Global Environmental Change 19 (2009) 366374

Contents lists available at ScienceDirect

Global Environmental Change

journal homepage: www.elsevier.com/locate/gloenvcha

Review

The role of social networks in natural resource governance:

What relational patterns make a difference?
rjan Bodin a,b,*, Beatrice I. Crona a
O
a
b

Stockholm Resilience Centre, Stockholm University, SE-106 91 Stockholm, Sweden

Centre for the Study of Institutional Diversity, Arizona State University, Tempe, AZ 85287-2402, USA

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 6 November 2008
Received in revised form 19 May 2009
Accepted 20 May 2009

Resent research has identied the existence of social networks as a common and important denominator
in cases where different stakeholders have come together to effectively deal with natural resource
problems and dilemmas. It has even been shown that social networks can be more important than the
existence of formal institutions for effective enforcement and compliance with environmental
regulations. However, all social networks are not created equal. On the contrary, the structural pattern
of relations (i.e. the topology) of a social network can have signicant impact on how actors actually
behave. This clearly has implications for actors abilities to manage environmental challenges. This
review aims to add more precision to initial insights and pending hypotheses about the positive impacts
of social networks on governance processes and outcomes, by reviewing and synthesizing empirically
based literature explicitly studying structural characteristics of social networks in natural resource
governance settings. It is shown that signicant differences in governance processes and outcomes can
be expected among networks experiencing structural differences in terms of density of relations, degree
of cohesiveness, subgroup interconnectivity, and degree of network centralization. Furthermore, the
review shows that none of these structural characteristics present a monotonically increasing positive
effect on processes of importance for resource governance, and that favoring one characteristic likely
occurs at the expense of another. Thus, assessing the most favorable level and mix of different network
characteristics, where most of the positive governance effects are obtained while undesired effects are
minimized, presents a key research and governance challenge.
2009 Elsevier Ltd. All rights reserved.

Keywords:
Natural resource governance
Social networks
Network analysis
Environmental management
Socialecological systems
Natural resource management

1. Introduction
Governance1 of ecosystems is inherently difcult since both
the natural environment and human societies are characterized
by uncertainties, complex dynamics, natural variations and scale
dependencies (e.g. Levin, 1998; Berkes et al., 2003). Furthermore,
they do not abide by human-made jurisdictions and administrative borders, and it is not possible to divide them into separate,
self-supporting, autonomous components. Many of the services
they provide are common pool resources with multiple actors2
competing for use, often leading to resource depletion or
management conicts (Hardin, 1968). Hence, management of

* Corresponding author at: Stockholm Resilience Centre, Stockholm University,

SE-106 91 Stockholm, Sweden. Tel.: +46 8 674 7671; fax: +46 8 674 70 20.
. Bodin).
E-mail address: [email protected] (O
1
By governance we refer to the management of natural resources, as well as the
structures and processes that provide the social and institutional environment in
which the management can take place.
2
By actors we refer to all possible stakeholders (e.g. resource users/extractors,
state agencies, NGOs, land owners, etc., see Fig. 1).
0959-3780/$ see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.gloenvcha.2009.05.002

any given resource would benet from actors agreeing on

common rules and practices, coordinating usage, engaging in
conict resolution, negotiating various tradeoffs, sharing information, and building common knowledge (e.g. Folke et al., 2005).
Research shows that top-down centralized management is poorly
suited for this (Ostrom, 1990; Holling and Meffe, 1996; Gunderson
et al., 1995; Berkes and Folke, 1998; Pretty and Ward, 2001).
Attention has therefore been directed at governing systems where
multiple actors to various degrees are involved in the governing
processes. These ideas are captured in the concept of comanagement (see, e.g. Carlsson and Berkes, 2005), where the
underlying rationale is that by involving different actors in the
governing process, the complexities inherent in both ecosystems
and the social arrangements constructed around these (i.e.
coupled socialecological systems, see Berkes and Folke, 1998)
can be more adequately addressed. Adaptive co-management is a
recent expansion of the co-management concept (e.g. Armitage
et al., 2009) with explicit focus on the adaptability of the joint
management process in response to environmental change and
the continuous acquisition of new knowledge (cf. adaptive
management, see Holling, 1978).

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

However, history, contemporary science and politics all suggest

that joint management and governance processes are often difcult,
albeit not impossible, to achieve in practice (e.g. Ostrom, 1990; Hahn
et al., 2006; McClanahan et al., 2008). To address contemporary
natural resource problems ranging from local sh stock depletions to
climate change and declining global ecosystem services, it is
therefore imperative to better understand how collaborative
barriers can be overcome (cf. Dietz et al., 2003). Resent research
has identied the existence of social networks (Fig. 1) as a common
and important denominator in cases where different stakeholders
have come together to effectively deal with natural resource
problems and dilemmas (e.g. Gunderson, 1999; Hahn et al., 2006;
Folke et al., 2005; Pretty and Ward, 2001; Olsson et al., 2008; Grafton,
2005; Scholz and Wang, 2006). It has even been shown that social
networks can be more important than the existence of formal
institutions for effective enforcement and compliance with environmental regulations (Scholz and Wang, 2006). Social networks can
improve collaborative governance processes by facilitating, (i) the
generation, acquisition and diffusion of different types of knowledge
and information about the systems under management (Crona and
Bodin, 2006; Isaac et al., 2007; Schusler and Decker, 2003), (ii)
mobilization and allocation of key resources for effective governance
(e.g. Carlsson and Sandstrom, 2008; Carlsson and Berkes, 2005;
Newman and Dale, 2007), (iii) commitment to common rules among
actors fostering willingness to engage in monitoring and sanctioning
programs (Dietz et al., 2003; Scholz and Wang, 2006), and (iv)
resolution of conicts (Hahn et al., 2006). However, all social
networks are not created equal (Bodin et al., 2006; Newman and
Dale, 2005). On the contrary, the structural pattern of relations (i.e.
the topology) of a social network can have signicant impact on how
actors actually behave (Degenne and Forse, 1999; Wasserman and
Faust, 1994).
Emerging recognition of the importance of social networks for
outcomes in natural resource governance has resulted in an
increase in empirical studies analyzing the structural characteristics of these networks, and more is under way. This increase is at

Fig. 1. Different sectors of society involved in the use and management of the
natural environment are schematically represented as different triangular slices.
The red dots represent individual organizations or persons (i.e. actors) within each
sector, and the lines represent relational ties among these. These relational ties can
contribute to better natural resource governance by, for example, facilitating
coordinated actions among different actors.

