Network Management and Governance in Policy Implementation The Case of Smoking Prevention Programs Susanne Hadorn Full Chapter

Network Management and Governance
in Policy Implementation: The Case of

Smoking Prevention Programs
Susanne Hadorn
Visit to download the full and correct content document:
https://ebookmass.com/product/network-management-and-governance-in-policy-impl
ementation-the-case-of-smoking-prevention-programs-susanne-hadorn/
INTERNATIONAL SERIES ON PUBLIC POLICY
Network
Management
and Governance in
Policy Implementation
The Case of Smoking
Prevention Programs
Susanne Hadorn
International Series on Public Policy
Series Editors
B. Guy Peters
Department of Political Science
University of Pittsburgh
Pittsburgh, PA, USA
Philippe Zittoun
Research Professor of Political Science
LET-ENTPE, University of Lyon
Lyon, France
The International Series on Public Policy - the official series of International
Public Policy Association, which organizes the International Conference
on Public Policy - identifies major contributions to the field of public
policy, dealing with analytical and substantive policy and governance issues
across a variety of academic disciplines.
A comparative and interdisciplinary venture, it examines questions of
policy process and analysis, policymaking and implementation, policy
instruments, policy change & reforms, politics and policy, encompassing a
range of approaches, theoretical, methodological, and/or empirical.
Relevant across the various fields of political science, sociology, anthro-
pology, geography, history, and economics, this cutting edge series wel-
comes contributions from academics from across disciplines and career
stages, and constitutes a unique resource for public policy scholars and
those teaching public policy worldwide.
All books in the series are subject to Palgrave’s rigorous peer review
process: https://www.palgrave.com/gb/demystifying-peer-review/
792492.
Susanne Hadorn
Network Management
and Governance in
Policy Implementation
The Case of Smoking Prevention Programs
Susanne Hadorn
KPM Center for Public Management
University of Bern
Bern, Switzerland
ISSN 2524-7301 ISSN 2524-731X (electronic)

International Series on Public Policy
ISBN 978-3-031-08807-0 ISBN 978-3-031-08808-7 (eBook)
https://doi.org/10.1007/978-3-031-08808-7
© The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer
Nature Switzerland AG 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the
Publisher, whether the whole or part of the material is concerned, specifically the rights of
translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information
in this book are believed to be true and accurate at the date of publication. Neither the
publisher nor the authors or the editors give a warranty, expressed or implied, with respect
to the material contained herein or for any errors or omissions that may have been made.
The publisher remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
This Palgrave Macmillan imprint is published by the registered company Springer Nature
Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Acknowledgments
I am highly grateful for the patient and continuous support by my supervi-

sor, Fritz Sager, whose expertise, pragmatism, and amicable supervision
were key to the mere existence of this book. A big thank you also goes to
my second supervisor, Daniela Christofoli, who took the time to review
my work and provided highly valuable comments to improve this book. I
also cannot thank enough my long-time research partner and friend,
Céline Mavrot, who has not only been an inspiration in terms of research,
but above all has provided invaluable moral support—often in the form of
a necessary pinch of (self-)irony. I would also like to thank my (former)
great team colleagues for their scientific support and the necessary and
always highly entertaining distraction, especially Lyn Pleger, Bettina
Stauffer, David Kaufmann, Stefan Wittwer, Markus Kaufmann, Eva
Thomann, Iris Stucki, Johanna Künzler, Michael Wicki, Caroline Schlaufer,
and all those whom I cannot mention by name now, but who rightly feel
addressed. Special thanks also go to Lisa Asticher and Michela Godenzi for
their great support in the final formatting and preparation of the manu-
script and to Sharon Berger, who advised me passionately in questions
about statistics. Additionally, I would like to specifically thank Oliver
Neumann, who spent weeks proofreading the original manuscript and
finally motivated me to submit it. Moreover, I would also like to mention
my family and friends, who have not only been a great moral support but
also shown me time and again what great things are waiting outside the
confines of my office.
Finally yet importantly, this book analyzes data collected as part of a
large-scale evaluation project of Swiss cantonal smoking prevention
v
vi ACKNOWLEDGMENTS
programs. Thus, of central importance to this book were the various actors
involved in this evaluation. First, the entire evaluation team, to whom I
would like to express my sincere thanks for the great cooperation and to
whom I am also indebted for allowing me to use the data for the analyses
presented here. Secondly, I would like to thank the Tobacco Prevention
Fund as well as all interview partners (including the respective network
managers and project managers) for the always fruitful cooperation in the
context of this evaluation.
Contents
1 Management
and Performance in Collaborative
Networks: A Review of the Literature and the
Contribution of This Book 1
2 Linking
Characteristics of Network Managers’
Work Context to Network Management and
Project-Level Output 27
3 A
Multi-method Approach to Analyze Network
Management and Policy Outputs 69
4 Swiss Smoking Prevention Policies 97
5 Connecting
Network Managers’ Work Contexts with
Network Management127
6 Linking
Network-Level and Project-Level Factors
to Policy Outputs183
7 Drivers
and Obstacles to Active Network Management
and Their Influence on Network Performance223
vii
viii Contents
8 How
to Improve Collaborative Policy Delivery
within Networks: Theoretical and Practical Insights237
Index255
List of Figures
Fig. 1.1 Research questions. (Notes: The numbers designate the

corresponding research questions) 3
Fig. 1.2 Schematic illustration of the policy delivery networks 6
Fig. 2.1 Basic model. (Notes: Arrows show expected causal effects. The
circled numbers designate the research questions) 29
Fig. 2.2 Evaluation model. (Source: Figure based on Sager et al. (2021),
adapted)47
Fig. 4.1 Interconnectedness of network managers 101
Fig. 4.2 Cantonal smoking prevention program of the canton of Uri 107
Fig. 4.3 Cantonal smoking prevention program of the canton of Zug.
(Notes: Please see Fig. 4.2 for an explanation of the illustration
and for the legend) 109
Fig. 4.4 Cantonal smoking prevention program of the canton of Zurich.
and for the legend. Projects P6 and P7 are co-managed by Züri
Rauchfrei and a cantonal working group active in the area of
youth protection) 110
Fig. 4.5 Cantonal smoking prevention program of the canton of
Solothurn. (Notes: Please see Fig. 4.2 for an explanation of the
illustration and for the legend) 111
Basel-Stadt. (Notes: Please see Fig. 4.2 for an explanation of the
Basel-Landschaft. (Notes: Please see Fig. 4.2 for an explanation
of the illustration and for the legend) 113
ix
x List of Figures
Fig. 4.8 Cantonal smoking prevention program of the canton of Jura.

Fribourg. (Notes: Please see Fig. 4.2 for an explanation of the
Fig. 4.10 Cantonal smoking prevention program of the canton of St.
Gallen. (Notes: Please see Fig. 4.2 for an explanation of the
Fig. 4.11 Cantonal smoking prevention program of the canton of Valais.
Fig. 4.12 Cantonal smoking prevention program of the canton of Vaud.
and for the legend. Projects P10, P11, and P12 were
co-managed by the CIPRET and the health agency) 120
Thurgau. (Notes: Please see Fig. 4.2 for an explanation of the
Fig. 4.14 Cantonal smoking prevention program of the canton of Ticino.
and for the legend. Projects P1, P3, P4, and P11 were
co-managed by the ASN and Radix) 122
Fig. 6.1 Actor constellation and intervention logic in the projects
‘Passive Smoke in Private Spaces’. (Notes: The dotted arrows
depict the direction of acquisition measures and the black
arrows display training and counseling measures. The gray
arrow illustrates the effect relationship between changes in the
target group behavior and the effects on the beneficiaries) 193
Fig. 6.2 Actor constellation and intervention logic in the projects
‘Public Relations’. (Notes: The dotted arrows depict the
direction of mobilization measures to involve partner
organizations and multipliers in the project. The black arrows
display information measures directly targeting the cantonal
population)199
Fig. 6.3 Actor constellation and intervention logic in the projects ‘Open
Youth Work’. (Notes: The dotted arrows depict the direction of
mobilization measures to involve partner organizations in the
project. The black arrows display substantive prevention
measures targeting adolescents through the youth centers) 204
List of Tables
Table 2.1 Possible configurations of four conditions characterizing

network managers’ work context 34
Table 2.2 Types of network management strategies 36
Table 2.3 Expected mechanisms between the four conditions and
network management 44
Table 3.1 Categorization grid‚ policy output 76
Table 3.2 Summary of the operationalization of conditions and variables 81
Table 3.3 Overview of the dataset 84
Table 4.1 Overview of 187 projects (N = 187)103
Table 5.1 Observed configurations of conditions 128
Table 5.2 Assignment of empirical cases to the theoretically possible
configurations129
Table 5.3 Summary of network management strategies applied in the
13 networks 133
Table 5.4 Overview about conditions and their expected influence on
the outcome 144
Table 5.5 Sufficient paths for occurrence and non-occurrence of active
network management 145
Table 5.6 Summary of findings ‘Solution Path I’ 148
Table 5.7 Summary of findings ‘Solution Path II’ 150
Table 5.8 Summary of findings ‘Solution Path III’ 153
Table 5.9 Summary of findings ‘Solution Path IV’ 155
Table 5.10 Summary of findings ‘Solution Path V’ 160
Table 5.11 Deviant cases for coverage 161
Table 5.12 Summary of findings ‘Contradictory and Unexplained Cases’ 165
Table 6.1 Models 1, 2, and 3 (Bayesian ordered logistic regressions) 185
xi
xii List of Tables
Table 6.2 Models 4 and 5 (Bayesian ordered logistic regressions) 188

Table 6.3 Model 6 and 7 (Bayesian ordered logistic regressions) 189
Table 6.4 Model 8 (Bayesian ordered logistic regression) 190
Table 6.5 Summary of the findings of the Bayesian ordered logistic
regressions191
Table 6.6 Summary of qualitative findings on extra-cantonal project
managers198
Table 6.7 Summary of qualitative findings on conceptual problems 203
Table 6.8 Summary of qualitative findings on the relationship between
network management and new projects 210
Table 6.9 Summary of results from testing the hypotheses 210
Table 6.10 Summary of findings from testing hypotheses 1 to 3 213
Table 6.11 Summary of findings from testing hypothesis 4 216
Table 6.12 Summary of findings from testing hypothesis 5 218
CHAPTER 1
Management and Performance

in Collaborative Networks: A Review of the
Literature and the Contribution of This Book
This book investigates the management of networks delivering public ser-

vices, one of the major themes in public management and policy research
in the last decades. Network management is an ever more relevant aspect
of both public policy research and practice given the steady increase of
collaborative forms of governance that can be observed all over the world
(Ansell & Torfing, 2021; Mandell, 2001). Collaborative governance is a
strategy used in policy making and implementation to coordinate and
integrate the objectives and interests of different actors (Ansell, 2012).
From a meta-perspective, it “seeks to restore trust in government and
expand democratic consent by deepening participation and deliberation in
public affairs” and changes the relationship between the state and non-
governmental stakeholders (Ansell, 2012, p. 498). An important manifes-
tation of this novel mode of interaction are networks (van Gestel &
Grotenbreg, 2021). Within such networks, network management is used
to steer partnerships between public and non-state organizations in order
to make public policies more effective. As such, network management
provides an alternative to the previously dominant forms of ‘hierarchical
management’ (Agranoff & McGuire, 2001a, p. 11).
In the attempt to understand the drivers of network performance—a
concept that scholars have defined and operationalized in different ways,
as I will discuss in this book—three main groups of determinants have
been studied so far (Turrini et al., 2010): network management activities,
© The Author(s), under exclusive license to Springer Nature 1

Switzerland AG 2022
S. Hadorn, Network Management and Governance in Policy
Implementation, International Series on Public Policy,
https://doi.org/10.1007/978-3-031-08808-7_1
2 S. HADORN
structural characteristics of networks, and contextual factors. The primary

focus of this study is on network management, while I also consider key
structural factors of networks and contextual factors to ensure a compre-
hensive perspective on network performance. The role of a central actor,
the so-called network manager, became increasingly important in order to
systematize and professionalize the coordination tasks within networks.
“The ‘network manager’ is primarily an intermediator and facilitator
[who] tries to create better conditions for problem-solving” (Kickert,
1997, p. 742), whereby scholars argued that these actors’ “managerial
skills have a direct impact on network performance” (Palomo-Navarro &
Navío-Marco, 2018, p. 873).
This book investigates two core aspects related to network manage-
ment: First, I will examine how different types of network managers can
be distinguished based on work-context-related factors and how these fac-
tors interact and result in either active or non-active network manage-
ment. Second, I will investigate the effects that network management
(active vs. non-active) has on network performance measured as the out-
put delivery performance of the networks. Network managers are the indi-
viduals responsible for coordinating the cooperation within networks since
“collaborations are, of course, enacted by individuals […]. [T]hese indi-
viduals are generally linked to the various organizations that form the col-
laboration. Commonly, they are acting in a representative function”
(Huxham & Vangen, 2005, p. 84). In other words, network managers
themselves are embedded in different forms of organizations that partici-
pate in networks, whereby the nature of the respective organization, in
turn, influences the activities of the network managers. I designate the
organization in which the network manager is employed ‘lead organiza-
tion’ hereafter.
In sum, this book addresses the following two major research questions
with the aim of filling current gaps in the research on network management.
1. How do different constellations of factors characterizing a network

manager’s work context affect the level of network management
activities (active vs. non-active network management)?
2. How do network-level characteristics including network manage-
ment and project-level characteristics affect the performance (out-
puts) of projects within the network?
1 MANAGEMENT AND PERFORMANCE IN COLLABORATIVE NETWORKS… 3
1 1 2
Fig. 1.1 Research questions. (Notes: The numbers designate the corresponding
research questions)
The main relationships analyzed are captured in Fig. 1.1 in simplified

form, while the research questions and the underlying gaps in the litera-
ture will be elaborated in more detail in Sect. 1.4.
The aim of this first chapter is to provide an overview about the current
body of literature in the field of network governance and in particular
about network performance and network management. In the first section
of this chapter, I will discuss the reasons for the rising importance of net-
works in public service delivery. The second section positions this book
within the literature by clarifying my understanding of the term network.
The third section outlines the research about determinants of performance
of collaborative networks. Finally, I identify current research gaps in the
fourth section, which represent the starting point of this book. Before
doing so, for the sake of clarity, Box 1.1 provides an overview of the data
and this books’ methodological approach (more details on the research
design can be found in Chap. 3).
Box 1.1 Data and Methodological Approach

