 74–86.
25, 29, 30, 31, 64 See also policy networks
managerialism 6, 7, 115, 125, 151, 161, 167–8, new institutional economics 161
184, 193–4. See also new public new institutionalism 44, 115
management (NPM) New Labour 42, 111, 115, 132, 134,
marketization 6, 42, 78, 111, 114, 115, 173, 135–6, 222
185, 199. See also contracting out; new new public management (NPM) 6, 113–15,
public management (NPM) 120, 124–6, 129, 135, 151, 160, 161,
markets 159, 170. See also quasi-markets 167–8
Maud Committee 119 Next Steps 122, 123, 125, 143, 147
meanings 101, 103, 108, 110, 205, 206 NHS. See National Health Service (NHS)
meso-level analysis of policy networks 18–21, NHS and Community Care Act 122–3
25, 27, 31, 39, 62
metagovernance 214–16, 217 objectivity 105, 106
micro-level analysis of policy oral administrative history 154
networks 15–17, 24, 29, 32, 39 organizational glue 83
minimal state 159–60, 165, 191. See also neo- organizational studies 6
liberalism organization theory 5
ministerial accountability 154
ministerial government 145–7, 150, 156 participation 171
MLG. See multi-level governance (MLG) partnerships 42, 43. See also public–private
modernist-empiricist tradition 5, 35, 49, 87, partnerships (PPPs)
106, 202, 211, 215, 216 performance measurement 6, 78
multi-level governance (MLG) 59, 59n7, personal networks 24, 32
60–1, 63, 67, 70–1 police reform 193–5
policy analysis 6
narratives 55, 87, 104, 105, 107, 110–13, 116, policy communities 23, 24, 26, 28–9, 32–3,
220, 224. See also storytelling 35, 37, 39–40, 58. See also policy
National Health Service (NHS) networks
accountability in 180 policy levers 182–3, 189
Community Health Services policy networks 5, 6–7, 199–200. See also
(CHS) 178–80, 189, 184 network governance; network
contracting out in 78, 122, 178–80 management
employment in 121, 127, 160 accountability of 53–4, 78–9, 128,
interview with chair of a primary care 171–2, 180
trust 93–6, 100–1 boundaries of 65
274 Subject Index
policy networks (cont.) public service ethic 96, 114, 125, 160
British contributions to study of 23–34 public spending 120, 133, 147, 150, 160, 177,
building of 77, 95, 100–1 184, 191
change and 49–52, 106, 113–15, 205–6 purchaser–provider split 114, 122–3, 159,
characteristics of 74–6, 183–9 178–80, 181, 183
comparison of 47–9
context of 204–5 quantitative methods 154, 156
criticisms of 202–6 quasi-markets 114, 115, 122, 123, 159,
definitions of 37, 66–7, 101 181, 183
as description 38–43
institutional void in 52–5 rational choice 44–5, 203
institutions and 64–5 reciprocity 76
as interest intermediation 38–40 Redcliffe-Maud report 119
intergovernmental relations in 19–21 resource exchange 184
interorganizational analysis of 18, 40–2 responsiveness 43
issue networks in 17–18, 37, 39, 40, 48 Rhodes model 24–7, 30–4, 62, 64, 65, 69–70
limits to 66–9 rowing 42–3, 54–5, 161, 212. See also steering
macro-level analysis of 21–3, 25, 29, 30,
31, 64 Scandinavian tradition 112
meso-level analysis of 18–21, 25, 27, 31, Scottish referendum 201
39, 62 segmented decision model 147–9, 150, 156,
micro-level analysis of 15–17, 24, 29, 206–7
32, 39 segmented state 42
neo-pluralist theory for 22–3, 30 self-organizing networks 164–72, 180,
participation in 171 213, 223
in the police 194–5 service delivery 53, 78, 111, 115, 119, 121–3,
policy levers in 182–3 128, 131, 164, 170, 181–2, 215. See also
political economy of 21–2, 30 contracting out; fragmentation
Rhodes model 24–7, 30–4 situated agency 206, 211, 216, 222, 223
social network analysis of 16–17 social anthropology 16
as theory 43–5 social constructivism 104–5, 108
types of 39 social interests 31
policy outcomes 66 socialist tradition 111–12
policy studies 5 social network analysis 16–17, 37
political economy 21–2, 30 social networks 95, 98–9
poll tax 177 social psychology 15
positivism 105 Social Science Research Council (SSRC).
power-dependence 43–4, 148, 153, 201, 203, See also Economic and Social Research
207, 209 Council (ESRC)
practices 101, 103, 105, 205, 211, 221, 223 initiative on central–local government
preferences 103, 205 relations 23, 24–7, 33
presidentialism 208–9 initiative on government–industry relations
prime ministerial cliques 142–4, 156 (GIR) 23, 27–31, 32, 33
prime ministerial government 141–2, 156 Panel on Central–Local Government
prime ministerial power 137–8, 139–42, 150, Relationships 4, 5
157, 208–11 Panel on Research into Local
privatization 119, 120–1, 132, 136, 159, 160, Government 4
176, 181, 183 social services
producer networks 26 quasi-markets in 122, 123
professional networks 26, 32, 58n4 social workers’ case studies 87–92,
public administration 3–4, 5 100, 101
public intervention 119, 120–1, 159–60 socio-cybernetics 162–4, 165
public–private partnerships (PPPs) 79, 191 sociology 18
public sector reform 45–7, 119–36, 147, 167. sour laws 52–5, 77–81, 85–6, 192–5, 201
See also hollowing out; new public sovereignty 123, 134–5
management (NPM) state. See also hollowing out
Subject Index 275
definitions of 216 Training and Enterprise Councils (TECs) 168
role of 211–14, 215 transnational policy networks 48, 123–4, 128
stateless 216 trust 32, 45, 53, 75, 159, 164–5, 166, 179, 184,
vs. civil society 42, 112, 170, 215, 216–17 185–7, 215
steering 42–3, 54–5, 161, 168–9, 181–3, 200,
212, 214–15, 216, 217. See also rowing unintended consequences 114, 126–9, 173,
storytelling 55, 84–5, 86, 116, 214, 217–20, 192–5, 199, 201
224. See also narratives
strategic planning 84 values 96, 186. See also public service ethic
subgovernments 17, 18–19, 29, 37, 38. voucher schemes 114
See also iron triangles
subsidiarity 124 water services 26
Wessex Regional Health Authority
territorial communities 26 180, 184
Thatcherism 120, 131 Westminster model 55, 173, 174–5, 199, 201,
thick descriptions 87, 100, 109, 206. See also 221, 222
ethnography Whitehall Programme 7–10, 156, 166n10
top-down initiatives 83 whole-of-government approach 45, 79–80
traditions 101, 105, 107, 110–13, 205–6, 211, wicked problems 111
221, 222 World Bank 162