367

least partly driven by the recognition that despite an obvious

decline in, or degradation of, natural resources, and in spite of the
best intentions, many governance initiatives around the world are
failing. Analyzing networks of various stakeholders helps tease
apart how social structures, created by the pattern of relations,
enhance or hinder these initiatives. The profound effect of social
networks on social processes in general is well documented by
social scientists, and has lead to the development of the eld of
social network analysis (SNA) (Freeman, 2004). To date, empirical
research in natural resource governance applying quantitative
analyses of networks characteristics is, however, still limited. Yet
valuable insights and hypotheses have started to materialize in this
evolving eld, and these are synthesized and discussed here. We
begin by reviewing how certain structural characteristics of social
networks inuence the key social processes which are of particular
interest for natural resource governance. We also discuss the
challenges in addressing the intricate balance between several of
these structural network characteristics for governance outcomes.
The paper concludes with ideas and perspectives for further
research in this area.
2. Relational patterns and processes
To better understand how social networks affect natural
resource governance processes, one needs to start by acknowledging some key characteristics which differentiate social networks. First, the pattern of relations will differ depending on which
network is in focus, i.e. depending on the type of relations involved.
We refer to these different patterns of relational ties as structural
characteristics of networks, and the effect they have on social
processes such as knowledge transfer, information sharing,
consensus building and power relations will be explored here.
Secondly, and as eluded to above, the content of the relational ties
between actors is fundamentally different in different networks.
The ties of a network of close kin will likely convey different
resources compared with a network of work colleagues. Similarly,
a network for transfer of ecological knowledge is different from a
network of relations used for accessing shing gear. Therefore,
when studying social networks, it is important to specify which
kind of relations are being studied, and how they relate to the
research question and governance issue at hand. To date, however,
most empirical studies addressing social networks in natural
resource governance have treated them as being either present or
absent, and rarely have structural characteristics been explicitly
measured and formally analyzed. Although network structure may
not be of primary interest by itself, social processes which
underpin the outcome of resource governance are enhanced or
inhibited by different structures. Likewise, but from another angle,
the same social processes can also, over time, lead to changes in the
network structures (see, e.g. Borgatti and Foster 2003 for a
discussion of how networks can be studied both as explanatory as
well as outcome variables). Analyses of the relation between
structure and process can thus help us understand the complexities of resource governance. This is the rationale behind this
review and in the following sections we will discuss how the
number of ties, degree of network cohesion, subgroup interlinkages, network centralization, as well as actor centrality affect
resource governance. This categorization of different structural
characteristics are typically used in social network analysis (see,
e.g. Wasserman and Faust, 1994), and are also reected in the
reviewed studies.
3. Few or many social ties?
Simply speaking, to turn a set of isolated actors into a set of
interacting actors, social relations have to be created among them.

368

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

Thus, a reasonable starting hypothesis would be that the more

social ties, the more possibilities for joint action and other kind of
collaborations that would help actors avoid erce resource
conicts and instead facilitate the development of common
resource regulations. Using network terminology, we can express
this as the higher the network density (i.e. the number of existing
ties divided by the number of possible ties), the more potential for
collective action. Increased levels of collective action would thus
result from increased possibilities for communication and, over
time, by increased levels of reciprocity and mutual trust (e.g. Pretty
and Ward, 2001; Janssen and Ostrom, 2006; Axelrod, 1997;
Putnam, 1993). Several empirical studies, across different disciplines, support this general hypothesis (see, e.g. Diani, 2003a).
In the natural resource governance literature, the positive
relationship between density and joint action is also supported.
Qualitatively, it has been demonstrated that by including various
stakeholders, and by fostering the development of relations among
them, chances for collaborations and joint action increases (Hahn
et al., 2006; Bebbington and Perreault, 1999; Olsson et al., 2004;
Gunderson, 1999). Using a quantitative approach Sandstrom
showed the impact of density on collective action among different
natural resource managers in northern Sweden (Sandstrom, 2008).
She observed that joint action beneted from increased tie density,
and that it was especially important that many relational ties
existed between actors of different kinds (e.g. between recreational shermen and governmental ofcials). Similarly, in a study
of shermen in rural Kenya, King (2000) showed how a group of
local shermen, by interacting with an increasing number of
government ofcials and other inuential individuals, and by
encouraging them to also interact among themselves, managed to
deal with a series of unfavorable developments related to the
shery.
Having social relations does not only lead to increased
possibilities for joint action, it may also enhance development
of knowledge and understanding through exposure to new ideas
and an increased amount of information. In agriculture and
agroforestry, it has been shown that important information on new
technologies and more sustainable management practices to a
large extent ows (or diffuses) through informal social ties (e.g.
Isaac et al., 2007; Conley and Udry, 2001). It is important to note
that the information which ows through such networks can be of
both external origin (e.g. new techniques in using pesticides in
agroforestry), as well as generated within the network (e.g. which
species are most suitable for the local ecosystem) (Isaac et al.,
2007). In addition, the study by Sandstrom (2008) showed that the
positive effects of social networks on knowledge development
seems to require less ties compared to what seems to be needed for
enabling collective action.
Some caution is warranted however, since there is also
evidence that the positive effect of network density in natural
resource governance is not necessarily continuously increasing
but might in fact decline at high densities. Very high tie density
can, in fact, reduce a groups effectiveness in collective action (Oh
et al., 2004). Furthermore, both empirical work and theoretical
multi-agent simulations of interacting resource users suggest that
excessively high network density can lead to homogenization of
information and knowledge which results in less efcient
resource use and/or reduced capacities to adapt to changing
conditions (Bodin and Norberg, 2005; Little and McDonald, 2007;
Ruef, 2002).
This section looked at the effect of the number of social ties per
se on governance processes. Social ties are, however, rarely
distributed equally among actors. In the following sections we
will discuss how more complex patterns of tie distribution can
affect processes such as consensus building, and knowledge
transfer and build-up.