To answer the questions raised in this book, I develop a theoretical
model and apply it to Swiss smoking prevention networks. The
methodological approach of this study is threefold: First, I identify
and empirically illustrate the critical characteristics distinguishing
network managers in terms of their work context. I thereby focus on
the level of the individual to identify the most important determi-
nants of a network manager’s actions which are related to her or his
work environment. The identification of the characteristics is based
on an abductive approach (see e.g., Reichertz. 2013), that is,
through a “combination of induction and deduction” (Eriksson &
Lindström, 1997, S. 197). Second, this book analyzes the relation-
ship between these identified factors characterizing network manag-
ers’ work context and the network management strategies applied.
(continued)
4 S. HADORN
Box 1.1 (continued)

To account for configurative effects that are caused by the interac-
tion between these different features, I conduct a qualitative com-
parative analysis (QCA) followed by a post-QCA case analysis. Third,
this book encompasses a theoretical model on how both network-
level factors including network management and project-level fac-
tors affect the performance of projects (measured as policy output)
that are implemented within the respective network. This second
question is examined quantitatively applying Bayesian ordered logis-
tic regressions, followed by qualitative case studies to cross-check the
main findings of the regressions (Seawright, 2016).
In the empirical analyses, I use data from 13 cantonal smoking
prevention networks and the 187 projects implemented in these net-
works. These networks were mandated and funded by a national
agency, the Tobacco Prevention Fund, and are composed of both
non-state and state actors that are either active at the national or the
cantonal level. Each of the networks had to select a network manager
responsible for network-level management with the goal of increas-
ing coordination among the various different organizations. These
cantonal networks provide an ideal case for the analysis of networks
and network management, since a comparison within the cantonal
structures in Switzerland enables what scholars have called a ‘labora-
tory’-like analysis (see, e.g., Freitag, 2005; Vatter, 2002). While
macro-structural aspects such as the political system and the funding
possibilities related to the policy field are constant among the cases,
the 13 networks vary substantially in terms of their structural design,
network management activities, and project configurations. The sta-
bility of macro-level factors thus enables the assessment of the effects
of network- and project-specific factors on network performance
(i.e., the policy outputs delivered).
This book uses data collected within the context of a wider evalu-
ation project, in which the evaluation team conducted 145 semi-
structured interviews with network members including the network
managers of each network as well as other implementation partners
cooperating with the networks. Furthermore, yearly self-evaluation
reports about the progress of the networks as a whole and each indi-
vidual project were analyzed. The reports include both quantitative
(continued)
Box 1.1 (continued)

data about goal achievement and qualitative information about the
implementation processes. Additionally, an analysis of the policy
concepts of each cantonal network provided insights about the poli-
cies’ coherence at the program level and the project level. Finally, to
be able to include contextual factors affecting the implementation
processes, an analysis of parliamentary debates within each canton
and a media analysis were conducted.
1.1 The Age of Collaborative Service Delivery

Throughout the last decades, new and more collaborative forms of gover-
nance have increasingly replaced traditional bureaucratic approaches.
Network governance, which has been widely discussed as one of the main
alternative forms of public service delivery, is characterized by a strong
inclusion of non-state actors in the provision of public policies (Kapucu
et al., 2017). We have observed the shift toward collaborative public man-
agement for a variety of reasons. First, increasingly complex societal prob-
lems cannot be solved by one single organization anymore, but need to be
addressed in new, collaborative ways. Through inter-organizational coop-
eration, more resources and expertise can be gathered to solve complex or
even wicked problems (Isett et al., 2011). Second, because of the increas-
ing importance of outsourcing, collaboration between public and non-
state organizations has grown (O’Leary & Vij, 2012, p. 509). Responding
to the constraints of such instruments of New Public Management, net-
works have evolved to satisfy the demand for more efficient forms of ser-
vice provision (O’Leary & Vij, 2012; Raab et al., 2015). Accordingly,
Graddy and Chen (2009, p. 53) state that “partnering with others is
believed to contribute to increased efficiency […]. And, to the extent this
is true, this movement intersects with and supports the increased focus on
performance.” As a consequence of these shifting priorities and challenges,
the importance and the number of networks in public service delivery
have risen significantly (Raab & Kenis, 2009). The most central character-
istic of this network approach is the interdependency between the various
actors involved in the policy making and delivery process (Kickert et al.,
1997). To achieve the policy goals, public agencies and non-state actors
have to repeatedly exchange information and resources. Public managers
6 S. HADORN
are thus increasingly confronted with the need for inter-organizational

management and coordination to make collaboration effective. Thereby,
networks have either formed voluntarily or have been mandated by state
authorities (McNamara, 2015). Given the large diversity of network con-
cepts discussed in the literature (Börzel, 1997), the following section pro-
vides a more detailed description and definition of the understanding of
networks on which this study is based.
1.2 Definition of the Concept ‘Networks’

For the sake of clarity, I will subsequently outline what I mean when refer-
ring to networks and explain the different components relevant to this
study. Figure 1.2 presents the structure of the networks examined in a
simplified manner. The rectangles at the top represent the organizations
that are active in the network. In the boxes below, the individuals employed
therein are depicted (written in italics). The dashed boxes represent the
implementation settings that are the immediate implementation context
of the projects. The curved arrows depict the network management tasks,
Fig. 1.2 Schematic illustration of the policy delivery networks

while the dashed vertical arrows illustrate the output delivery (per proj-
ect). Finally, the different stick figures represent the different target groups
of the projects.
This study is concerned with networks of at least three interdependent
but still autonomous actors that jointly deliver public services to achieve a
common objective (Raab et al., 2015). While certain aspects of the policy
design phase will be included in the analysis because of the strong interde-
pendency of different stages of the policy cycle (Jann & Wegrich, 2007), I
here focus on the implementation of policies. This study thus distinguishes
itself from the literature linking network theory primarily to questions of
collaborative policy making (see, e.g., Edelenbos & Klijn, 2005). Thereby,
the services delivered by the networks are policy mixes composed of indi-
vidual projects (Mavrot et al., 2018) that are implemented by the various
network members. Each of the project implemented has individual attri-
butes (e.g., the background of the project manager, the type of policy
instrument, the quality of the underlying policy concept) which can have
an effect on the services (i.e., the outputs) delivered (depicted as dashed
vertical arrows in Fig. 1.2). In this sense, network performance is the accu-
mulated performance of these projects, while the pooling of resources and
the coordination of actions are expected to increase the performance of
the network as a whole (see, e.g., Isett et al., 2011).
Importantly, network members can, on the one hand, be understood as
the participating organizations, which include both state and non-state
actors (depicted at the top of Fig. 1.2 in the rectangles) (van Waarden,
1992). I focus on networks operating under a lead organization (Provan
& Kenis, 2008): This central organization within a network is not only
responsible for network management but itself also responsible for deliver-
ing certain services (for more details, see Sect. 1.3). On the other hand,
the term ‘network member’ can also be used to designate the individuals
working within these organizations (depicted in the second upper rectan-
gle in Fig. 1.2) (van Waarden, 1992). Thereby, the number of individuals
may differ from the number of organizations, since several individuals
affiliated with the same organization can operate within a network.
Regarding these individuals, I differentiate between the network manager
(one per network) and the various project managers. One of the key ele-
ments of the following analysis is that in lead organization networks, the
network manager often has a dual role: The respective person is usually
not only responsible for network management (depicted as the curved
horizontal arrows in Fig. 1.2), but also holds—to varying
8 S. HADORN
degrees—project management responsibilities (referred to as ‘caseload’ in

the following).
Finally, as shown in Fig. 1.2, the implementation context in which the
projects are delivered is of central importance in this study. In the field of
public health, there is a general trend that projects are implemented in
different settings (i.e., different implementation contexts depicted as
dashed boxes) to best reach various target groups (illustrated with stick
figures) (Kickbusch & Gleicher, 2012). Thereby, actors that do not
directly belong to the networks, so-called intermediaries or multipliers,
may play a crucial role in shaping policy delivery. In many projects, these
multipliers (who may be street-level bureaucrats if employed by the state)
represent the link between the network and the target groups. I define
multipliers “as bottom-up policy implementers that can also support the
establishment of new actor networks and articulation of interests to bring
about change” (Kivimaa, 2014, p. 1371). For instance, projects imple-
mented in the setting ‘school’ often require the cooperation of teachers,
school nurses, or heads of schools to reach the students which represent
the target groups. Different professional groups may act as multipliers
depending on the nature of the project and the chosen implementation
setting. Before turning to the analytical model in Chap. 2 examining the
relationships between these elements within networks and network per-
formance, the current state of the literature about determinants of net-
work performance and existing research gaps will be explored in the
following sub-chapters.
1.3 Determinants of Network Performance

Given the rising importance of effectiveness and efficiency in the delivery
of public services, the assessment of network performance is central.
However, Kenis and Provan (2009, p. 440) claimed that the ‘euphoria’
about high performance of networks has in a large part of the literature
been based on the assumption that networks “are something positive per
se.” Various authors argued that an empirical examination of the actual
effectiveness of whole networks and potential determinants of success has
been widely neglected (Kenis & Provan, 2009; Turrini et al., 2010; van
Gestel & Grotenbreg, 2021). Nonetheless, a number of studies have
strived to explain network performance empirically. In these studies, per-
formance was often operationalized as the perceived network-level success
(often assessed by network members) or measured as the effects on the
target groups (Jennings Jr & Ewalt, 1998; Meier & O’Toole, 2001, 2003,
2010; O’Toole & Meier, 2004; Provan & Milward, 1995; Raab et al.,
2015; Verweij et al., 2013; Ysa et al., 2014). Cristofoli and Macciò (2018,
p. 900) provide an overview of measures of network performance used in
the literature and argue that “there is a prevalence of subjective/percep-
tual (based on single actors’ perception) measures to appreciate the net-
work output/outcome.” In addition, the rather fragmented stream of
literature has shown that performance of networks is not inherently high.
Rather, as Brown et al. (1998, p. 523) argue, “partnerships are neither
inevitable successes […] nor predictable failures […]” but are “deter-
mined by a series of more detailed –and partially controllable– institu-
tional features.” In their literature review on determinants of network
performance, Turrini et al. (2010) have summarized that three groups of
factors were found to have an impact on network effectiveness: (1) contex-
tual factors, (2) structural characteristics of the network, and (3) network
functioning characteristics (i.e., network management).
First, the two main contextual factors that were found to influence the
performance of networks are the availability of the necessary resources and
the broader communities’ support for a given network. As regards
resources, the existence of, for instance, national funding was found to
positively affect network outcomes in terms of access to public services
and the quality of service delivery (Conrad et al., 2003). Other studies
found that a lack of resources is detrimental for successful cooperation,
however, argued that the existence of funding is not sufficient for network
success (see, e.g., Bazzoli et al., 2003; Provan & Milward, 1995). The
second contextual factor, the wider communities’ support for a network,
is linked to the question whether networks operate in an environment in
which the value of cooperation is appreciated or not (e.g., based on previ-
ous experiences) (Turrini et al., 2010).
Second, numerous structural characteristics of networks have been
studied in regard to their effects on network performance. Both the size
of a network and the selection of network members were found to be rel-
evant for performance, whereby increasing the number of participants was
found to decrease perceived success (Hasnain-Wynia et al., 2003). In
addition, a careful selection of network participants, and where necessary
an exclusion of undesirable actors, has been suggested to ensure better
outcomes (Agranoff & McGuire, 2001b). A further study found that
increasing network age is positively related to network performance, since
it takes time to legitimize collaboration within a network before collective
10 S. HADORN
activities can be started (Raab et al., 2015). Another structural aspect rel-
evant for network effectiveness is the formalization of the cooperation.
The degree of formalization is shaped by the (non-)existence of clear rules
and processes, for example in terms of evaluation processes (Turrini et al.,
2010). This is strongly related to the question of how and if network par-
ticipants can be held accountable by other stakeholders, one of the most
frequently addressed challenges of network governance (Agranoff &
McGuire, 2001b). Since some of the traditional mechanisms of account-
ability are altered or replaced through new mechanisms as a result of the
‘shared power’ context (Klijn & Koppenjan, 2014, p. 244), assuring
accountable behavior becomes more complex in networks. Drivers for this
issue are that networks function without strict hierarchies and that infor-
mation about each other’s actions are restricted (Acar & Robertson,
2004). In addition, instability and fragmentation within networks can lead
to “diffused responsibility for actions and outcomes” (Huang & Provan,
2007, p. 170).
The majority of the previously discussed structural determinants of net-
work success are related to—and can be influenced by—a central coordi-
nating actor, that is, a network manager. A network manager can activate
or exclude participants and thereby influences both the size and the com-
position of a network. A network manager can set rules for cooperation
and monitoring, for instance to increase accountability. It thus seems little
surprising that Provan and Milward (1995) found that the existence or
non-existence of a central steering actor, or in other words of a network
manager, affects network performance. The authors found that the exis-
tence of a central coordinator helps to improve coordination among and
integration of network members. In this regard, Provan and Kenis (2008)
have proposed a typology of three different governance modes: Self-
governed networks, lead organization networks, and networks with a net-
work administrative organization (NAO). Whereas in the self-governed
mode networks members share the management task in a rather informal
manner, management tasks are allocated to a certain actor in a more for-
malized way in the latter two modes. NAOs are organizations that have
the sole task of managing a network and are not directly involved in the
delivery of a specific public service (Provan & Kenis, 2008). In contrast,
lead organizations are characterized by their dual role in which they have
not only duties on the network level (i.e., network management) but are
themselves service providers. This book focuses on this last governance

