Theoretical framework

To analyse the recent changes in the governance of oil transhipment in Statia in relation to the marine ecosystem, this chapter presents the concept of marine community (Van Bets, Van Tatenhove, and Mol, 2016), which enables us to study the transformation of a SIDS into an environmental state in the context of the three-layered power concept of Arts and Van Tatenhove (2004).

Whereas in ecological sciences, the concept refers to a group of interacting living organisms sharing a populated marine environment (Bertness et al., 2000), in this thesis, a marine community is based on social science definitions and insights, emphasising social and political dynamics. A marine community is a community of socio-economic and policy actors and institutions organised around a certain maritime activity that influences or will be affected by the (marine) ecosystem in which the activity occurs (Van Bets, Van Tatenhove, and Mol, 2016).

Analytically, in a marine community, two interdependent communities can be distinguished: a user community and a policy community, as shown in Figure 3.2.

Actors, and their institutions, can be part of both user and policy communities, but each community has a distinct logic, rationality and purpose. According to Crow and Allan (1994) and Smith (2001), a user community refers to a network of

63 interdependent actors that executes, and is affected by, the maritime activity and which makes use of the goods and services provided by the marine ecosystem.

Figure 3.2. The relationship between user and policy communities within a marine community

A policy community comprises actors who are part of (in)formal institutions and governance arrangements that regulate the maritime activity to achieve sustainable use and management of marine ecosystems (Atkinson and Coleman, 1992; Sørenson and Torfing, 2005; Walker, 1989). The marine community concept is different from a social-ecological system for the following reasons. First, the marine ecosystem is problematized and taken into consideration only when maritime activities are performed. Although it is acknowledged that the marine ecosystem is affected by the maritime activity, the focus is on how the marine community governs these environmental changes. Second, the marine community does not adopt the systemic perspective of a social-ecological system; rather, it focuses on the agency of users and policy makers and how they interact with each other to address environmental issues.

Marine communities consist of interdependent state, market, civil society and scientific actors (operational at different levels) who interact with each other in user and/or policy communities. Bringing the environmental state back in provokes the existing power relationships between actors in the marine community to influence policy-making and decision-making. To study the power dynamics within marine communities in detail, the three-layered power concept of Arts and Van Tatenhove

64

(2004) will be applied. They define power as “the organisational and discursive capacity of agencies, either in competition with one other or jointly, to achieve outcomes in social practices, a capacity which is however co-determined by the structural power of those social institutions in which these agencies are embedded” (Arts and Van Tatenhove, 2004, p. 347). Inspired by Clegg’s three circuits of power (Clegg, 1989), they distinguish between relational, dispositional and structural power.

Relational power refers to the capability of actors to achieve outcomes through interactions. This form of power posits actors, resources, outcomes and interactions as the constitutive elements of power. However, the capacities of actors to realise or to influence outcomes in interaction are unequally distributed because of unequal access to certain resources.

Dispositional power shapes the “agency’s capacity to act” (Clegg, 1989, p. 84).

Institutional rules and the unequal distribution of resources define and position actors vis-à-vis each other within the user and policy community and between the user and policy community. Rules determine whether it is legitimate for an agent to claim a certain position, while the (uneven) distribution of resources determines the (relative) autonomy and dependency of an agent in a certain position. These positions co-determine what agents may achieve in terms of relational power in a marine community. The unequal distribution of resources, how resources are applied, and the positioning of actors in the user and policy community in relation to the marine activity are all factors that are the result of structural power.

Structural power refers to the way macro-societal and political structures shape the nature and conduct of agents, being both individuals and organisations.

Structural power works through orders of signification, legitimisation and domination, which ‘materialise’ in discourses as well as in political, legal and economic institutions in societies (Giddens, 1984). “Mediated by these discourses and institutions, (collective) agents give meaning to the social world, consider some acts and thoughts legitimate and others not, and are enabled and constrained to mobilise resources to achieve certain outcomes in social relationships” (Arts and Van Tatenhove, 2004, p. 351). By applying this multi-layered power model to understand the governance dynamics of marine communities, power is not restricted to the ability of actors in the user and policy community to mobilise resources or to realise outcomes; it also includes the way actors are positioned vis-à-vis each other and the unequal division of resources as a result of structural power.

65 3.3 Methodology

To gain insights into the relationships, complexities and institutional settings within a marine community, a case study design was chosen. Case studies allow researchers to derive an in-depth understanding of a research object by examining a range of factors and potential causal connections and how they change over time. In addition, this methodology is useful to address actors’ motives, interpretations, constructions of reality and behaviour (Swanborn, 2010; Yin, 2003). Statia was selected as a case study because of its maritime activity, an oil transhipment and oil storage terminal, operational since 1982 and managed by the American company NuStar Terminals N.V. since 2005 and because of the institutional changes when Statia became a special municipality of the Netherlands.

The semi-structured interviews with key informants served for mapping the marine community structure and for identifying how each actor is positioned in this community. All the semi-structured interviews were conducted using the same topic list, covering questions on stakeholders’ contacts and interactions in the community, access to knowledge and information, rule compliance, conflict resolution and their perceptions, distracted from the theoretical framework (see Appendix 1). Key informants represent the main stakeholder groups, such as national and local governmental authorities, NuStar, fisheries, port authorities, NGOs, dive shops, local inhabitants and researchers. In total, 25 interviews were conducted in total, of which 20 were conducted with a single interviewee and four with multiple interviewees. Some key interviewees were interviewed on more than one occasion. The limited population size at Statia enabled the selection of relevant and sufficient interviewees. Twenty-four interviews were conducted face-to-face, 23 in Statia and one in the Netherlands. Most of the interviews were conducted in interviewees’ offices, and some were conducted out in the field; for example when interviewing a dive shop owner on the beach. The interviews lasted between 25 and 90 min. Before the fieldwork started, eight interview appointments were already scheduled with selected stakeholders. Other interviewees resulted from a snowball sampling method influenced by the preselected interviewees.

Almost all the interviews were recorded with a voice recorder; in some cases, the circumstances did not allow for this. The interviews were transcribed verbally as soon as possible after the interview took place, preferably the same day or the next day. The anonymity of the interviewees was guaranteed by means of a coding system used for referencing the interviews in this chapter (see Appendix 2).

Although the interviewees are categorised in governmental, market, civil society and research actors, some of them have or had multiple roles. In those cases, the

66

most dominant category was chosen.

The data were structured through coding with ATLAS.ti. Several rounds of coding occurred, including top-down coding, based on the interview topic list, the distinct actors in the user and policy community and the three different layers of power, and bottom-up coding in the three different phases, i.e., before 2010, 2010-2015 and after April 2010-2015, which the analysis revealed (see Appendix 4).

The interactions among actors in the marine community and the means they use to interact were visualised through mind mapping, which resulted in the community composition shift presented in Figure 3.5. Data triangulation started with double checking by asking different interviewees the same information or by consulting key interviewees on more than one occasion. Afterwards, the information was also cross-referenced with policy documents, literature, (participatory) observations and news items. In the end, triangulation is determined by an iterative process; thus multiple rounds of structuring and cross-referencing occurred (Bogdan and Biklen, 2006; O’Donoghue and Punch, 2003). Fieldwork at