4. Level of network cohesion

One important characteristic of a social network is the level of
cohesion, i.e. to what extent the network hangs together instead
of being divided into separate cohesive subgroups (Wasserman
and Faust, 1994). Thus, a network with high structural cohesion
lacks a set of clearly distinguishable subgroups (see Fig. 2A and B).
The existence of subgroups can pose challenges for joint action
aimed at governing a common natural resource, due to the risk of
us-and-them attitudes among actors (e.g. Borgatti and Foster,
2003). The factors behind the formation of subgroups in networks
are many. There might be geographical boundaries that distinguish
sets of actors from others (e.g. villages, see, e.g. Ramirez-Sanchez,
2007), division of labor and/or specialization may lead to the
formation of subgroups (Frank, 1995; Crona and Bodin, 2006), or it
may simply be the result of the inherent limitation of actors to
uphold too many concurrent relations (see, e.g. Gladwell, 2002).
Thus, subgroups tend to inevitably develop, and this has
consequences for governance outcomes.
In network terms a subgroup can be dened as having
signicantly more ties between its group members than between
members and non-members, and various methods exists to
identify them (e.g. LS sets, Lambda sets, k-cores, see Borgatti
et al., 1990 for an overview). A group can also be identied
algorithmically by iteratively removing ties which are situated in
between many pair of actors, thus essentially unfolding the
underlying subgroup structure by fragmentation (community
structure, see Girvan and Newman, 2002). Regardless of how
groups are dened and identied, the underlying assumption is
that groups distinguish themselves from the rest of the network
through their internal tie distribution, i.e. they make up more or
less distinguishably islands in the relational landscape. Ties within
such cohesive subgroups are also referred to as bonding ties (cf.
Newman and Dale, 2007).
Relating this to the previous discussion of the (mostly)
positive effect of network density on collaborative processes we
can conclude that if a social network is less cohesive, i.e. there
are several clearly distinguishable subgroups present, the
density of relational ties between groups can be regarded as
low, and this could have negative effects on the capacity for
collaborative processes among subgroups (Granovetter, 1973).
However, if actors connecting subgroups have the willingness,
capacity and motivations to coordinate subgroup activities
towards a common goal, this limitation could be overcome (as
will be discussed in the next section). Formation and maintenance of subgroups also have implications for a process
deemed important for resource governance, namely generation
and transfer of tacit knowledge, including, e.g. understanding of
complex ecological linkages.
To develop tacit knowledge of complex systems, such as
ecosystems, a high degree of information exchange among more or
less specialized actors (such as, e.g. lagoon gillnet shermen) is
benecial (e.g. Crona and Bodin, 2006; Hamel, 1991). Research
suggests continuous and persistent interaction is needed for tacit
knowledge transfer. This has been shown in studies of organizational performance (Reagans and McEvily, 2003; Hamel, 1991;
Hansen, 1999), and through comparison of local ecological
knowledge held by different groups of small-scale shermen
(Crona and Bodin, 2006; Crona, 2006). The cognitive capacity of
any one individual is limited, and a constant inux of less relevant
information from numerous other actors may in fact hinder the
development of specialized knowledge. Thus, the presence of
multiple subgroups may enhance both (1) the development of
knowledge itself by providing opportunities for high degrees of
interaction among similar others, and (2) contribute to the
development of a diversity of knowledge by enabling different

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

369

knowledge to develop in different subgroups. Both of these factors

are thought to affect outcomes of natural resource governance in a
desirable way (e.g. Walters, 1986; Moller et al., 2004). However,
developing isolated sets of specialized knowledge is of limited use
in governing complex ecosystems since systemic and boundaryspanning understanding and actions are often needed (Berkes and
Folke, 1998). Thus, it is apparent that different subgroups should
interact, and that social networks conducive for governance
processes therefore need to balance between an overall structural
cohesion on the one hand, and allowing for the presence of
subgroups on the other. In the next section we examine the
structural characteristics of social networks that facilitate or
impede such subgroup interactions.
5. Connecting beyond your subgroup
As seen above, cohesive subgroups can be said to consist of
bonding ties. Bridging ties, on the other hand, refers to ties
connecting different subgroups (see Fig. 2D). The primary
argument for this distinction has been that bonding ties promote
trust, reciprocity and thus cohesion within communities, which is
benecial for consensus building and conict resolution, two
important prerequisites for natural resource governance (Ostrom,
1990). Bonding ties are also often required for tacit knowledge
transfer as discussed above. Bridging ties, on the other hand,
provide access to external resources of various kinds, and are often
needed to help actors initiate or support collective action (e.g.
Granovetter, 1973; Newman and Dale, 2007; Lin, 2002), both of
which are vital for resource governance.
The value of bridging ties has been shown in several disciplines.
In organizational science, Ruef (2002) explicitly showed that
increased connectivity between different subunits in corporations
enhanced productivity and innovative capacity. Krishna (2002)
used a large dataset of rural villages in India to conclude that a key
factor in collective action and economic development was the
ability of a village to link up with various external actors and
agencies. Warriner and Moul (1992) also showed that farmers who
were tied to dense and closed subgroup (i.e. had less bridging ties
in favor of more bonding ties), were less inclined to adopt the new
conservation tillage practices.
Similar insights have begun to emerge in research of resource
governance. In a case study in a semi-urban setting, it was
suggested that one of the most inuential factors behind the
successful establishment of the adaptive co-management process
was the existence of trustful bridging relational ties among various
different actors and stakeholders in the region (Hahn et al., 2006).
This is supported by other studies where the positive impacts of
such boundary-spanning network characteristics (i.e. networks
rich in bridging ties) on resource governance outcomes have also
been assessed (e.g. Pretty and Ward, 2001; Newman and Dale,
2007; Westley and Vredenburg, 1997; Ernstson et al., 2009).
In addition to these studies, where the bridging characteristics
of the networks were more qualitatively assessed, quantitative
investigations of specic structural characteristics of boundary
spanning social networks in natural resource governance have
been conducted. A case study of rural shing villages in Mexico
(Ramirez-Sanchez, 2007; Ramirez-Sanchez and Pinkerton, 2009)
showed that, in addition to bonding ties, bridging ties were also
used for critical information exchange of sh stock abundance and
location. Furthermore, Sandstrom (2008) demonstrated that the
presence of bridging ties enhanced the different natural resource
managers collective ability to mobilize various different kinds of
know-how for lake sheries management. Another study showed
that a group of shermen, particularly well connected to a range of
other subgroups of shermen dened by specialized gear use,
seems to have been able to utilize those bridging ties to develop a

Fig. 2. Schematic presentation of some archetypical network topologies. (A)

Represents a network without any clearly distinguishable subgroups (high
cohesiveness), whereas (B) presents a network that is divided in two isolated
subgroups (low overall cohesiveness, also described as a high level of modularity).
(C) Represents a highly centralized network (the node in the middle has much
higher degree centrality than the rest of the nodes), and (D) presents a network with
two distinguishable groups (dotted lines), which are interconnected via two
bridging ties.

better systemic understanding of ecological processes in the region

(Crona and Bodin, 2006).
The positive effect of bridging ties in natural resource
governance extends beyond the exchange of information and
knowledge. They can foster trust among previously unconnected
groups which facilitates collective actions among different types of
actors, such as farmers and governmental ofcials. This is crucial in
natural resources governance which typically affects many
different sectors of society. Schneider et al. (2003) explicitly
assessed, using a series of case studies in the USA, how some
structural characteristics of social networks among various
government authorities and other stakeholders affected estuary
governance processes. They found that if an increasing number of
different types of actors were tied together in boundary spanning
networks, there were positive effects on their beliefs in collaborative action, they developed more trust in the governing processes,
they were more condent in being able to solve conicts, and
hence conditions for efcient coordination among governmental
agencies and others seemed to be signicantly improved.
Implicitly stated, bridging ties thus help putting heterogeneous
actors in contact with each other. The positive effect of such
bridging ties in enabling collective action is also tentatively
supported by the previously mentioned Mexican case study
(Ramirez-Sanchez, 2007). Another example can be found in the
study by King (2000), who showed how the establishment of
bridging social ties to various external and formally more powerful
actors enabled a local group of shermen to inuence decisionmaking processes in their favor. This particular kind of bridging
ties, which vertically connect different hierarchical levels of
authority, are often discussed in terms of linking social capital,
and are suggested as being particularly relevant in order to
leverage resources, ideas and information beyond the level of the
community (e.g. Woolcock, 2001). Interestingly, King (2000)
showed that real progress in the decision-making processes
occurred only when local shermen were able to involve external
authorities beyond those directly involved in natural resource
governance. Thus, it appears that bridging ties in some cases need
to vertically link several different levels of authorities in order to be
really effective.