mode, heeding the call by Chen and Graddy (2010, p. 419) to conduct
further research to understand the drivers of effectiveness in lead organiza-
tion networks. Note that for the sake of simplicity, I will use the term
network manager to designate the person responsible for network man-
agement within the lead organization hereafter.
Since many structural factors may be affected by the existence of a cen-
tral coordinator, they are automatically also strongly linked to what Turrini
et al. (2010) identified as the last group of determinants of network per-
formance: network management. There are numerous studies that have
looked at management issues in networks (Agranoff & McGuire, 2001b;
Herranz, 2008; Klijn et al., 1995, 2010). Thereby, effective management
is seen as a major challenge because of the networks’ inter-organizational
nature, where structures and relationships between actors are often fluid.
Agranoff and McGuire (2001b), for instance, claim that individual net-
work members at times refuse to follow management interventions from
other actors (e.g., network managers). Despite the importance of network
management and the acknowledgment of the related challenges, Willem
and Lucidarme (2014) pointed out that many of the studies on the topic
are theoretical contributions. Hence, management challenges have often
been described, but rarely been studied empirically. For instance, various
scholars have proposed typologies of activities and strategies that network
managers can use to make cooperation more fruitful (e.g., Agranoff &
McGuire, 2001b; Gage & Mandell, 1990; Klijn et al., 2010).
As regards empirical evidence for a positive relationship between net-
work management and performance, Meier and O’Toole (2001) is one of
the most influential studies. They found that the more often network
managers engage in interactions within the network, the higher network
outcomes are (measured as behavioral changes in the target groups).
Similarly, findings of an online survey among participants of public health
networks of Peters et al. (2017) showed that the perceived level of net-
work management activities was positively correlated with perceived net-
work outcomes. A more recent contribution by Cepiku et al. (2021,
p. 1496) also confirmed this positive relationship, arguing that “good net-
work management in the form of having a dedicated network manager” is
of importance to high network effectiveness at the community level.
Notably, scholars argued that not only the number of interactions affects
12 S. HADORN
network outcomes, but also the type of strategy applied: Actively ‘con-
necting’ actors (as opposed to “solely setting the organizational condi-
tions in place”) was found to be the most effective driver of performance
(Klijn et al., 2010, p. 1077). These and similar studies support the propo-
sition that active network management fosters performance, whereby the
studies either capture performance as the success perceived by network
members (see, e.g., Peters et al., 2017) or in terms of effects on target
groups or the community as a whole (Cepiku et al., 2021; Hicklin et al.,
2007; Meier & O’Toole, 2001, 2003, 2007).
Finally, in the last years, several notable attempts were made to bring
the different influencing factors together and explain configurational
effects on network performance. First, Raab et al. (2015) investigated
how network effectiveness is affected by interactions between governance
modes in a network, the structure of a network (including the degree of
integration), and contextual factors (resources and stability). The study
found that centralized integration, stability, and a higher network age
improve network outcomes. Second, the study by Cristofoli and Markovic
(2016) also included contextual, structural, and managerial factors in
their configurational analysis of network performance. In networks that
are equipped with adequate funding for their activities, the coexistence of
a central coordinating agency and active network management was found
to be crucial for high network performance. The study thus supports
findings of previous research suggesting that network management mat-
ters. Thereby, it uncovers that network management alone is not suffi-
cient for high performance but needs to coincide with other factors (such
as sufficient resources and central integration) to take effect. Finally, the
configurational analysis by Cristofoli et al. (2019) showed that depending
on the development stage of a network, the level of trust, and connectiv-
ity within the network, network managers should adopt specific combina-
tions of network management strategies to make networks more effective
in terms of the perceived outcomes. Overall, these recent contributions
have taken research on network performance a step further by providing
new insights about the combined effects of key determinants of network
performance.
1.4 Deficits in the Network Literature

This book focuses on two particular aspects that were insufficiently con-
sidered in earlier research on network management and network perfor-
mance, which I will discuss in more detail below. First, previous research
has largely disregarded the relevance of different factors distinguishing
network managers in terms of their work context. Consequently, there is a
lack of knowledge about the influence of these factors on the managers’
capacity to actively manage a network. Second, despite the many theoreti-
cal contributions in this area, there is still a dearth of empirical studies
examining the influence of network management on network performance
measured in terms of their policy delivery performance.
Both structural aspects of networks and management strategies that
foster successful collaboration have so far been examined to explain net-
work performance. The starting point of the first research question exam-
ined in this book is that there is only very limited research on the question
who manages a network and what the effects of this choice are (notable
exceptions are Herranz, 2008; Juenke, 2005). In other words, evidence is
missing on how network managers can be distinguished from each other
in terms of their work environment and how the interaction of these fac-
tors affects the network managers’ ability to manage the collaboration
successfully. This is particularly surprising in view of the importance of
network managers and their organizations: “the lead agency plays a critical
role in effectively developing and sustaining […] networks. For public
funding agencies, a well-articulated process for qualification should be in
place when selecting lead agencies” (Chen & Graddy, 2010, p. 419).
Similarly, O’Leary and Vij (2012, p. 511) argue that one of the major
challenges in collaboration is the question in what ways and by whom (net-
work manager) a network is governed, as the design of the governance
structure is an extremely complex process. However, to date, there is only
little literature that goes beyond the mere assertion that this choice is deci-
sive. McGuire (2011, pp. 440–441), for example, has explicitly asked the
question “Who is the network manager?” but has focused his answer on
the specific tasks that have to be completed by this actor within a network.
He finds that “the network manager is thus the person or persons who
assume a network-level perspective.”
Nevertheless, what exactly characterizes these individuals and distin-
guishes them from one another has remained unclear in network research
and related assumptions were made without testing them empirically. For
14 S. HADORN
instance, Provan and Kenis (2008, p. 244) argued in a conceptual study

that network managers in a lead organization have one central advantage:
“the lead organization will, typically, already have legitimacy as an organi-
zation, and it can leverage that legitimacy on behalf of the network as a
whole.” However, the questions of whether network managers in lead
organizations always benefit from existing legitimacy and what happens if
there is a lack of legitimacy were not analyzed further. Acknowledging this
deficit in the literature, Müller-Seitz (2012, pp. 438–439) argues in his
literature review on inter-organizational networks that future studies
should analyze which “attributes [of network managers] are relevant for
the successful leading of a network in the first place.” Further scholars
have recently pointed out this research gap, suggesting certain specific
aspects of network managers that should be looked at. On the one hand,
“in a world of shared governance it could be interesting to see if form
(public, nonprofit, or private firm) makes a difference in terms of perfor-
mance” (Turrini et al., 2010, p. 548), which also applies to the activities
of network managers. First insights on differences between management
activities of public and non-state network managers were provided by
Herranz (2008) and will be analyzed further in the present study. On the
other hand, “the network managers’ background, [and the] relation to
the other actors […] might be an interesting avenue for research to pursue
in the near future” (Peters et al., 2017, p. 218). Overall, the most preva-
lent questions are, first, based on what characteristics network managers—
particularly in terms of their work context—can be differentiated. Second,
it is of key interest how these different factors interact with one another
and affect the way in which the actors manage a network. This means that
a configurational approach is most suitable to understand how different
factors affect each other and ultimately shape the network managers’
actions.
The second main gap pointed out by several scholars is the empirical
assessment of the effects of network management on network outcomes.
Specifically, Ysa et al. (2014, p. 650) argue that the positive relationship
between management and outcomes has only been theorized in much of
the literature, while “there is very little empirical evidence that confirms
the theory in practice” (see also Willem & Lucidarme, 2014). Hence, the
authors have called for more empirical studies and for greater diversity in
methodological approaches since a large part of empirical network research
is based on only a few cases (Ysa et al., 2014). Another key limitation of
the current body of literature is that, as shown in Sect. 1.3, most of the few
empirical studies examining the connection between network manage-

ment and network performance captured performance as success per-
ceived by single groups of actors or in terms of the effects on the target
groups. Thereby, measuring perceived success suffers from common over-
estimations of the own performance in a self-assessment (Vabo et al.,
2011). Equally problematic is a direct attribution of changes in the target
groups’ behavior (i.e., the outcomes of a policy) to actions of implement-
ing actors (such as network management activities) since this contradicts
the logic of well-established approaches in the field of policy evaluation
(Pawson & Tilley, 1997; Sager et al., 2021; Sager & Rüefli, 2005). If the
actions of implementing actors are linked directly to behavioral changes in
the target group, the analysis does not empirically test whether the policy’s
intervention hypothesis, capturing how the policy outputs are hypothe-
sized to change the target groups’ behavior (Sager et al., 2018), is accu-
rate or not. As such, these studies do not adequately capture the effects of
network management on the service delivery performance of the networks
(a notable exception is Cristofoli & Markovic, 2016). In Sect. 2.4.1, I will
discuss the limitations of the aforementioned performance indicators as
well as the benefits of an alternative performance measurement, namely
policy outputs, in more detail.
Because of choosing a rather novel performance criterion, namely pol-
icy outputs, this study further contributes to the literature on network
performance, paying particular attention to the multilevel structure of net-
works. As depicted in Fig. 1.2, networks are composed of different, inter-
dependent actors that “remain autonomous entities concentrating on
manageable pieces of the full service that best fit their specific area of
expertise” (Provan & Milward, 1991, pp. 394–395). These ‘manageable
pieces’ can take the form of projects, while it is at this project level where
policy outputs are delivered and ultimately target groups are reached. In
this sense, networks can only perform well if as many projects as possible
are delivering the planned outputs. Thus, I argue that the policy delivery
performance of networks can be assessed most adequately when measur-
ing project outputs, despite the view of certain scholars that networks
should be assessed on the network level rather than on the level of indi-
vidual organizations’ performance (Kenis & Provan, 2009; Turrini
et al., 2010).
If we strive to understand the effects of network-level factors such as
management activities on project-level outputs—and with this on net-
works’ problem-solving capacity—we also need to include project
16 S. HADORN
characteristics that might affect policy implementation. A further impor-

tant contribution of this book is thus the identification and empirical anal-
ysis of key project-level factors that might influence the service delivery
within networks. Connecting this second overall gap with the previous
research gap suggests that there is the need for a closer examination of,
first, how the work context of network managers affects their activities,
and, second, if project outputs differ depending on whether a network is
managed actively or not. An analysis of project outputs should consider
the multilevel character of networks and thus include potentially
performance- relevant project-level characteristics besides the usually
included network-level factors.
In sum, I identified the following two main research gaps: Previous
studies have stressed the importance of the network managers did, how-
ever, (1) not comprehensively study the key factors distinguishing network
managers in terms of their work context and how different constellations
of these factors affect network management. Thereby, numerous studies
have theorized the impact of network management on network perfor-
mance, while only little evidence exists on this relationship. The few
empirical studies on this topic have mostly measured network performance
as policy outcomes (i.e., changes in target group behavior or community-
level effects) or as the network success perceived by the network members,
while both approaches have their weaknesses. Consequently, (2) we know
very little about how network-level characteristics including network man-
agement as well as project-level characteristics affect the measured output
delivery performance of projects implemented within networks.
This book is structured as follows: In the following Chap. 2, I develop
and explain the theoretical model covering the two research questions.
Chapter 3 outlines the methods and data used in the empirical analyses,
followed by Chap. 4 that provides a description of the case studied, namely
the Swiss smoking prevention policy. In Chaps. 5 and 6, I will empirically
analyze the research questions and discuss the findings for each chapter. In
Chap. 7, I provide a discussion interlinking the different parts of the book.
Finally, in the conclusion in Chap. 8, I summarize the overall results, open
up the discussion related to the findings of this book, and present direc-
tions for further research. Before turning to the theory chapter, Box 1.2
provides a brief overview of the book’s core findings for ‘quick readers’.
Box 1.2 The Book’s Results in a Nutshell