2017 - Rhodes - Network Governance and The Differentiated Polity Selected Essays, Volume I

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2017 - Rhodes - Network Governance and The Differentiated Polity Selected Essays, Volume I

Uploaded by

Copyright:

Available Formats

NETWORK GOVERNANCE

This collection of essays is in two volumes. Volume I is a retrospective. It

David Levi-Faur (Hebrew University of Jerusalem)

List of Figures xiii

PART I. POLICY NETWORKS

PART II. GOVERNANCE

Appendix: Bibliography on the Anglo-Governance Debate 225

2.1 Approaches to networks 16

3.1 Types of policy networks 39

What a Long Strange Trip It’s Been

I WAS SO MUCH OLDER THEN

It is intimidating to look back over a 45-year career as a political scientist. Has

In the Beginning, 1970–1976

A Professional Political Scientist at Last, 1976–1988

the Public Administration Committee’s (PAC) Conference on 13–15 September

resources. The organization’s dominant coalition employs strategies within

As I began to explore policy networks, Margaret Thatcher was intent on

From Government to Governance, 1988–1998

so I had to be different. My fortunes hung by a slender thread and I am ever

Programme gave me the opportunity to branch out on my own and do

In 1996, I had a downbeat view of the state of my discipline. Hood (1999:

Policy Networks in Britain

THE L I TE RATURE ON NE TW ORKS

The ﬁrst problem is to provide an organizing format for the literature on

The Micro-level of Analysis

Levels of Micro Meso Macro

Figure 2.1 Approaches to networks.

Social Network Analysis

‘terminological confusion’ increased. As Whitten and Wolfe (1972: 729–32)

The Meso-level of Analysis

American policy-making to refer to ‘small groups of political actors, both

Inevitably, intergovernmental theory has weaknesses. First, as Tarrow

The Macro-level of Analysis

BRITISH CONTRIBUTIONS TO THE S TUDY

The primary objective of this section is to provide a summary of a growing

The Rhodes Model

Rhodes (1981, especially chs 1 and 5) provided an explicit application of

spokespersons for local government (Rhodes 1981: 111–25). Competition is

a cluster or complex of organisations connected to each other by resource-

change. It is argued that the ‘national government environment’ conditions the

The GIR Initiative

of policies, processes and industrial sectors to provide an empirical basis for

CONCLUSIONS: P ROBLEMS AND DEVELOPMENT

Third, cross-national studies cannot assume that national variations are

INTRODUCTION: THE UBIQUITY OF NETWORK S

THE L I TERA TU RE ON P OL I CY NE TWORKS

When describing government policy-making, the term ‘policy network’ refers

Networks as Interest Intermediation

Membership: Very limited number, some Large

A policy community has the following characteristics: a limited number of

Networks as Interorganizational Analysis

between individuals in those institutions. At its simplest, interorganizational

Policy Networks as Theory

There is a large theoretical literature on policy networks in Britain (see Rhodes

by power-dependence; that is, ‘any organisation is dependent on other orga-

permanent, rule-governed relationships. The agreed rules build trust and

Policy Networks as Reform

Focus Improving steering Cooperation Network arrangements

time-consuming, objectives can be blurred, and outcomes can be indeﬁnite.

DEBATES AND CHALLENGES

Describing and Comparing Networks

The notion of a policy network can be dismissed as mere metaphor. It is not a