370

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

Organizational and sociological studies show that the inherent

diversity of ideas and perspectives that emerge with the presence
of bridging ties enhance the capacity for innovations and for
nding solutions to complex problems, and thus adaptive capacity
(e.g. Page, 2007; Davidson-Hunt, 2006). In his study of Mexican
shing villages, Ramirez-Sanchez (2007) described how bridging
ties (i.e. between villages) provided access to shing grounds in
distant shing villages. These ties, which from a hierarchical
perspective of authority can be seen as horizontal, provided access
to a resource which could buffer uctuations in the local shery,
but they also had a benecial effect on social integration between
interacting communities. Thus the net effect was a potentially
increased adaptive capacity stemming from facilitation of consensus building and reduced costs of conict resolution between
villages, in combination with shared resources to deal with
resource uctuations spanning spatial and temporal scales.
6. Network position and inuence
So far the discussion has focused on structural characteristics at
the level of whole networks. However, it is often equally relevant
to assess structural characteristics at the level of individual actors
(i.e. the nodes of the network) to understand how actors can use
their structural position to inuence the natural resource
governance process. By occupying certain central positions in a
social network (Fig. 2C), actors are able to exert inuences over
others in the network, and are better situated to access valuable
information which can put them at an advantage (Burt, 1992,
2004; Degenne and Forse, 1999). There are various ways to dene
and measure centrality in social networks. Here we focus on two
important and distinct types of centrality: degree and betweenness. The number of ties an actor possesses (degree centrality,
Fig. 2C) has been shown to have a positive effect on that actors
inuence (Degenne and Forse, 1999). There are, however, possible
drawbacks of having too many ties. An actor may feel obligated to
please all, or most, of its numerous network neighbors thus
constraining his/her possibilities for action (cf. Frank and
Yasumoto, 1998).
Centrality can also be seen as the degree to which an individual
actor connects other actors who would otherwise not be linked
(Burt, 1992). In SNA the degree to which an actor indirectly
connects other actors is often quantied using the metric
betweenness centrality (Freeman, 1979). An actor who sits
between many other actors in the network is said to have a high
betweenness centrality. This implies that the actor could act as a
bridge between these others (ties that span between otherwise
disconnected sets of actors are by denition bridging ties). It
follows from this that high betweenness centrality grants the actor
the ability to inuence the ow of resources between others, and it
also provides him/her with a diversity of resources provided by the
bridging ties (Burt, 2004; Granovetter, 1973).
The impact of structural position has been investigated in the
literature on natural resource governance. The semi-urban case
study mentioned earlier (Hahn et al., 2006) showed, qualitatively,
that the critical bridging ties among actors in the region were to a
large extent created and maintained by a very small set of key
individuals. This shows that the existence of bridging (or bonding)
ties per se may not be enough to improve governance processes, it
is equally important that the central actors use these in ways that
benet other actors in the network (cf. Krishna, 2002; Bebbington
and Perreault, 1999). Who occupies these positions, and how they
utilize their favorable position will therefore have an impact on
governance outcomes. If individuals occupying inuential positions in the social networks are unaware of the need for, or
unwilling to engage in, collective action they may end up,
deliberately or not, blocking initiatives by others. This has been

put forth as a possible explanation for the lack of common

initiatives among shermen to effectively deal with sh stock
degradations (Crona and Bodin, 2006; Bodin and Crona, 2008), and
it also points to the power asymmetries among actors resulting
simply from occupying different network positions. On the other
hand, an example where a central position is used to improve
governance for the whole community is shown by Isaac et al.
(2007), where centrally positioned farmers engaged in information
acquisition and development of ecological knowledge, while also
passing this information on to other farmers in their personal
networks.
Possessing a favorable position in a social network may not,
however, by itself necessarily lead to higher inuence. Furthermore, an individual can very well be inuential without necessarily
occupying a favorable position (for example, if the individual
possesses formal authority). Interestingly, King (2000) showed
that the combined effect of network position (degree centrality)
and formal level of authority coincided with perceived level of
actual inuence in the decision-making processes, and that
individuals that did not have both were ranked lower in terms
of inuence. In their study of a rural shing village Bodin and Crona
(2008) also illustrated this interplay between structural positions
and formal authority. They identied two nodes of power. One was
represented by the government ofcial, who was formally
recognized by the state but who occupied an only marginally
inuential position in the village networks. The other was
represented by the village chairman, who had no ofcial power
vis-a`-vis the state but was rmly embedded and very central in the
village networks and therefore highly inuential in all village
matters. Thus, the village as a whole was highly dependent on how
and if these two individuals, whos inuence was based on very
different grounds, chose to collaborate on village matters. Further
investigation into the incentives, interests and objectives of such
key individuals should therefore be of interest in network research
for natural resource governance.
A rst step in approaching these questions can be to identify
common characteristics, such as age, ethnicity, level of education,
occupational/hierarchical position, and wealth, among actors
holding inuential network positions. Empirical work has shown
that by doing this, i.e. analyzing which attributes are shared by
central actors, better understanding of factors affecting resource
governance issues in small-scale sheries and agroforestry, such as
knowledge generation and transfer, can be achieved (Isaac et al.,
2007; Bodin and Crona, 2008). Furthermore, central actors may not
well reect the statistical distribution of member characteristics in
the network at large. In fact, they may display quite a skewed, but
internally homogenous, representation with effects on natural
resource governance and agency at the village level as shown by
Bodin and Crona (2008). This shows, once more, the importance of
including analysis of actor representation and effects on agency for
change in studies of social networks and resource governance. A
more elaborative approach to the study of central actors is
therefore to look at strategies or methods employed by them.
Which strategies appear more or less efcient given an actors
specic position? For example, are there any commonalities
between successful leadership strategies, used by key actor
upholding positions characterized by high centrality?
6.1. Network centralization
Centrality can also be measured at the level of the whole
network. The degree of network centralization assesses the
degree to which centralities of constituent actors differ among
each other, i.e. the variability in centrality amongst network
members (see, e.g. Wasserman and Faust, 1994). Thus, networks
including actors with both very high and very low centrality will