The results of the study show that network managers can be differ-
entiated along four main factors in terms of their work context: First,
the legal status or in other words, whether the network manager is
affiliated with a public or non-state organization. Second, the degree
of integration between the organization with which the network
manager is affiliated and other network partners measured in terms
of existing or non-existing cooperation experiences. Third, the net-
work managers’ caseload as regards their direct project implementa-
tion tasks. As such, I capture the network managers’ relative
weighting of network management tasks versus project implementa-
tion responsibilities. The fourth distinguishing factor is the quality of
the policy design process, relating to whether the network manager
operates in a context characterized by a conflict between different
organizations regarding the structure of the network and its objec-
tives or not. With regard to the coexistence of these factors in Swiss
smoking prevention networks, there existed multiple different pat-
terns among the cases. On the one hand, public network managers
were typically not integrated because they had not played a role in
smoking prevention before the establishment of the mandated net-
works, while network managers who were affiliated with non-state
lead organizations could usually base their actions on high integra-
tion due to the NGOs’ historical role in the policy field. On the
other hand, network managers embedded in not-integrated organi-
zations experienced a conflict in the policy design phase in substan-
tially more cases than integrated network managers did. This suggests
that existing relationships between actors help to prevent conflicts
regarding, for instance, the definition of goals in the policy
design phase.
Importantly, if considered individually, any of these four factors
(legal status, integration, caseload, and policy design phase) may
either inhibit or benefit the network management activities, depend-
ing on their nature. Based on previous studies analyzing individual
factors, I expect non-state network managers, integrated network
managers, network managers with a low caseload, and network man-
(continued)
18 S. HADORN
Box 1.2 (continued)

agers experiencing a conflict-free policy design phase to have better
conditions to actively manage a network compared to public net-
work managers, non-integrated network managers, network manag-
ers with a high caseload, and network managers experiencing a
conflictual policy design phase (see e.g., Herranz, 2008; Juenke,
2005; Kenis & Provan, 2009; Raab et al., 2015; Raab & Kenis,
2009; Saz-Carranza et al., 2016; Schaap & van Twist, 1997). The
configurative analysis presented in this book demonstrates, however,
that no single factor is sufficient for active nor for non-active net-
work management. Instead, various configurations of multiple fac-
tors explain active network management. Several combinations such
as, for instance, non-governmental network managers with a low
caseload and a conflict-free policy design phase led to active network
management in the networks studied. This finding is consistent with
the assumptions regarding the direction of the effect of the individ-
ual factors. More surprising is the result that the combination of
three inhibiting properties (public network manager with high case-
load and conflictual policy design phase) also led to active network
management. The evidence shows that in situations with very bad
preconditions, network managers are put under particular pressure
to act. As a reaction, one of the network managers reduced her case-
load (i.e., shifted project implementation responsibilities to another
network partner) to increase the network management capacities. In
this vein, the qualitative analysis (which complements the QCA)
demonstrates that a low caseload was not a sufficient but a necessary
condition for active network management in the 13 networks stud-
ied. This finding has important implications for practitioners as it
illustrates that investing a sufficient amount of resources is key to
achieving active network management. Policy makers should be
aware of the challenges that the dual role of network managers—the
balancing of network management and direct policy implementation
responsibilities (in this book called ‘caseload’) —entails.
A further central result is that trust does not always improve
cooperation in networks (see, e.g., Isett & Provan, 2002; Vangen &
Huxham, 2003), but that network managers’ desire to build trust
can also lead to negative dynamics and thus non-active network
(continued)
Box 1.2 (continued)

management. The fear of losing previously established trust through
overly extensive interferences in the actions of the network partners
may prevent certain network managers from actively managing their
networks. The evidence shows that the danger of such a dynamic
exists especially in networks with public network managers: Due to
their affiliation with the public administration, they strive to avoid
being perceived as too top-down by the non-governmental partners.
To achieve this and to keep the trust of the partners, certain network
managers reduced their network management activities, which may
be perceived as an illegitimate intervention affecting the autonomy
of the network partners. Too much harmony and the desire to create
trustful relationships within networks may thus lead to a self-imposed
deadlock of network managers in certain contexts. This result differs
from Brummel et al. (2012) and Muir and Mullins (2015) who
found that in mandated networks, network management tasks are
delegated to public actors who use top-down management strategies
to bring power back to the state. The empirical evidence provided in
this book, however, suggests that public actors do not in all circum-
stances strive to take over power and often do not use their institu-
tional authority when managing the networks.
In contrast, the management activities of non-state network man-
agers were limited primarily in cases where a fundamental conflict
between the vision of the mandated agency and those of the network
managers existed. Indications of such differences were already pres-
ent in the conflict-laden policy design phase, but the disagreements
were not resolved. This shows that conflicts between actors within
the network and the mandating agency may affect performance.
Divergent visions must either be aligned before the implementation
phase or need to be carefully considered in the choice of a motivated
network manager if active network management is to be achieved.
With regard to the decisive role of the policy design process, also the
stability of requirements on the part of the mandating agency plays
a particularly important role. The evidence shows that unclear and
changing rules with regard to financing conditions can reduce the
motivation of the network members, which in turn hampers network
management activities.
(continued)
20 S. HADORN
Box 1.2 (continued)

Finally, this book provides crucial evidence on the question of
whether active network management actually has an impact on net-
work performance. The results show that active management by net-
work managers does not—as previously assumed—have a positive
impact on all projects, but that active network management primar-
ily affects a particular group of projects. Specifically, the analysis of
the 187 projects shows that the output delivery performance of new
projects is higher when the corresponding network manager actively
manages the cooperation compared to new projects in contexts of
non-active network management. According to the qualitative case
studies, the reason for this difference is that the performance of new
projects strongly depends on how quickly and well the network part-
ners can establish new relationships with implementation partners.
Network managers can make a significant contribution, especially in
this important initial phase, if they provide their network partners
with access to target groups and connect different stakeholders. In
contrast, the potential contribution network managers can make in
existing projects is generally lower, since the project structures
including the necessary partnerships usually already exist, thus elimi-
nating the need for high initial investments in these aspects. In exist-
ing projects, active network managers can provide beneficial support
primarily in exceptional situations (e.g., in phases of crisis or transi-
tion). Practitioners can use these new insights to design more tar-
geted network management interventions in order to increase
network performance. In addition, the analysis of the output deliv-
ery performance of networks shows that project-level factors also
affect performance. Future analyses of network performance should
therefore no longer exclusively consider network-level factors such
as the age and size of the networks, but also take into account proj-
ect-level factors such as the quality of their concepts or whether the
project is new or existing that influence the delivery of policies
directly at the implementation level.
References
Acar, M., & Robertson, P. J. (2004). Accountability challenges in networks and
partnerships: Evidence from educational partnerships in the United States.
International Review of Administrative Sciences, 70(2), 331–344. https://doi.
org/10.1177/0020852304044260
Agranoff, R., & McGuire, M. (2001a). After the network is formed: Process,
power, and performance. In M. Mandell (Ed.), Getting results through collabo-
ration: Networks and network structures for public policy and management
(pp. 11–29). Quorum Books.
Agranoff, R., & McGuire, M. (2001b). Big questions in public network manage-
ment research. Journal of Public Administration Research and Theory, 11(3),
295–326. https://doi.org/10.1093/oxfordjournals.jpart.a003504
Ansell, C. (2012). Collaborative governance. In D. Levi-Faur (Ed.), Oxford hand-
book of governance (pp. 498–511). Oxford University Press. https://doi.
org/10.1093/oxfordhb/9780199560530.013.003
Ansell, C., & Torfing, J. (2021). Co-creation: The new kid on the block in public
governance. Policy & Politics, 49(2), 211–230. https://doi.org/10.133
2/030557321X16115951196045
Bazzoli, G. J., Casey, E., Alexander, J. A., Conrad, D. A., Shortell, S. M., Sofaer,
S., Hasnain-Wynia, R., & Zukoski, A. P. (2003). Collaborative initiatives:
Where the rubber meets the road in community partnerships. Medical Care
Research and Review, 60(4 Suppl), 63S–94S. https://doi.
org/10.1177/1077558703259082
Börzel, T. A. (1997). What’s so special about policy networks? An explora-
tion of the concept and its usefulness in studying European governance.
European Integration Online Papers, 1(16). http://eiop.or.at/eiop/
texte/1997-016a.htm
Brown, M. M., O’Toole, L. J., & Brudney, J. L. (1998). Implementing informa-
tion technology in government: An empirical assessment of the role of local
partnerships. Journal of Public Administration Research and Theory, 8(4),
499–526. https://doi.org/10.1093/oxfordjournals.jpart.a024394
Brummel, R. F., Nelson, K. C., & Jakes, P. J. (2012). Burning through organiza-
tional boundaries? Examining inter-organizational communication networks in
policy-mandated collaborative bushfire planning groups. Global Environmental
Change, 22(2), 516–528. https://doi.org/10.1016/j.gloenvcha.2011.12.004
Cepiku, D., Giordano, F., Mastrodascio, M., & Wang, W. (2021). What drives
network effectiveness? A configurational approach. Public Management Review,
23(10), 1479–1503. https://doi.org/10.1080/14719037.2020.1764084
Chen, B., & Graddy, E. A. (2010). The effectiveness of nonprofit lead-organization
networks for social service delivery. Nonprofit Management and Leadership,
20(4), 405–422. https://doi.org/10.1002/nml.20002
22 S. HADORN
Conrad, D. A., Cave, S. H., Lucas, M., Harville, J., Shortell, S. M., Bazzoli, G. J.,
Hasnain-Wynia, R., Sofaer, S., Alexander, J. A., Casey, E., & Margolin,
F. S. (2003). Community care networks: Linking vision to outcomes for com-
munity health improvement. Medical Care Research and Review, 60(4),
95–129. https://doi.org/10.1177/1077558703259096
Cristofoli, D., & Macciò, L. (2018). To wind a skein into a ball: Exploring the con-
cept and measures of public network performance. Public Management Review,
20(6), 896–922. https://doi.org/10.1080/14719037.2017.1363904
Cristofoli, D., & Markovic, J. (2016). How to make public networks really work:
A qualitative comparative analysis. Public Administration, 94(1), 89–110.
https://doi.org/10.1111/padm.12192
Cristofoli, D., Trivellato, B., & Verzillo, S. (2019). Network management as a
contingent activity. A configurational analysis of managerial behaviors in differ-
ent network settings. Public Management Review, 92(3), 1–26. https://doi.
org/10.1080/14719037.2019.1577905
Edelenbos, J., & Klijn, E. H. (2005). Managing stakeholder involvement in deci-
sion making: A comparative analysis of six interactive processes in the
Netherlands. Journal of Public Administration Research and Theory, 16(3),
417–446. https://doi.org/10.1093/jopart/mui049
Eriksson, K., & Lindström, U. Å. (1997). Abduction—a way to deeper under-
standing of the world of caring. Scandinavian journal of caring sciences, 11(4),
195–198. https://doi.org/10.1111/j.1471-6712.1997.tb00455.x.
Freitag, M. (2005). Labor Schweiz: Vergleichende Wahlbeteiligungsforschung bei
kantonalen Parlamentswahlen. KZfSS Kölner Zeitschrift für Soziologie und
Sozialpsychologie, 57(4), 667–690. https://doi.org/10.1007/s11577-005-
0220-7
Gage, R. W., & Mandell, M. (1990). Strategies for managing intergovernmental
policies and networks. Praeger.
Graddy, E. A., & Chen, B. (2009). Partner selection and the effectiveness of inter-
organizational collaborations. In R. O’Leary & L. B. Bingham (Eds.), The col-
laborative public manager: New ideas for the twenty-first century (pp. 53–70).
Georgetown University Press.
Hasnain-Wynia, R., Sofaer, S., Bazzoli, G. J., Alexander, J. A., Shortell, S. M.,
Conrad, D. A., Chan, B., Zukoski, A. P., & Sweney, J. (2003). Members’ per-
ceptions of community care network partnerships’ effectiveness. Medical Care
Research and Review, 60(4 Suppl), 40S–62S. https://doi.org/
10.1177/1077558703260272
Herranz, J. J. (2008). The multisectoral trilemma of network management.
Journal of Public Administration Research and Theory, 18(1), 1–31. https://
doi.org/10.1093/jopart/mum004
Hicklin, A., O’Toole, L. J., & Meier, K. J. (2007). Serpents in the sand: Managerial
networking and nonlinear influences on organizational performance. Journal of
Public Administration Research and Theory, 18(2), 253–273. https://doi.