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

have a high overall network centrality; i.e. they are highly

centralized (Fig. 2C). Early hall-mark studies of small group
performance have shown that ability to solve simple problems is
positively correlated with high network centralization, while
solving more complex problems requires more diverse structures
(Leavitt, 1951). It has been tentatively shown that high degrees of
network centralization appears positively correlated with collective action in resource governance, mainly through the positive
effect on central actors abilities to prioritize and coordinate
activities (Sandstrom, 2008).
There are, however, several issues of concern with highly
centralized networks. The uneven distribution of ties in itself leads
to asymmetric relations of inuence and power (Ernstson et al.,
2009; Diani, 2003b) which means that issues of legitimacy and
accurate representation of peripheral actors need to be brought
into the analysis. In addition, highly centralized networks may not
be appropriate for governing socialecological systems over time
since they are less suited for solving complex tasks. They are also
more vulnerable to the removal or dysfunctionality of the few
central actors (e.g. reference Frank et al., 2007). As suggested by
Bodin et al. (2006), the degree of network centralization most
benecial for natural resource governance may differ depending
on the phase of the governance process. For example, mobilizing
and coordinating actors at the start of a process may require higher
degrees of centralization, while engaging various actors to resolve
management of complex ecosystem processes may be favored by
less centralized networks.
7. Coreperiphery networks
A special kind of centralized network, so-called coreperiphery
network, results when highly central (core) actors are very densely
tied to each other, while actors in the periphery are connected only
to the core actors, and not directly to other actors in the periphery
(e.g. Borgatti and Everett, 1999) (Fig. 3). This structural characteristic has, as we discuss here, implications for e.g. information
diffusion, but also for access to diverse knowledge and for
mobilization of support at critical moments in the resource
governance process. It has been reported that adoption of new
innovations generally tends to trickle-down from highly interconnected core actors to more loosely connected peripheral actors
(see Abrahamson and Rosenkopf, 1997 and references therein).
Similar effects of coreperiphery structures on the exchange of

Fig. 3. A schematic representation of the coreperiphery network of voluntary

organizations in Stockholm (adapted from Ernstson et al., 2009). The core (on the
top, encircled with a dotted line) consists of a small set of highly interconnected
organizations. The periphery (bottom, encircled with a dotted line) consists of a
larger set of organizations typically not linked to each other, but to the core. The
core group of organizations posses much more ties to political actors (as shown by
the left column) whereas the larger group of peripheral organizations spends more
time in the park (right column).

371

information and knowledge have been specically reported in

resource governance contexts. For example, Isaac et al. (2007)
found that the advise networks among farmers in four agroforestry-dominated villages in Ghana exhibited coreperiphery
structures. Furthermore, the core actors were found to be
signicantly more engaged in acquisition of new information
and knowledge than periphery actors. These core actors acquired
information from external sources beyond the local community
(e.g. from governmental institutions) but also from other, more
peripheral, farmers in their own village. The core actors where thus
acting both as (i) bridges; bringing in new information and
knowledge to the village, and (ii) communication hubs; disseminating this new information within, but also beyond, the core.
There are potentially positive effects of such information
dissemination with regard to the adoption of new, more
sustainable farming techniques and agricultural output at the
village level, and the communication channels, once established
between the core and peripheral actors, could be benecial for
other forms of communication of value to resource governance in
the future. However, these are hypotheses that have not been
explicitly tested.
As mentioned, coreperiphery networks may affect not only
information and knowledge but also acquisition and dissemination. There is also evidence of effects on collective action in general.
This has been explicitly demonstrated in a comprehensive study of
governance processes involving green area management and
planning in an urban landscape (Ernstson et al., 2009). The study
showed how a number of diverse and mostly volunteer based
organizations were brought together in an organizational network
by their common interest to protect an urban green area from
further exploitations. The coreperiphery structure of the resulting
network coincided with a clearly marked division of labor (Fig. 3).
Core organizations had many ties to political, government and
administrative agencies, and were using these ties (along with
other strategies, such as media exposure) to inuence the
governance processes by effectively acting as watch-dogs for
urban exploitation plans. Through their central position they were
able to frame the discourse and to a large degree set the decisionmaking agenda, thus they effectively channeled and exerted the
inuence of the complete set of 62 volunteer organizations with
their approximately 10,000 individual members. The peripheral
organizations, on the other hand, spent more time actively using
the green areas, and had more introvert and user-oriented
activities. They acted as sensors of ecological change and/or
small-scale exploitations, and they habitually reported what they
saw directly to the core organizations who could then bring the
issue into the governing processes. This also shows how the
increased power often associated with central actors can actually
be utilized by peripheral actors in certain situations. There are,
however, issues of legitimacy that need to be considered when a
set of centrally positioned actors dene the agenda on behalf of the
majority of network members. For instance, and as suggested by
Ernstson et al. (2009), the same coreperiphery structure effective
for protecting the urban green area may have hampered the
emergence of co-management since user groups in the periphery,
with local ecological knowledge, have been marginalized on
decision-making arenas.
Finally, in comparison with networks with multiple centers or
subgroups, coreperiphery structures seem less likely to suffer
from extensively us-and-them attitudes among actors. Although
actors in both the core and periphery might sense a social
distinction between the two groups, there is only one centre-ofgravity. This reduces the risk for e.g. conicts which could hamper
collaborative processes (Johnson et al., 2003). However, as
mentioned previously, the lack of distinctive subgroups risks
leading to homogenization and lack of diversity of, e.g. knowledge

372

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

It is clear form this review that structure does make a

difference, although the literature on how structural social
networks characteristics affect natural resource governance is
still limited. Therefore, network studies of successful cases can
reveal key network characteristics that benet governance
processes. Below we briey relate how these insights can inform
future transdisciplinary research on the effects of social networks
on natural resource governance as well as more practical
applications of resource governance.

accomplish this include providing an arena for interaction,

encouraging broader participation, and funding for coordinators/
facilitators.
From another angle, agencies can use SNA to make sure that the
most relevant representatives of different, sometimes even
marginalized, subgroups are being invited and engaged in
participatory processes (Prell et al., 2008). A study of shermen
in southern US actually showed a poor correlation between the
shermen who were judged as inuential by external governmental agencies, and those shermen who appeared inuential
based on social network characteristics such as centrality (Maiolo
and Johnson, 1989; Maiolo et al., 1992). Social network analysis
can thus be used in guiding governing agencies communication
and engagement efforts to maximize efciency, and/or to target
specic subgroups (see also Mertens et al., 2005).

8.1. Balancing structures

8.3. Social network dynamics and cross-scale interactions

Reecting on the discussions above, it is apparent that many of

the different structural characteristics presented here do not
present a monotonically increasing positive effect on processes of
importance in the governance of natural resources. For example, if
only few ties exist among actors, joint action is hard to achieve, but
too many ties can foster actor homogenization and reduce the
capacity for effective collective action to deal with changing
conditions. Thus, assessing the most favorable level where most of
the positive effects are obtained while undesired effects are
minimized presents a key research challenge. Furthermore, there is
a need to balance several different and often opposing structural
characteristics of social networks in order for them to be effective
and efcient in the governance of natural resources (cf. Prell et al.,
in press; Bodin et al., 2006; Janssen et al., 2006; Frank et al., 2007).
The study by King (2000) serves as an interesting example of the
need for both bonding and bridging ties. Local shermen engaged
with several external actors in order to amass support for their
cause (bridging). Successful use of such external ties typically
requires well-coordinated collective action and agreement among
the shermen themselves. This in turn requires, at least
hypothetically, internally well-connected actors (bonding ties).
However, how much of each type of tie is needed, and what an
optimal ratio of bonding and bridging ties might be needed, also
needs further investigation (cf. Bodin and Crona, 2008). Additionally, as stated above, the current phase of the governance process
(e.g. initiation, reorganization, consolidation), will dictate which
structural characteristics are most likely to be benecial.

Social networks evolve over time. For example, actors who are
ultimately successful in furthering their goals often engage in
actively linking up with others in order to pursue their interests (cf.
King, 2000). One can therefore expect the structural characteristics
to change over time. Maiolo and Johnson (1989) showed that
structural characteristics of shermens networks differed
between cases where the shery was more or less developed
(different societal sectors were more or less represented in the
different networks). The study suggests the process of sheries
development may have co-evolved with changes in the network
structure. It has been explicitly suggested that a successful
management strategy in governing natural resources is one where
actors, during periods of stability, develop new relational ties with
various other actors and stakeholders which can be drawn upon in
times of change (Olsson et al., 2006), see also (Hirschmann, 1984;
Gunderson, 1999). This resonates with the proposition that
informal networks are especially useful in times of changes (Frank
et al., 2007; Ramirez-Sanchez, 2007; Bebbington and Perreault,
1999). In this context, we suggest that, in combination with
empirical studies, theoretical models and simulations of various
behavioral characteristics of individuals can provide important
insights on how different networks structures can emerge, and
how the emerging structures might co-evolve with behavioral
changes of the individuals (cf. Hanaki et al., 2007; Skyrms and
Pemantle, 2000). We also think it is important to point out that not
only the structures of a network can evolve, the content of what is
transferred through the ties can also change over time. A relational
tie that, initially, is used only for the exchange of some specic kind
of information, e.g. sustainable farming practices, can evolve into
deeper social relationships which in turn can facilitate the
development of common norms and values (cf. the discussion
on homophily and social ties in McPherson et al., 2001).
Furthermore, even in cases when a tie that is used strictly for a
specic kind of information exchange does not develop into
something deeper, the conveyed information can trigger some
change of behavior. Such a behavioral change could, eventually,
lead to changes in norms and values which, in turn, also shape
governance outcomes.
Viewed from a different perspective, networks which consist of
a set of distinguishable subgroups can be seen as networks of
subgroups. The networks thus exhibit different hierarchical levels
of scales (in this case, individual nodes at the lowest scale, and
interacting subgroups at a higher scale). Different processes can be
dominating at different scales. For example, how individual actors
interact within dense subgroups can be quite different from how
subgroups interact. This can be seen as different scales providing
different functions in the system. However, the different levels of
scale are not isolated from each other; they continuously feed back

and experiences, which, in turn, can reduce the collective capacity

of the actors to adapt to changing conditions (Bodin and Norberg,
2005).
8. Summary and outlook

8.2. Network creation and stakeholder participation

Given that social networks and their structural characteristics
do make a difference in governing natural resources; can we help
build them, and can we simultaneously make sure they develop
favorable structural characteristics? Technically, networks can,
and are, being created and designed in various forms. Theoretical
examples show that changing the conguration of ties between
otherwise identical actors could signicantly improved the
aggregated management outcomes (e.g. Bodin and Norberg,
2005). However, for social network to provide more than just
information transfer, and for them to be sustained over time, the
relational ties must be voluntary. Human agency will inevitably
make actors resist imposed design and they will tend to choose
their contacts based on personal preferences, which are often
related to how similar the contacts are to the actors themselves (cf.
homophily, McPherson et al., 2001). Encouragingly though,
Schneider et al. (2003) have shown that facilitating the development of effective boundary-spanning networks among different
kinds of actors is feasible. Some of the key issues needed to