org/10.1093/jopart/mum009
Huang, K., & Provan, K. G. (2007). Structural embeddedness and organizational
social outcomes in a centrally governed mental health services network. Public
Management Review, 9(2), 169–189. https://doi.org/10.1080/
14719030701340218
Huxham, C., & Vangen, S. (2005). Managing to collaborate: The theory and prac-
tice of collaborative advantages. Routledge.
Isett, K. R., Mergel, I. A., LeRoux, K., Mischen, P. A., & Rethemeyer, R. K. (2011).
Networks in public administration scholarship: Understanding where we are
and where we need to go. Journal of Public Administration Research and
Theory, 21(1), 157–173. https://doi.org/10.1093/jopart/muq061
Isett, K. R., & Provan, K. G. (2002). The evolution of interorganizational net-
work relationships over time: Does sector matter? Journal of Public
Administration Research and Theory, 15(1), 149–165. https://doi.
org/10.5465/APBPP.2002.7519436
Jann, W., & Wegrich, K. (2007). Theories of the policy cycle. In F. Fischer,
G. J. Miller, & M. S. Sidney (Eds.), Handbook of public policy analysis: Theory,
politics, and methods (pp. 43–62). Routledge.
Jennings, E. T., Jr., & Ewalt, J. A. G. (1998). Interorganizational coordination,
administrative consolidation, and policy performance. Public Administration
Review, 58(5), 417–428. https://doi.org/10.2307/977551
Juenke, E. G. (2005). Management tenure and network time: How experience
affects bureaucratic dynamics. Journal of Public Administration Research and
Theory, 15(1), 113–131. https://doi.org/10.1093/jopart/mui006
Kapucu, N., Hu, Q., & Khosa, S. (2017). The state of network research in public
administration. Administration & Society, 49(8), 1087–1120. https://doi.
org/10.1177/0095399714555752
Kenis, P., & Provan, K. G. (2009). Towards an exogenous theory of public net-
work performance. Public Administration, 87(3), 440–456. https://doi.
org/10.1111/j.1467-9299.2009.01775.x
Kickbusch, I., & Gleicher, D. E. (2012). Governance for health in the 21st century.
WHO Regional Office for Europe.
Kickert, W. J. M. (1997). Public governance in the Netherlands: An alternative to
Anglo-American ‘managerialism’. Public Administration, 75(4), 731–752.
https://doi.org/10.1111/1467-9299.00084
Kickert, W. J. M., Klijn, E. H., & Koppenjan, J. F. M. (1997). Introduction: A
management perspective on policy networks. In W. J. M. Kickert, E. H. Klijn,
& J. F. M. Koppenjan (Eds.), Managing complex networks: Strategies for the
public sector (pp. 1–13). Sage publications.
Kivimaa, P. (2014). Government-affiliated intermediary organisations as actors in
system-level transitions. Research Policy, 43(8), 1370–1380. https://doi.
org/10.1016/j.respol.2014.02.007
24 S. HADORN
Klijn, E. H., & Koppenjan, J. F. M. (2014). Accountable networks. In M. Bovens,

R. E. Goodin, & T. Schillemans (Eds.), The oxford handbook of public account-
ability (1st ed., pp. 242–257). Oxford University Press. https://doi.
org/10.1093/oxfordhb/9780199641253.013.0029
Klijn, E. H., Koppenjan, J. F. M., & Termeer, K. (1995). Managing networks in
the public sector: A theoretical study of management strategies in policy net-
works. Public Administration, 73(3), 437–454. https://doi.org/10.1111/
j.1467-9299.1995.tb00837.x
Klijn, E. H., Steijn, B., & Edelenbos, J. (2010). The impact of network manage-
ment on outcomes in governance networks. Public Administration, 88(4),
1063–1082. https://doi.org/10.1111/j.1467-9299.2010.01826.x
Mandell, M. (Ed.). (2001). Getting results through collaboration: Networks and
network structures for public policy and management. Quorum Books.
Mavrot, C., Hadorn, S., & Sager, F. (2018). Mapping the mix: Linking instru-
ments, settings and target groups in the study of policy mixes. Research Policy,
48(10), Article 103614. https://doi.org/10.1016/j.respol.2018.06.012
McGuire, M. (2011). Network management. In M. Bevir (Ed.), The SAGE hand-
book of governance (pp. 436–453). Sage publications.
McNamara, M. W. (2015). Unraveling the characteristics of mandated collabora-
tion. In J. C. Morris & K. Miller-Stevens (Eds.), Advancing collaboration the-
ory: Models, typologies, and evidence (pp. 65–85). Routledge.
Meier, K. J., & O’Toole, L. J. (2001). Managerial strategies and behavior in net-
works: A model with evidence from US public education. Journal of Public
org/10.1093/oxfordjournals.jpart.a003503
Meier, K. J., & O’Toole, L. J. (2003). Public management and educational perfor-
mance: The impact of managerial networking. Public Administration Review,
63(6), 689–699. https://doi.org/10.1111/1540-6210.00332
Meier, K. J., & O’Toole, L. J. (2007). Modeling public management: Empirical
analysis of the management–performance nexus. Public Management Review,
9(4), 503–527. https://doi.org/10.1080/14719030701726630
Meier, K. J., & O’Toole, L. J. (2010). Beware of managers not bear-
ing gifts: How management capacity augments the impact of manage-
rial networking. Public Administration, 88(4), 1025–1044. https://doi.
org/10.1111/j.1467-9299.2010.01853.x
Muir, J., & Mullins, D. (2015). The governance of mandated partnerships: The
case of social housing procurement. Housing Studies, 30(6), 967–986. https://
doi.org/10.1080/02673037.2014.995070
Müller-Seitz, G. (2012). Leadership in interorganizational networks: A lit-
erature review and suggestions for future research. International
Journal of Management Reviews, 14(4), 428–443. https://doi.
org/10.1111/j.1468-2370.2011.00324.x
O’Leary, R., & Vij, N. (2012). Collaborative public management: Where have we
been and where are we going? The American Review of Public Administration,
42(5), 507–522. https://doi.org/10.1177/0275074012445780
O’Toole, L. J., & Meier, K. J. (2004). Public management in intergovernmental
networks: Matching structural networks and managerial networking. Journal of
org/10.1093/jopart/muh032
Palomo-Navarro, A., & Navío-Marco, J. (2018). Smart city networks’ gover-
nance: The Spanish smart city network case study. Telecommunications Policy,
42(10), 872–880. https://doi.org/10.1016/j.telpol.2017.10.002
Pawson, R., & Tilley, N. (1997). Realistic evaluation. Sage publications.
Peters, D. T. J. M., Klijn, E. H., Stronks, K., & Harting, J. (2017). Policy coordi-
nation and integration, trust, management and performance in public health-
related policy networks: A survey. International Review of Administrative
Sciences, 83(1), 200–222. https://doi.org/10.1177/0020852315585061
Provan, K. G., & Kenis, P. (2008). Modes of network governance: Structure,
management, and effectiveness. Journal of Public Administration Research and
Theory, 18(2), 229–252. https://doi.org/10.1093/jopart/mum015
Provan, K. G., & Milward, H. B. (1991). Institutional-level norms and organiza-
tional involvement in a service-implementation network. Journal of Public
org/10.1093/oxfordjournals.jpart.a037100
Provan, K. G., & Milward, H. B. (1995). A preliminary theory of interorganiza-
tional network effectiveness: A comparative study of four community mental
health systems. Administrative Science Quarterly, 40(1), 1–33. https://doi.
org/10.2307/2393698
Raab, J., & Kenis, P. (2009). Heading toward a society of networks: Empirical
developments and theoretical challenges. Journal of Management Inquiry,
18(3), 198–210. https://doi.org/10.1177/1056492609337493
Raab, J., Mannak, R. S., & Cambré, B. (2015). Combining structure, governance,
and context: A configurational approach to network effectiveness. Journal of
org/10.1093/jopart/mut039
Reichertz, J. (2013). Die Abduktion in der qualitativen Sozialforschung: Über die
Entdeckung des Neuen. Wiesbaden: Springer-Verlag.
Sager, F., Hadorn, S., Balthasar, A., & Mavrot, C. (2021). Politikevaluation: Eine
Einführung. Springer VS. https://link.springer.com/content/
pdf/10.1007%2F978-3-658-32490-2.pdf; https://doi.org/10.1007/
978-3-658-32490-2.
Sager, F., Ingold, K. M., & Balthasar, A. (2018). Policy-Analyse in der Schweiz:
Besonderheiten, Theorien, Beispiele (2nd ed.).
Sager, F., & Rüefli, C. (2005). Die Evaluation öffentlicher Politiken mit föderalis-
tischen Vollzugsarrangements. Eine konzeptionelle Erweiterung des
26 S. HADORN
Stufenmodells und eine praktische Anwendung. Swiss Political Science Review,

11(2), 101–129. https://doi.org/10.1002/j.1662-6370.2005.tb00357.x
Saz-Carranza, A., Salvador Iborra, S., & Albareda, A. (2016). The power dynam-
ics of mandated network administrative organizations. Public Administration
Review, 76(3), 449–462. https://doi.org/10.1111/puar.12445
Seawright, J. (2016). Multi-method social science: Combining qualitative and
quantitative tools. Cambridge: Cambridge University Press.
Schaap, L., & van Twist, M. (1997). The dynamics of closedness in networks. In
W. J. M. Kickert, E. H. Klijn, & J. F. M. Koppenjan (Eds.), Managing complex
networks: Strategies for the public sector (pp. 63–78). Sage publications. https://
doi.org/10.4135/9781446217658.n4
Turrini, A., Cristofoli, D., Frosini, F., & Nasi, G. (2010). Networking literature
about determinants of network effectiveness. Public Administration, 88(2),
528–550. https://doi.org/10.1111/j.1467-9299.2009.01791.x
Vabo, S. I., Røiseland, A., & Nyseth, T. (2011). Evaluating performance in urban
development networks—The Nordic context. Urban Research & Practice,
4(1), 72–84. https://doi.org/10.1080/17535069.2011.550543
van Gestel, N., & Grotenbreg, S. (2021). Collaborative governance and innova-
tion in public services settings. Policy & Politics, 49(2), 249–265. https://doi.
org/10.1332/030557321X16123785900606
van Waarden, F. (1992). Dimensions and types of policy networks. European
Journal of Political Research, 21(1–2), 29–52. https://doi.org/10.1111/
j.1475-6765.1992.tb00287.x
Vangen, S., & Huxham, C. (2003). Nurturing collaborative relations: Building
trust in interorganizational collaboration. The Journal of Applied Behavioral
Science, 39(1), 5–31. https://doi.org/10.1177/0021886303039001001
Vatter, A. (2002). Kantonale Demokratien im Vergleich: Entstehungsgründe,
Interaktionen und Wirkungen politischer Institutionen in den Schweizer
Kantonen. Wiesbaden: Springer.
Verweij, S., Klijn, E. H., Edelenbos, J., & van Buuren, A. (2013). What makes
governance networks work? A fuzzy set qualitative comparative analysis of 14
Dutch spatial planning projects. Public Administration, 91(4), 1035–1055.
Willem, A., & Lucidarme, S. (2014). Pitfalls and challenges for trust and effective-
ness in collaborative networks. Public Management Review, 16(5), 733–760.
https://doi.org/10.1080/14719037.2012.744426
Ysa, T., Sierra, V., & Esteve, M. (2014). Determinants of network outcomes: The
impact of management strategies. Public Administration, 92(3), 636–655.
CHAPTER 2
Linking Characteristics of Network

Managers’ Work Context to Network
Management and Project-Level Output
The aim of this theoretical chapter is to develop the analytical framework

of this book. Based on the previously discussed gaps in the literature, I
strive to answer two main research questions. First, I will analyze which
factors characterize the network managers’ work context within different
networks. The goal is to identify the central factors affecting the way in
which the network managers manage a network. Based on this, this book
examines how various differently equipped network managers manage net-
works. Thereby, a focus will be on the configurational effects of the identi-
fied factors: Rather than solely looking at how each work-context-related
feature individually affects the network managers’ activities, I will investi-
gate how these features interact in different constellations and thereby
affect network management. Second, I focus on the problem-solving
capacity of networks. Specifically, I will analyze how network characteris-
tics and network management affect the performance of a network (at the
project level) to provide the planned services, that is, policy outputs. In
addition, characteristics of the respective projects will be included in the
analysis to control for performance differences that are caused on this level.
The structure of this chapter is as follows: In the first sub-chapter, I
shortly summarize the analytical framework before starting with the
detailed development of the model in the next sub-chapters. In Sect. 2.2,
I identify the key features distinguishing network managers in terms of
their work context, which I then put in relation with network
© The Author(s), under exclusive license to Springer Nature 27

Switzerland AG 2022
S. Hadorn, Network Management and Governance in Policy
Implementation, International Series on Public Policy,
https://doi.org/10.1007/978-3-031-08808-7_2
28 S. HADORN
management activities in the third sub-chapter. The aim is to make theo-

retical assumptions about the influence of these different factors on net-
work managers’ capacities to manage networks more or less actively. In the
fourth sub-chapter, I develop a framework linking both network charac-
teristics including network management and project-level characteristics
with network performance measured as project outputs. The final section
provides a summary of the model and the research questions guiding
this book.
2.1 A Synthesis of the Analytical Framework