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

to each other. The need for better understanding of such crossscale interactions in natural resource governance is put forward by
many scholars (e.g. Berkes, 2008; Cash et al., 2006; Ostrom, 2005).
Such cross-scale interactions also affect how social networks
inuence governance processes. As stated by Frank et al. (2007),
local resource extractors are increasingly linked to global networks
of trade (large scale), but the structure of the local social networks
(small scale) largely determines who gets to participate and under
what conditions. A social network perspective holds great
potential in enabling analyses of various cross-scale interactions.
It could therefore be of great value in researching natural resource
governance processes ranging from the local to the global thus
enabling understanding of various factors driving global environmental change.
Acknowledgement
The authors thank Tim Daw, Henrik Ernstson, Carl Folke and
Maria Tengo for helpful comments and support. The research was
funded by the Swedish Research Council Formas and the
Foundation for Strategic Environmental Research, MISTRA.
References
Abrahamson, E., Rosenkopf, L., 1997. Social network effects on the extent of
innovation diffusion: a computer simulation. Organization Science 8, 289309.
Armitage, D.R., Plummer, R., Berkes, F., Arthur, R.I., Charles, A.T., Davidson-Hunt,
I.J., Diduck, A.P., Doubleday, N.C., Johnson, D.S., Marschke, M., McConney, P.,
Pinkerton, E.W., Wollenberg, E.K., 2009. Adaptive co-management for
socialecological complexity. Frontiers in Ecology and the Environment 6,
95102.
Axelrod, R., 1997. The Complexity of Cooperation. Princeton University Press,
Princeton.
Bebbington, A., Perreault, T., 1999. Social capital, development, and access to
resources in highland Equador. Economic Geography 75, 395418.
Berkes, F., 2008. Commons in a multi-level world. International Journal of the
Commons 2, 16.
Berkes, F., Folke, C. (Eds.), 1998. Linking Social and Ecological Systems. Cambridge
University Press, Cambridge.
Berkes, F., Folke, C., Colding, J. (Eds.), 2003. Navigating SocialEcological Systems:
Building Resilience for Complexity and Change.. Cambridge University Press.
., Crona, B., 2008. Community-based management of natural resources
Bodin, O
exploring the role of social capital and leadership in a rural shing community.
World Development 36, 27632779.
., Crona, B., Ernstson, H., 2006. Social networks in natural resource manageBodin, O
mentWhats there to learn from a structural perspective? Ecology & Society
11, r2.
., Norberg, J., 2005. Information network topologies for enhanced local
Bodin, O
adaptive management. Environmental Management 35, 175193.
Borgatti, S.P., Everett, M.G., 1999. Models of core/periphery structures. Social Networks 21, 375395.
Borgatti, S.P., Everett, M.G., Shirey, P.R., 1990. LS sets, lambda sets and other
cohesive subsets. Social Networks 12, 337357.
Borgatti, S.P., Foster, P.C., 2003. The network paradigm in organizational research: a
review and typology. Journal of Management 29, 9911013.
Burt, R., 1992. Structural Holes: The Social Structure of Competiton. Harvard
University Press, Cambridge, MA.
Burt, R.S., 2004. Structural holes and good ideas. American Journal of Sociology 110,
349399.
Carlsson, L., Berkes, F., 2005. Co-management: concepts and methodological implications. Journal of Environmental Management 75, 6576.
Carlsson, L., Sandstrom, A., 2008. Network governance of the commons. International Journal of the Commons 2, 3354.
Cash, D.W., Adger, W.N., Berkes, F., Garden, P., Lebel, L., Olsson, P., Pritchard, L.,
Young, O., 2006. Scale and cross-scale dynamics: governance and information in
a multilevel world. Ecology & Society 11, 8.
Conley, T., Udry, C., 2001. Social learning through networks: the adoption of new
agricultural technologies in Ghana. American Journal of Agricultural Economics
83, 668673.
Crona, B.I., 2006. Supporting and enhancing development of heterogeneous ecological knowledge among resource users in a Kenyan Seascape. Ecology &
Society 11, 32.
., 2006. WHAT you know is WHO you know? Communication
Crona, B.I., Bodin, O
patterns among resource extractors as a prerequisite for co-management.
Ecology & Society 11, 7.
Davidson-Hunt, I.J., 2006. Adaptive learning networks: developing resource management knowledge through social learning forums. Human Ecology 34, 593
614.

373

Degenne, A., Forse, M., 1999. Introducing Social Networks. Sage Publications,
London.
Diani, M., 2003a. In: Diani, M., McAdam, D. (Eds.), Social Movements and Networks.
Relational Approaches to Collective Action. Oxford University Press, Oxford, pp.
119.
Diani, M., 2003b. In: Diani, M., McAdam, D. (Eds.), Social Movements and Networks.
Relational Approaches to Collective Action. Oxford University Press, Oxford, pp.
105122.
Dietz, T., Ostrom, E., Stern, P.C., 2003. The struggle to govern the commons. Science
302, 19071912.
Ernstson, H., Sorlin, S., Elmqvist, T., 2009. Social Movements and Ecosystem Servicesthe Role of Social Network Structure in Protecting and Managing Urban
Green Areas in Stockholm. Ecology & Society 13, 39.
Folke, C., Hahn, T., Olsson, P., Norberg, J., 2005. Adaptive governance of social
ecological systems. Annual Review of Environment and Resources 30, 441
473.
Frank, K.A., 1995. Identifying cohesive subgroups. Social Networks 17, 2756.
Frank, K.A., Mueller, K., Ann Krause, Taylor, W., Leonard, N., 2007. In: Taylor,
W.W., Schechter, M.G., Wolfson, L.G. (Eds.), Globalization: Effects on Fisheries Resources. Cambridge University Press, New York, pp. 385423.
Frank, K.A., Yasumoto, J.Y., 1998. Linking action to social structure within a system:
social capital within and between subgroups. American Journal of Sociology
104, 642686.
Freeman, L., 1979. Centrality in social networks. Conceptual clarications. Social
Networks 1, 215239.
Freeman, L.C., 2004. The Development of Social Network AnalysisA Study in the
Sociology of Science. Empirical Press, Vancouver.
Girvan, M., Newman, M.E.J., 2002. Community structure in social and biological
networks. Proceedings of the National Academy of Sciences of the United States
of America 99, 78217826.
Gladwell, M., 2002. The Tipping Point: How Little Things Can Make a Big Difference.
Abacus.
Grafton, R.Q., 2005. Social capital and sheries governance. Ocean and Coastal
Management 48, 753766.
Granovetter, M., 1973. The strength of weak ties. American Journal of Sociology 76,
13601380.
Gunderson, L.H., 1999. Resilience, exibility and adaptive managementantidotes
for spurious certitude? Conservation Ecology 3.
Gunderson, L.H., Holling, C.S., Light, S.S., 1995. Barriers and Bridges to the Renewal
of Ecosystems and Institutions. Columbia University Press, New York.
Hahn, T., Olsson, P., Folke, C., Johansson, K., 2006. Trust-building, knowledge
generation and organizational innovations: the role of a bridging organization
for adaptive comanagement of a wetland landscape around Kristianstad, Sweden. Human Ecology 34, 573592.
Hamel, G., 1991. Competition for competence and inter-partner learning within
international strategic alliances. Strategic Management Journal 12, 83103.
Hanaki, N., Peterhansl, A., Dodds, P.S., Watts, D.J., 2007. Cooperation in evolving
social networks. Management Science 53, 10361050.
Hansen, M.T., 1999. The search-transfer problem: the role of weak ties in sharing
knowledge across organization subunits. Administrative Science Quarterly 44,
82111.
Hardin, G., 1968. The tragedy of the commons. Science 162, 12431248.
Hirschmann, A., 1984. Getting Ahead Collectively: Grassroots Development in Latin
America. Pergamon Press, Oxford.
Holling, C.S., 1978. Adaptive Environmental Assessment and Management. John
Wiley & Sons, New York.
Holling, C.S., Meffe, G.K., 1996. Command and control and the pathology of natural
resource management. Conservation Biology 10, 328337.
Isaac, M.E., Erickson, B.H., Quashie-Sam, S.J., Timmer, V.R., 2007. Transfer of knowledge on agroforestry management practices: the structure of farmer advice
networks. Ecology & Society 12, 32.
., Anderies, J.M., Elmqvist, T., Ernstsson, H., McAllister, R.R.,
Janssen, M.A., Bodin, O
Olsson, P., Ryan, P., 2006. A network perspective on the resilience of social
ecological systems. Ecology & Society 11, 15.
Janssen, M.A., Ostrom, E., 2006. In: Baland, J.M., Bardhan, P., Bowles, S. (Eds.), Inequality, Cooperation and Environmental Sustainability. Princeton University
Press, pp. 6096.
Johnson, J.C., Boster, J.S., Palinkas, L.A., 2003. Social roles and the evolution of
networks in extreme and isolated environments. Journal of Mathematical
Sociology 27, 89121.
King, A., 2000. Managing Without Institutions: The Role of Communication Networks in Governing Resource Access and Control. Department of Biological
Sciences, University of Warwick, Coventry.
Krishna, A., 2002. Active Social CapitalTracing the Roots of Development and
Democracy. Columbia University Press, New York.
Leavitt, H., 1951. Some effects of certain communication patterns on group performance. Journal of Abnormal and Social Psychology 46, 3850.
Levin, S.A., 1998. Ecosystems and the biosphere as complex adaptive systems.
Ecosystems 1, 431436.
Lin, N., 2002. Social Capital: A Theory of Social Structure and Action. Cambridge
University Press, Cambridge.
Little, L.R., McDonald, A.D., 2007. Simulations of agents in social networks harvesting a resource. Ecological Modelling 204, 379386.
Maiolo, J.R., Johnson, J.C., 1989. In: Thomas, J.S., Maril, L., Durrenberger, E.P.
(Eds.), A Conference on Social Science Issues. University of South Alabama
Publication Services, Mobile, Alabama.