Before elaborating the individual parts of the theoretical model, Fig. 1.1
offers an overview of the entire analytical framework. First, I argue that
four main factors distinguish network managers in terms of their work
context: (1) The legal status of the organization they are affiliated with
(i.e., is it a public or non-state actor); (2) the caseload (i.e., do the network
managers have a high workload due to many direct project management
tasks in addition to their network management task or not); (3) the level
of integration of the network managers’ lead organization (i.e., did the
organization in which the network manager works cooperate with the
other network members before the studied policy program or not); (4)
and the nature of the policy design phase (i.e., was there a conflict between
different organizations during the planning stage or not). Additionally, I
argue that the individual effects of the factors ‘legal status’, ‘caseload’,
‘integration’, and ‘policy design phase’ on network management should
not be considered in isolation. Rather, these factors affect network man-
agement activities depending on the combination with the other factors.
A configurational analysis is thus the most promising approach to fully
understand the underlying mechanisms. Second, based on previous net-
work management and policy analysis literature, I develop a model to ana-
lyze network performance measured as project-level outputs. The
framework includes network-level characteristics including network size,
network age, and network management as well as project-level character-
istics such as project status (i.e., new or existing), the quality of the policy
concept (i.e., conceptual problems or no conceptual problems), and proj-
ect managers’ characteristics (i.e., is it the same actor as the network man-
ager or not) to explain policy outputs. In the following sections, I develop
the model displayed in Fig. 2.1 in more detail.
2 LINKING CHARACTERISTICS OF NETWORK MANAGERS’ WORK CONTEXT… 29
Factors distinguishing Network Network Network-level factors

network managers in management characteristics
terms of their work - Size
context 1 - Age
- Integration
- Caseload
- Legal status
- Policy design phase
Project-level factors
2 2 2
Project characteristics
- Status
- Double role
- Level
Policy output 2 - Concept
(Project-level) - Instrument type
Fig. 2.1 Basic model. (Notes: Arrows show expected causal effects. The circled
numbers designate the research questions)
2.2 Conditions Distinguishing Different Network

Managers in Terms of Their Work Context
As discussed previously, studies about network management have tradi-
tionally focused on the activities that have to be carried out to manage
collaborations. Only a few scholars have explicitly discussed by whom
these tasks are performed. The aim of this section is to outline how net-
work managers and specifically their work context may differ from one
another by drawing on existing literature and extensive observations from
practice. The term ‘work context’ refers to the fact that network manag-
ers—understood as individuals who always act as representatives of the
organization they belong to within networks (Huxham & Vangen,
2005)—may be embedded in very different work environments. Overall,
four central factors characterizing network managers’ work context were
identified that are relevant for a manager’s role within the network:
Integration, legal status, caseload, and the quality of the policy design
phase. Since I will examine how these factors influence network manage-
ment by means of a qualitative comparative analysis (QCA), the four fac-
tors are called ‘conditions’ in the following in accordance with the
corresponding literature (see, e.g., Sager & Rielle, 2013).
30 S. HADORN
2.2.1 Integration
The relationship between the network manager and network members can
vary substantially depending on the emergence process of a network.
Networks can form as a result of a joint and voluntary effort by the net-
work members or if a third party, for example a state actor, mandates
them. In addition, an initially voluntarily developed network can later be
modified through a mandate into a more formalized type of collaboration.
As networks develop over time, changes are likely to take place in the way
a network is organized. For instance, a shift from a self-governance net-
work to a network that is managed by a lead organization might occur
(Provan & Kenis, 2008).
Emergence processes can have strong effects on the way a network
operates and on the attitudes of network participants toward the joint
endeavor (McNamara, 2015). As such, the lead organizations can either
be part of the network from the very beginning of the joint action or they
can be deployed at a later stage, for example after the network has changed
its governance mode as a result of a mandate. In this regard, it has been
argued that lead organizations are usually selected as a result of their high
legitimacy (Provan & Kenis, 2008). This argument is, however, contrasted
by the observation that—particularly in mandated networks—the deploy-
ment of a network manager is often a question of power (McNamara,
2015). In these cases, the specific choice might be the result of a top-
down process in which an actor is selected who has no existing ties within
the respective policy field. To summarize, a lead organization and the
therein appointed network manager can have strong ties with the rest of
the network or can be a new actor within the respective policy field that
has no relationships with the other network members. The condition
‘integration’ encompassing these two opposing starting points of network
managers is subsequently categorized as ‘integrated’ versus ‘not integrated’.
2.2.2 Legal Status

Traditionally, public organizations have played a key role in the delivery of
public services. However, there has been a trend toward a stronger involve-
ment of non-state organizations in public service delivery. As a result, col-
laborative networks are characterized by their cross-sectoral composition
in which public organizations and non-state actors work together toward
a common goal (Mischen, 2015). While the multi-sectoral nature of
2 LINKING CHARACTERISTICS OF NETWORK MANAGERS’ WORK CONTEXT… 31
network members is undisputed, the majority of the network (manage-

ment) literature does not make a clear analytical distinction between net-
work managers that are state actors and those who are non-state actors. In
many cases, a public agency acts in the role of the lead organization
employing the network manager to improve collaboration with other pub-
lic and non-state organizations. Empirical studies, however, show that lead
organizations can also be non-state actors, such as non-profit organiza-
tions (Chen & Graddy, 2010).
The development process of a network and the applied selection criteria
influence whether a public or a non-state actor becomes the lead organiza-
tion deploying the network manager. McNamara (2015), for example,
argued that mandates are often used to place a public actor into a network
to gain back state control. In other cases, a central non-state actor may be
selected for the role of the lead organization (e.g., through the network
members) due to its high legitimacy within the respective policy field
(Provan & Kenis, 2008). This is not to say that only non-state lead orga-
nizations can have a high legitimacy within a given policy field or that
public lead organizations are always selected to gain back state control.
However, it illustrates that a network manager can be installed either by a
public or a non-state lead organization.
2.2.3 Caseload
The definition of lead organizations proposed by Provan and Kenis (2008)
establishes a key property that differentiates a lead organization from
other organizations deploying network managers: The former are not only
responsible for the management of the networks, but are themselves
service-delivering actors within the network. This distinct feature distin-
guishes lead organization networks from the remaining two modes of net-
work governance. In these other modes, there either exists no central
agency but a collective steering process (self-governance) or the network
manager is exclusively responsible for network-level tasks and was estab-
lished for this specific purpose (network administrative organization). The
definition of lead organizations indicates that their activities—and particu-
larly the duties of the network manager—can vary substantially depending
on the prioritization of the two task areas (i.e., network management and
service delivery tasks). The network managers may provide a substantial
part of the overall services of the network themselves and minimize the
resources invested into network management. Alternatively, they may
32 S. HADORN
decide to primarily act as network managers and only be involved periph-

erally into policy delivery.
Similarly referring to resource allocation, Raab et al. (2015) have dis-
tinguished network managers based on their financial means available for
network management. In their study, they assess the total amount of funds
that network managers receive from their partners for network manage-
ment. However, this indicator does not account for the fact that the actual
allocation of funds is often difficult to assess. This limitation is particularly
important in the case of lead organizations, where different tasks of net-
work managers (network management and service delivery) are closely
interlinked. In practice, funds that are budgeted for one purpose (e.g.,
network management) may easily be allocated for the other task area (e.g.,
service delivery). It is therefore important to not only look at the bud-
geted amount but to also consider the previously discussed weighting of
the tasks. By considering both aspects, the amount of resources invested
in network management can be assessed most accurately. In the following,
I will distinguish between ‘high caseload’ in cases where network manag-
ers are responsible for a comparably high number of projects (i.e., when
they have many service delivery tasks) and ‘not high caseload’ where the
network managers only manage a few projects (i.e., when they have few
service delivery tasks).1 Note that I will use the term ‘low caseload’ inter-
changeably with ‘not high caseload’ for better readability.
2.2.4 Policy Design Phase

The fourth condition characterizing the work context of the network
manager is the ‘policy design phase’ and specifically refers to the inter-
organizational context during the planning phase of the cooperation. It
reflects the quality of the starting phase of the partnership within the net-
work and particularly the relationship between the various stakeholders.
Despite this book’s focus on implementation processes, the quality of the
design phase needs to be considered when studying policy implementa-
tion because policy design and policy implementation are two interdepen-
dent process steps of the policy cycle (Jann & Wegrich, 2007). Since I am
interested in which aspects affect the way a network manager manages a
network, the quality of the policy design phase is highly relevant, particu-
larly in mandated networks where studies showed that conflicts are
1
The operationalization of this binary variable is displayed in Sect. 3.2.
Another random document with
no related content on Scribd:
name of it, but a traveller figured in it, who took his servant with him
to the heart of Africa. The latter, who was passionately fond of
travelling, and took an eager interest in all the doings and
adventures of explorers, made but one request, and that was to be
allowed to change his name of Joseph to that of Mohammed Ben
Abdullah. “It was more euphonious,” he said, and the audience
roared with laughter.
Well, Joseph was quite right, and if Barth had not done as he did,
the negroes and Tuaregs would never have remembered his
European name, it would never have become engraved on their
memories, it would never have been transmitted to their
descendants, and I should not have been able to solve all difficulties,
however great, and emerge safely from every situation, however
embarrassing, by the simple words “I am the son, or rather the
nephew of Abdul Kerim.”
It is impossible to admire too much the lofty, upright character of
Barth, which so impressed all with whom he came in contact on his
journey, that nearly half a century after his death the mere fact of his
having traversed a district—poor as he was, and exposed to all
manner of dangers, the friendship of Beckay his only safeguard—
should be enough to open the way for a pretended relation of his.
How few travellers could boast of having done as much, even in
modern times. Too many explorers have indeed, after forcing their
way through a country against the will of the natives, left behind
them a legacy of increased difficulty and danger to their successors.
I was very anxious to secure the services of a political agent with
a thorough knowledge of the country, and the language of the
Tuaregs. I wished to send him, if I could find him, in advance of our
party to take letters to the chiefs, or to plead our cause with them.
Acting on the advice of Hamadi, I chose a certain Sidi Hamet,
distantly connected with the Kuntas, and then employed in the
Custom House at Timbuktu, under one Said, the interpreter of the
Post-Office.
I must do this justice to Said, he yielded with anything but a good
grace to the employment of his subordinate on our service, and did
more to dissuade him than to further our wishes. We had to invoke
the aid of Commandant Rejou, and later, at Tosaye, Sidi Hamet
piteously entreated me to let him go back, and I expect Said’s
objection to his joining us had something to do with his faltering.
However, I forgive him with all my heart. Sidi Hamet was the
interpreter’s right hand, his chief source of information on every
subject, and he found it hard work to fulfil his own duties, even those
of an interpreter, without him.
On the 16th I went back to spend a day at Kabara, where I had
invited all the notables of Timbuktu to come and listen to the
wonders of the phonograph. It was an exhibition which long dwelt in
the memory of those present. Amongst the most attentive listeners
were the two sons of the chief of the Eastern Kuntas, who lives at
Mabrok. I felt sure that the rumour of the extraordinary things I had
done would precede me.
Commandant Rejou had already warned Sakhaui, or Sarrawi,
chief of the Igwadaren Aussa, the first Tuareg tribe we should meet
on our way down the river, of our approach. In the evening two
envoys from this chief arrived with a missive, which it was almost
impossible to decipher, but from which, in spite of its ludicrous
phraseology, we managed to make out two things, one being that
Sakhaui had no desire to see us, the other that he was very much
afraid of us.
We did our best to reassure and impress the messengers, and
finally succeeded in convincing them that we had no evil intentions
with regard to the Igwadaren, and armed with a fresh document from
us they set off to return to their chief.
Meanwhile Sidi Hamet, who had been well coached in what he
was to say and do, had started on his way to Aluatta, to ask him to
meet us at Kagha, a little village on the right bank about thirty-one
miles from Timbuktu. For the first time I now announced my
pretended relationship with Abdul Kerim, taking myself the Arab
name of Abd el Kader, or the servant of the Most High.
This mission with the Kuntas accomplished, Sidi Hamet was to go
to the Igwadaren of Sakhaui and wait for us.
Having settled everything to the best of our ability, visited the
boats, and repaired any little damage which had been done by the
way, we had now only to give ourselves up to the current of the river
and to the will of God.
It was not without a certain emotion that, on Wednesday, January
22, we started from Kabara, seen off by all our brother officers of the
garrison of Timbuktu, and escorted to our boats by a great crowd of
natives, who, with more or less enthusiasm, invoked the protection of
Allah on our behalf.
WE LEAVE KABARA.
As long as our boats were in sight of the station we could see

handkerchiefs and helmets waving to us in token of adieu, and when
the flag of the fort disappeared from view our hearts felt somewhat
oppressed, for we were leaving all that in our exile from France
represented our native country. Henceforth we five white men, with
our twenty-eight black followers, were thrown on our own resources,
and had to depend upon ourselves alone. How many of us would
return? How many of us were destined to sleep our last sleep
beneath the soil of Africa?
AT TIMBUKTU.
DROVE OF OXEN.
CHAPTER III
FROM TIMBUKTU TO TOSAYE
On January 22 we made a very short stage, and moored about