374

O. Bodin, B.I. Crona / Global Environmental Change 19 (2009) 366374

Maiolo, J.R., Johnson, J.C., Grifth, D., 1992. Application of social science theory to
sheries management: three examples. Society and Natural Resources 5, 391
407.
McClanahan, T., Castilla, J., White, A.T., Defeo, O., 2008. Healing small-scale sheries
by facilitating complex socio-ecological systems. Reviews in Fish Biology and
Fisheries 19, 3347.
McPherson, M., Smith-Lovin, L., Cook, J.M., 2001. Birds of a feather: homophily in
social networks. Annual Review of Sociology 27, 415444.
Mertens, F., Saint-Charles, J., Mergler, D., Passos, C.J., Lucotte, M., 2005. Network
approach for analyzing and promoting equity in participatory ecohealth
research. EcoHealth 2, 115.
Moller, H., Berkes, F., Lyver, P.O.B., Kislalioglu, M., 2004. Combining science and
traditional ecological knowledge: monitoring populations for co-management.
Ecology & Society 9, 2.
Newman, L., Dale, A., 2005. Network structure, diversity, and proactive resilience
building: a response to Tompkins and Adger. Ecology & Society 10, r2.
Newman, L., Dale, A., 2007. Homophily and agency: creating effective sustainable
development networks. Environment, Development and Sustainability 9, 79
90.
Oh, H., Chung, M.-H., Labianca, G., 2004. Group social capital and group effectiveness: the role of informal socializing ties. Academy of Management Journal 47,
860875.
Olsson, P., Folke, C., Berkes, F., 2004. Adaptive comanagement for building resilience
in social and ecological systems. Environmental Management 34, 7590.
Olsson, P., Folke, C., Hughes, T.P., 2008. Navigating the transition to ecosystembased management of the Great Barrier Reef, Australia. Proceedings of the
National Academy of Sciences 105, 94899494.
Olsson, P., Gunderson, L.H., Carpenter, S.R., Ryan, P., Lebel, L., Folke, C., Holling, C.S.,
2006. Shooting the rapids: navigating transitions to adaptive governance of
socialecological systems. Ecology & Society 11, 18.
Ostrom, E., 1990. Governing the Commons: The Evolution of Institutions for
Collective Action. Cambridge University Press.
Ostrom, E., 2005. Understanding Institutional Diversity. Princeton University Press,
Bloomington.
Page, S.E., 2007. The Difference. Princeton University Press, Princeton.
Prell, C., Hubacek, K., Quinn, C., Reed, M., 2008. Whos in the network? When
stakeholders inuence data analysis. Systemic Practice and Action Research 21,
443458.
Prell, C., Hubacek, K., Reed, M., in press. Stakeholder analysis and social network
analysis in natural resource management. Society and Natural Resources.

Pretty, J., Ward, H., 2001. Social capital and the environment. World Development
29, 209227.
Putnam, R.D., 1993. Making Democracy Work. Civic Traditions in Modern Italy.
Princeton University Press, Princeton, NJ.
Ramirez-Sanchez, S., 2007. A social Relational Approach to the Conservation and
Management of Fisheries: The Rural Communities of the Loreto Bay National
Marine Park, BCS, Mexico. School of Resource and Environmental Management,
Simon Fraser University, British Columbia, Canada.
Ramirez-Sanchez, S., Pinkerton, E., 2009. The impact of resource scarcity on bonding
and bridging social capital: the case of shers information-sharing networks in
Loreto, BCS, Mexico. Ecology & Society 14, 22.
Reagans, R., McEvily, B., 2003. Network structure and knowledge transfer: the
effects of cohesion and range. Administrative Science Quarterly 48, 240267.
Ruef, M., 2002. Strong ties, weak ties and islands: structural and cultural predictors
of organizational innovation. Industrial and Corporate Change 11, 427449.
Sandstrom, A., 2008. Policy Networks: The Relation Between Structure and Performance. Department of Business Administration and Social Sciences, Lulea
University of Technology, Lulea, Sweden.
Schneider, M., Scholz, J., Lubell, M., Mindruta, D., Edwardsen, M., 2003. Building
Consensual Institutions: networks and the National Estuary Program. American
Journal of Political Science 47, 143158.
Scholz, J.T., Wang, C.-L., 2006. Cooptation or transformation? Local policy networks
and federal regulatory enforcement. American Journal of Political Science 50,
8197.
Schusler, T.M., Decker, D.J., 2003. Social learning for collaborative natural resource
management. Society and Natural Resources 15, 309326.
Skyrms, B., Pemantle, R., 2000. A dynamic model of social network formation.
Proceedings of the National Academy of Sciences of the United States of America
97, 93409346.
Walters, C.J., 1986. Adaptive Management of Renewable Resources. McGraw Hill,
New York.
Warriner, G.K., Moul, T.M., 1992. Kinship and personal communication network
inuences on the adoption of agriculture conservation technology. Journal of
Rural Studies 8, 279291.
Wasserman, S., Faust, K., 1994. Social Network AnalysisMethods and Applications.
Cambride University Press, Cambridge.
Westley, F., Vredenburg, H., 1997. Interorganizational collaboration and the preservation of global biodiversity. Organization Science 8, 381403.
Woolcock, M., 2001. The place of social capital in understanding social and economic outcomes. Canadian Journal of Policy Research 2, 117.