mid-day at Geïgelia, a little village picturesquely perched on rising
ground of a reddish hue, a little below the mouth of the stream,
which, as I have said, gives access to Day and Kabara.
We determined to spend the afternoon in making things snug.
Hitherto we had put off from day to day the task of arranging our
cabins. Now our three little craft are all the world to us, the floating
castles which must drift down with us to the sea, Inch Allah! (God
willing), as the Mussulmans say. We must shake down in them as
comfortably as possible.
I occupy the fore-cabin of the Davoust. On one side is my plank
bed, with, for mattress, the wrappings of the presents which we take
out one by one as required, replenishing them from the bales in the
hold.
On the other side is a big table, with packing cases serving as
legs. Everywhere books and instruments, an iron chest containing
the more valuable presents: caftans, velvet burnous, gold-
embroidered turbans, etc.
On the mats which cover the partitions, I have fixed the
photographs of a famous singer, purchased in the Rue de Rivoli, in a
moment of musical enthusiasm. I found them by chance at the
bottom of a trunk, into which they must have slipped when I was
leaving France. These portraits, as will be seen, played a part in the
politics of the Niger. Opposite them, an engraving of the President of
the Republic, or rather, Sultan of France, as they call him here. Nor
must the phonograph in its ingenious case be forgotten, with the
voltaic piles, Geissler’s tubes, little electric lamps, forming a fairy
crown, which is lighted on pressing a button. Such articles as have
hitherto figured in the baggage of none but the passengers of Jules
Verne.
The after-cabin is Père Hacquart’s sanctum, and also the arsenal.
The Father rests peacefully on a couch of rice and biscuit tins, with
the conventional bedding of package wrappings; on the partitions,
the ceiling, everywhere, I have fixed guns for presentation, revolvers,
etc., for exchange; a goodly number of cases of cartridges,
moreover, give this retreat the appearance of an armoury. On the
bridge, all round the machine-gun, are more cases, which serve as
benches for the oarsmen.
Our hold is a masterpiece of packing, due to Baudry’s ingenuity. I
defy the most skilful to insert as much as a needle more. On board
the Aube, the fore-cabin protects Baudry and Bluzet; there is not
much elbow-room for two, in such a confined space. The hinder one
is reserved for Taburet and his medicine chests.
The little barge Dantec, also provided with a shelter, will carry all
surplus articles. At present, until it is used as an infirmary, which I
trust may be as seldom as possible, it is the refuge of the destitute,
where Suleyman, the interpreter, and the Arab translator, Tierno
Abdulaye Dem, are quartered.
I may now describe more minutely our dusky auxiliaries. At first
there were twenty coolies, reduced to nineteen by the defection of
Matar Samba. Their head-man Digui, of whom I have spoken, will be
judged by his deeds; there is no need now to mention all the
blessings he deserves.
THE ‘AUBE’ AND HER CREW.
Suleyman Gundiamu and Tierno are the scholars of the party.

Suleyman almost speaks French, although he says la noce for un
os, cherchicane for certificat, and always translates keffir (Infidel) by
Christian. As for Tierno, he is a sly, cunning dog, of whose fidelity I
have often had my doubts: evidence is against me, however, and I
must do him the justice to say, that on all occasions he has sided
with us against his co-religionists, his compatriots, and even his
relations. Idle as a dormouse in everything but writing Arabic, but
isn’t that just what he is for? Altogether he is not a bad boy, and we
should scarcely find a better fellow amongst his people.
Our carpenter, Abdulaye, is a big Wolof, strong as Hercules,
intelligent, only idle by fits, and not very serious ones either.
“Abdulaye, something has gone wrong with your working hand!”
This is how we call him to order; if the appeal is not attended to, a
good blow follows as punishment; Abdulaye is aware of his guilt,
sets to work again, and does the tasks of four.
Abdulaye is certainly not a marabout. He is even addicted to
spirituous liquors, but he has not had many opportunities on the
journey of indulging this taste; he was, however, overcome on our
arrival at Dahomey. For six days we never set eyes on him, for he
was never sober.
My first acquaintance with Abdulaye arose from his love of the
bottle. In May 1894, when I took command of the flotilla, Abdulaye
having found the door of the store-room open, gave way to his
propensity, and I found him dead drunk beside a very respectable
number of empty bottles. The awakening was anything but pleasant,
and Abdulaye never forgets the capers he cut on that occasion.
Such is our staff, or I should rather say, these are the native officers
of our expedition. Besides this, each of us has his own servant. Mine
is Mamé, an intelligent Saracolais, who speaks Songhay, the
language of the blacks on the banks of the Niger from Jenné to
below Say. He is a very faithful and devoted lad; the point about him
is the excessive deliberation of every motion, which gives him
something of the appearance of a chameleon. Lucky fault, or rather
precious gift, which all who have been served by Sudanese will
appreciate. Thanks to it, Mamé has never broken anything of mine.
Baudry’s servant’s name is Mussa; his father is head-man of
Diamu, a village on the banks of the Senegal. He is the philosopher,
the learned man of our military establishment. He reads and writes
French pretty correctly, but his studies occupy some of the attention
due to his master. If Baudry has employed his talents as a teacher to
the full on a most willing pupil, in return his boots have seldom been
blacked—or rather greased—in the course of the voyage.
Fate decreed that Bluzet should have as servant a son of the
blacksmith of Mussa’s father. Fily is his name, and by reason of his
parentage he is the confidant and devoted slave of Mussa.
Provided he is treated firmly, Fily is an excellent servant, and a
cook of the first order (for that country, be it understood), and the
cakes we used to call his nougats aux arachides, have often been
fully appreciated at our table.
Lastly, Father Hacquart and Taburet have two boys at their
disposal, both answering to the name of Mamadu; to distinguish
them one is called Father Mamadu, the other Doctor Mamadu.
Add to these a yellow dog, Meyer by name, why so called I cannot
say, and the menagerie is complete.
We did possess two cats, one an excellent swimmer, in spite of all
preconceived notions; but these little animals, who behaved
themselves anything but decently on board, disappeared in the
course of a very few days.
In spite of his denials, I have always suspected Bluzet, a sworn
enemy of the feline race, of aiding and abetting their desertion, for
they seemed to have a special grudge against him.
I have forgotten old Abdul Dori, but he did not make a long stay
on board. I have already mentioned that I suspected him of evil
designs in taking service with us. He got me to advance him a pretty
round sum on the voyage to Massina, which he said he owed to one
of his countrymen, and desired to repay before he entered upon a
venture so full of danger. As soon as the sly rogue had gained his
end, he changed his tactics. From Sego, according to him, the
voyage would be comparatively easy. His debt paid, he attempted to
terrify my coolies, telling them the most ridiculous tales about the
ferocity of the Tuaregs, and giving the most discouraging account of
the rapids, which in the end we unfortunately found partly true.
He soon discovered he was wasting his time. My men came of
their own accord, and reported that Abdul was trying to dishearten
them. I soon made him understand I would not stand that kind of
thing. Seeing the failure of these manœuvres, and in no way anxious
to remain with us, he shammed sickness, pretending to be attacked
with dysentery. The doctor soon discovered the trick, and I told him
that, ill or well, he would have to follow me.
His plan having miscarried, he set about making himself really ill,
and lay down to sleep without any covering on the chilliest nights. At
this game, if he did not procure the dysentery of his dreams, he at
least contracted inflammation of both lungs, which developed the
very day of our departure. He remained two days longer with us;
then really seriously ill, he became delirious. Moved with pity, I
decided to send him back to Timbuktu in a canoe hired at the village
of Burrem. I don’t know what became of him, but I advise those who
may come across him hereafter, and are deceived by his honeyed
words and ways, to beware of him. As far as we are concerned, I
consider it a blessing that his cowardice overcame the desire for
doing evil. He might have proved a great source of danger,
especially at Say, his native place, where he would have aided and
abetted our enemies.
The first and most important object of the expedition was to trace
as correctly as possible the course of the river which we had to
follow. For this purpose I had observing instruments of very accurate
construction made for each barge, which would afford us the means
of making a triangulation of the river en route. Two barges were to
coast along the banks, while the third kept in the deep channel.
We tried this plan on January 23, the first day on which we
navigated an almost unknown region. It was soon found
impracticable. By evening we had gone less than four and a half
miles. At this rate, counting necessary stoppages, it would take a
year to reach the mouth of the river. We therefore adopted the
following plan: the Davoust followed the left bank; the Aube the right
one while on surveying duty, the two barges frequently taking their
places.
At the same time, Baudry on the Dantec tacked about in search of
the deep channel, taking frequent soundings. Any inaccuracies were
guarded against by taking the mean draught of the two larger
vessels, and constantly determining the position by astronomical
observations.
This system was invariably followed down to Ansongo, that is, for
the whole navigable course of the Niger. Though we did not secure
the accuracy of a regular survey, still to me it appeared quite
enough; for the first vessels that might come after us, will possess an
indication of the position of the deep channel relatively to the banks
and their configuration, the distances from one point to another, the
position of the villages, and the peculiarities of the soil, etc.
Below Ansongo, in the region of the rapids, Baudry and I had to
abandon all survey work, and devote our attention exclusively to the
boats. Bluzet completed the map, which is of no practical value, as it
is impossible to determine any navigable channel, especially for
steamers, in those dangerous rapids. The only object of its existence
is to prove that a navigable channel does not exist. So that all that
can be done is to choose the least undesirable means of access to
the Western Sudan from among the many that have been proposed.
After passing the villages of Koa, Burrem and Bori, where the
people came out in canoes with presents of goats, sheep, eggs and
poultry, we arrived abreast of Kagha, about one o’clock on the 25th.
The moment we reached the mouth of the creek which leads to it—
for the village is not on the main stream, but a little inland—we were
hailed from a canoe by a great giant with an intelligent face and
woolly hair, forming a halo round his head, which was more
picturesque than clean. He was a Kunta, knew French, had been in
the villages of Mediné and Nioro, in the French Sudan, and even
spoke a little Soninké, the maternal language of most of our coolies.
He acted as pilot for us, but, in spite of all his efforts, we could not
get up to Kagha, for there was not sufficient depth of water; so we
had to pitch our camp at the foot of a little hill covered with dwarf
palms rather more than a mile from the nearest huts.
A deputation of the Kuntas of the village soon joined us, who told
us that Sidi Hamet had arrived two days before with my letter for
Aluatta; but the latter was from home, and no one knew exactly
where to find him, nor if my missive had reached him.
In fact, fifteen days before, a band of Kel Gossi, a Tuareg tribe
whose territory is about the centre of the bend of the Niger, had
carried off a hundred head of cattle belonging to the chief of the
southern Kuntas; Aluatta had set off to overtake the raiders, and
induce them in the name of Allah and Mahomet to restore their ill-
gotten gains.
However extraordinary the following custom may appear, it is
actually prevalent in the Tuareg districts. One tribe steals from a
neighbour all or part of his herds; if the latter is not strong enough to
recover by force that which he has been deprived of, he tries
conciliation, and generally regains, if not all, at least a portion of his
chattels. This invariably occurs when the injured party is a marabout,
and be it remembered these raids do not involve war: the same Kel
Gossi will be quite prepared to come the next day to ask Aluatta to
implore for them the protection of Heaven, and to purchase
talismans from him.
Whatever the result, this troublesome episode made me fear I
should not see Aluatta. Unable to confer with him, I betook myself to
his relations and endeavoured to secure their friendship, telling them
the story of my connection with Barth, or Abdul Kerim.
This produced a marvellous change in their demeanour; reserved
before, they became most cordial. To strengthen the effect still
further I brought the phonograph into play. One of the head Kuntas
sang an Arab song in his tent. It was really the battle hymn of Hamet
Beckay, the friend of my “uncle,” and it was really something to see
the amazement of all when the instrument repeated the song. From
that time we were the best of friends. All expressed their regret that I
could not have a palaver with their chief. “Not wishing to deceive
you,” said they, “we will not promise a visit from Aluatta, but, if you
like to wait, you shall see his brother, Abiddin, who at this moment is
at Arhlal, about twelve miles away. We will send and fetch him at
once.”
The proposal pleased me too much to be refused, and the
messengers departed.
Along with our friends the Kuntas, there came a little band of
Tuareg Kel Temulai, who lived further down stream in the direction of
Ganto, who were evidently sent to give information.
They were tall, strong fellows, spare and active. As this tribe has
no camp on the banks of the river, I told them I should ascend the
creek which leads to Ganto for the purpose of seeing them. In fact, I
wished to ascertain their intentions. The Kel Temulai were one of the
two tribes which divided the dominion of the region around Timbuktu;
Kabara and the southern portion of the plain which surrounds the city
belong to them. The French drove them from it, and they fell back
towards the east, gathering round their chief Madunia, who lived
near Ganto and was more than a hundred years old.
On the next day, the 26th, a despatch actually arrived, which the
Commandant of Timbuktu had managed to send on to us by canoe.
A fortnight later we were to receive yet another at Rhergo, and our
delight may be imagined, for we had had no news from home for ten
months.
In the afternoon Abiddin arrived. Tall, strong, and well-made, he
looked anything but amiable, and was far from communicative. I
confess his first appearance struck me as anything but pleasing. He
was by no means anxious to get into our good graces, and replied
very dryly to my protestations of friendship. We talked together for
about an hour, but I failed altogether to mollify him, and I began to
despair of bringing him round.
In the evening I found out something more about him, and the
position he occupies in the country. He is older than Aluatta, but from
his very boyhood he showed such a warlike disposition, and one so
very unlike the gentle nature which is naturally expected of a
marabout, that his father named Aluatta his successor instead of
him, refusing him the baraka or paternal blessing usually bestowed
on the first-born. Does not this remind one of the story of Jacob and
Esau?
However, Abiddin did not seem to mind the elevation of his
brother to what should have been his own position as religious chief
of the Kuntas, but devoted himself gladly to the direction of the
warlike expeditions of his tribe.
He seems to excel as a leader, and the Kel Antassar, the tribe
which longest resisted French influence in the districts round
Timbuktu, knew something of his valour. At the head of a little body
of men he surprised their camp at least a hundred times, and I now
began to understand the real reason why Abiddin had treated me so
coldly; he would have liked to have been allowed to take his part in
the play now that, after what he thought our culpable inactivity of
more than a year, we had again made up our minds to act. This
would have given him a fine chance of revenging himself on his old
enemy N’Guna, the chief of the Kel Antassar. It really was a pity that
the authorities at Timbuktu had ignored the existence and the
character of such a man. If only as guides, he and his Kuntas would
have been admirable auxiliaries for us.
We concocted a diplomatic plan to win the confidence of Abiddin.
When he came to see us the next morning I dwelt much upon my
relation to Abdul Kerim, and I roused his curiosity by showing off the
phonograph. Then when his manner became a little less churlish, I
held my peace and let Father Hacquart have his turn. The father
began by taking him roundly to task in Arabic for his want of
politeness and amiability. He actually brought Abiddin to
acknowledge himself in the wrong, and ended by getting him to
promise not only to help us himself, but to give us recommendations
to his friends. In the evening he actually returned bringing us three
letters, one for Salla Uld Kara, another for a certain sheriff named
Hameit, whom we should meet beyond Al Walidj, and the third and
most important for Madidu, chief of the Awellimiden Tuaregs.
This letter for Madidu simply delighted me. I was to some extent
already acquainted with the various tribes we should have to deal
with on our way down the river. The first were the Igwadaren, divided
into two sections hostile to each other, under two chiefs, brothers,
though enemies, named Sakhaui and Sakhib. Beyond them we
should come to the Kel Es Suk, marabouts of the great Tuareg
family, a small tribe of the Tademeket Kel Burrum, to whose chief,
Yunes by name, Abiddin also gave us a letter; and beyond them, that
is to say, after passing Tosaye, we should enter the territory of the
great Awellimiden Confederation, but how far it extends I did not as
yet know.
Abiddin, who had passed a month with the chief of the
Awellimiden a year ago, could not say enough in his praise, whilst,
on the other hand, he warned us very earnestly against the small
tribes addicted to pilfering through whose districts we should have to
pass to begin with. “Madidu,” he said, “is a lion, the other chiefs are
mere jackals!”
“Madidu,” he added, “makes war, and of course the plunder he
takes in war is a lawful prize, but he would scorn to pillage peaceful
folk, such as the negro cultivators of the soil, or inoffensive
merchants with no one to fight for them, in the reckless manner of
the Kel Temulai or the Igwadaren. There is no one higher than
Madidu unless it be God.”
Of course I knew that Abiddin exaggerated, as all Orientals do,
and that much of his enthusiasm for Madidu was only cupboard love,
the result of the good cheer he had enjoyed in his camp. Still I
gathered from what he said that his chief really was somebody worth
reckoning with. Writing to the Lieutenant-Governor of the French
Sudan by the returning canoe which had brought our despatches I
said, “I am now pretty well convinced that if Madidu really wishes it
we shall pass without hindrance, but that if he opposes us we shall
have the greatest difficulty in going down the river.” This was,
however, but a façon de parler, for I was mentally resolved that, with
or without Madidu’s help and permission, we would go down the
Niger, though if he did try to prevent us, we should most likely leave
our bones in the river.
It will readily be understood how much this passage through the
Awellimiden district occupied our thoughts. It was the chief subject of
all our talks with Abiddin, and we had every reason to congratulate
ourselves on having so far won him over. But we meant to do far
more than that. He was altogether our friend now, and never left the
boat except to eat. I reminded him of the former grandeur of his race,
of Sidi Moktar and his brothers, who had acted as mediators
between the tribes of the neighbourhood, and pointed out to him that
it was the outburst of fanaticism, against which his grand-uncle had
struggled so hard, which had led to the decrease of the influence of
the Kuntas. We too, I told him, had to contend against those who
propagated the doctrines declared by Hamet Beckay to be false and
contrary to the true morality of Islam, and we had succeeded in what
that great man wished to accomplish, for we had driven back the
invading Toucouleurs.
If, I urged, we whites, who had considerable forces at our
disposal, made a firm alliance with the Kuntas, who would in their
turn place at our service all their religious influence, their ancient
power would be restored, they would be our trustworthy agents,
working loyally for the pacification of the country, which would owe to
them all the benefits of peace, for which they would never cease to
be grateful.
On the other hand, I pointed out, that if we made an alliance with
the Awellimidens, whose lands we did not in the least covet, all the
small pilfering tribes, such as the Igwadaren and the Kel Temulai,
would be compelled to cease their depredations, because all the
merchants on the river would be under the protection of the French,
or of their new friends. Placed as they would necessarily be between
us and the Awellimiden, they could not without risk of destruction, or
at least of severe reprisals, insult either of the two contracting
parties.
Abiddin seemed much taken by my arguments, which appealed
forcibly to his sympathies and intelligence. He was a decidedly
clever fellow, and I struck whilst the iron was hot, by adding that it
seemed to me that this proposal, if made to the Kunta chief, would
solve the problem of the pacification, and add immensely to the
value of the districts surrounding Timbuktu.
We should very soon relieve those under our protection from all
fear of molestation by the Tuaregs, we should promote the creation
of centres of commerce and outlets for trade, and moreover, we
should greatly reduce our expenses at Timbuktu, for our gains would
help us to pay and support the troops quartered in that town.
“It is evident,” answered Abiddin, “that if you could come to terms
with Madidu, and be really friends with him, it would be a very good
thing for us all. We shall, however, want somebody to act as go-
between, but the question is, whom could we choose.”
“Houa!” (thou), said Father Hacquart, suddenly striking into the
conversation. Abiddin started; the idea that he might go himself had
evidently not occurred to him. The father now put out all his
eloquence to persuade him, and finally won his consent.
Abiddin spent the whole of the next day with us, and asked the
doctor to give him some medical advice, for he suffered greatly from
rheumatism and cystitis. I arranged with him that we should go to
Rhergo, and there wait for news from him. If he should send us word
to go on we should know that he had already passed us, and was en
route for Madidu’s camp.
On the 29th, despairing of seeing Aluatta, who was still
negotiating with the Kel Gossi, we decided to leave Kagha, but we
had scarcely left our moorings when we were met by such a violent
wind that it was absolutely impossible to proceed, and we went to
take refuge in an opening on the left bank. It was not until after a
delay of two hours that we were at last able to go on and anchor
opposite Milali. We were asleep, when our watch aroused us with
the news that a canoe was approaching, the man in which was
shouting out something at the top of his voice. It turned out to be a
courier from Aluatta, who had at last received our despatch, and
would come the next day to Kagha, where he begged us if possible
to return.
Only too glad to hear from him at last, we went back the next day,
and about four o’clock in the afternoon Aluatta came to see us with
his retinue of followers. He was a handsome young fellow, with a
very dark skin and a most intelligent face, a gentle but rather proud
expression. He is supposed to have the gift of prophecy, and to be
able to perform miracles. It is said that he predicted the death of
Tidiani, a former chief of the Massina, a year before it took place.
Everything having already been settled with Abiddin, Aluatta had
only to ratify our agreement with his brother, and this he did readily.
Of course we showed off our phonograph and bicycle to our visitor,
and a telescope greatly aroused his admiration, because he was
able to see and recognize the people of Kagha through it. We spent
the whole January 30 with Aluatta, and then, this time in earnest, we
resumed our voyage.
INTERVIEW WITH ALUATTA.
We were dreadfully hindered by a strong contrary wind from the

east, and it was not until February 3 that we arrived at Kunta, where
we were to see the Kel Temulai.
At our approach the negroes of the village (the Tuaregs have their
encampment on the opposite side of the river some little distance
inland) at once begun carefully to sweep the bank where we should
disembark, and very soon our tent was up, our camp-stools were
beneath its shelter, and our visitors the Kel Temulai arrived, including
R’alif, the brother of R’abbas, chief of the tribe, with the two sons of
the latter and a small retinue.
The palaver was carried on under difficulties for want of some one
understanding the Ta-Masheg or the Tuareg language, and we had
to converse in Songhay, our servant Mamé acting as interpreter. This
was the first time we had seen the Tuaregs in their own land, and we
were all deeply interested in them. They are many of them very finely
built fellows, and their features, all you can see of them, for the lower
part of their faces is always obstinately hidden by the tagelmust or
veil, are of a purer Kel Temulai type than I have ever seen
elsewhere. They all wear breeches coming down to the instep, and
mantles, or as they call them bubus, of dark blue material. The more
important members of the tribe have a kind of pocket of red flannel
on their breasts. In the right hand they hold an iron spear some six
feet long, and on the left arm a dagger is kept in place by a bracelet
without causing its owner the slightest inconvenience, so that it is
always within easy reach of the hand, and can be used at a
moment’s notice. Lastly, a few of them also have a straight sword
with a cross for a hilt, reminding us of those in use in the Middle
Ages, and which is hung on the left side by a rope.
A LITTLE SLAVE GIRL OF RHERGO.

Network Management and Governance in Policy Implementation The Case of Smoking Prevention Programs Susanne Hadorn Full Chapter

Uploaded by

Document Information

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

Network Management and Governance in Policy Implementation The Case of Smoking Prevention Programs Susanne Hadorn Full Chapter

Uploaded by

Copyright:

Available Formats

Network Management and Governance

in Policy Implementation: The Case of

ISSN 2524-7301 ISSN 2524-731X (electronic)

I am highly grateful for the patient and continuous support by my supervi-

4 Swiss Smoking Prevention Policies 97

Fig. 1.1 Research questions. (Notes: The numbers designate the

Fig. 4.8 Cantonal smoking prevention program of the canton of Jura.

Table 2.1 Possible configurations of four conditions characterizing

Table 6.2 Models 4 and 5 (Bayesian ordered logistic regressions) 188

Management and Performance

This book investigates the management of networks delivering public ser-

© The Author(s), under exclusive license to Springer Nature 1

structural characteristics of networks, and contextual factors. The primary

1. How do different constellations of factors characterizing a network

The main relationships analyzed are captured in Fig. 1.1 in simplified

Box 1.1 Data and Methodological Approach

Box 1.1 (continued)

Box 1.1 (continued)

1.1 The Age of Collaborative Service Delivery

are thus increasingly confronted with the need for inter-organizational

1.2 Definition of the Concept ‘Networks’

Fig. 1.2 Schematic illustration of the policy delivery networks

degrees—project management responsibilities (referred to as ‘caseload’ in

1.3 Determinants of Network Performance

themselves service providers. This book focuses on this last governance

1.4 Deficits in the Network Literature

instance, Provan and Kenis (2008, p. 244) argued in a conceptual study

empirical studies examining the connection between network manage-

characteristics that might affect policy implementation. A further impor-

Box 1.2 The Book’s Results in a Nutshell

Box 1.2 (continued)

Box 1.2 (continued)

Box 1.2 (continued)

Public Administration Research and Theory, 18(2), 253–273. https://doi.

Klijn, E. H., & Koppenjan, J. F. M. (2014). Accountable networks. In M. Bovens,

Stufenmodells und eine praktische Anwendung. Swiss Political Science Review,

Linking Characteristics of Network

The aim of this theoretical chapter is to develop the analytical framework

© The Author(s), under exclusive license to Springer Nature 27

management activities in the third sub-chapter. The aim is to make theo-

2.1 A Synthesis of the Analytical Framework

Factors distinguishing Network Network Network-level factors

2.2 Conditions Distinguishing Different Network

2.2.2 Legal Status

network members is undisputed, the majority of the network (manage-

decide to primarily act as network managers and only be involved periph-

2.2.4 Policy Design Phase

As long as our boats were in sight of the station we could see

FROM TIMBUKTU TO TOSAYE

On January 22 we made a very short stage, and moored about

Suleyman Gundiamu and Tierno are the scholars of the party.

We were dreadfully hindered by a strong contrary wind from the

A LITTLE SLAVE GIRL OF RHERGO.

You might also like

4 Swiss Smoking Prevention Policies 97

Table 6.2 Models 4 and 5 (Bayesian ordered logistic regressions) 188