The purpose of the sociological research within the Integrated Project FLOODsite is to better understand the impact of floods on communities and the latter’s capability to respond during, and to recover from, such events. The concept “community” comprises two distinct meanings: it refers, firstly, to a locally based group of people (e.g. a village) and, secondly, to social networks of individuals belonging together because of specific interests and objectives as well as of ties based on kinship or positive emotions. Community-based approaches to flood mitigation aim to build the capacity of local people to respond quickly and effectively. Understanding how communities cope in flood events, how they respond, how they behave, etc. is valuable information to share with those yet to be impacted and with time to prepare, as well as with those agencies responding to flood events. Thus, the major objectives of FLOODsite Task 11 are (i) to characterise types of communities with regard to their preparedness, vulnerability and resilience related to flood events; (ii) to understand the driving forces of human behaviour before, during, and after floods, and (iii) to learn lessons from case studies in Germany, Italy and the U.K.
The outcome of these efforts will provide a better understanding of the role of subjective and intersubjective perceptions and situational interpretations, pre- and post-disaster preparedness as well as the capability and capacity of communities to recover from a hazardous event. Since FLOODsite is a project developed and dominated by natural scientists and engineers, it should be pointed out that our approach differs from mainstream flood research: We strongly focus on a bottom-up perspective, i.e. the residents of flood-prone and, in most cases, recently flood-affected areas. Their points of view in many respects differ from experts’ evaluations with regard to the way flood risk management should work on several scales.
The main Deliverable represents a major outcome of FLOODsite Task 11. It summarises the main findings of three Milestone reports and in-depth analyses at the regional level in the river catchments Vereinigte Mulde (Germany), Adige (Italy) and in England and Wales (U.K.):
• Part A: Country Report Germany (case study Mulde)
• Part B: Country Report Italy (case study Adige)
• Part C: Country Report U.K. (case study England and Wales)
This report represents the results of Part C, the UK case study. The structure of the Country Reports is as far as possible similar, although some research questions are focused on in more detail in certain sections, because they arose out of the specific context of the respective case study. All Country Reports have a common introduction setting out the theoretical background of the basic concepts (Chapter 1.2). After a description of the research locations and the methodological approach, main empirical findings are presented. It has to be taken into account that Parts A and B are based on primary empirical investigations within the framework of the FLOODsite project, while Part C mainly builds upon secondary analyses of data stemming from other research projects.
The Country Reports represent the first milestone of our analyses. The next step will focus on cross-national comparisons and lessons to be learned from the different experiences.
1.2 Theoretical approaches and main concepts
In the following chapter, the most important concepts of our analyses will be explained and defined. These are (social) vulnerability, social capital (including social networks) and risk construction. All of them stem from rather distinct strands of the social sciences and are only exceptionally brought together in disaster research, especially in the classical sociological tradition (e.g. Quarantelli and Dynes 1977; Drabek 1986; Quarantelli 1987; Kreps 1989; Dynes and Tierney 1994; Quarantelli 1998; Tierney et al. 2001). However, we will lay down some good reasons for their interrelatedness. Further context-specific concepts will be introduced in the course of the single Country Reports (Parts A, B and C).
1.2.1 Social vulnerability
Vulnerability has been defined as the major topic of FLOODsite Subtheme 1.3. However, this is not the only reason why it deserves some conceptual consideration. More important is that within just a few years, “vulnerability” has become a buzzword applied in distinct contexts in order to describe and explain almost everything. Some years ago, Weichselgartner (2001, 88) presented 24 more or less different definitions of vulnerability. He categorised them into three approaches: vulnerability as exposure to risks or hazards, vulnerability as social response and vulnerability of places (ibid., 87; with reference to Cutter 1996).
“Official” FLOODsite terminology refers to the first conceptualisation. Vulnerability is defined as the “characteristic of a system that describes its potential to be harmed. This can be considered as a combination of susceptibility and value” (Language of Risk 2005, 27). With its focus on potential or actual damage due to a hazardous event, this describes a very common and widespread understanding of vulnerability from the point of view of natural scientists, engineers, disaster managers and economists (for the latter: Messner and Meyer 2006). From a social science perspective, namely, sociology, geography and political science, however, this framing of vulnerability has some severe shortcomings: First of all, it does not explicitly take into account people’s behaviour, their assumptions, their knowledge and non-knowledge or processes of sense-making. Secondly, the definition does not pay attention to the temporal dimensions of a disaster, its emergence out of and rootedness in daily routines, which in their own are related to the political context and conditioned by policy choices (Sarewitz et al. 2003).
In order to avoid (further) conceptual confusion in this multi-faceted debate, in the following we restrict our efforts to a concept of social vulnerability building mainly upon approaches from sociology and geography. This goes back to a central notion of the term—its emergence “as a concept for understanding what it is about the condition of people that enables a hazard to become a disaster” (Tapsell et al. 2005, 3). Also in the reports, our focus will be on the social dimension of vulnerability. However, we are fully aware that the impact of a flood depends not only on social aspects but also on event characteristics (such as flood depth, duration, contamination, speed of onset etc.), context-specific conditions (functioning of warning system and evacuation measures, dike-breaches, daytime, location) as well as certain parameters which might gain importance in the course of a flood (e.g. type of housing, having handicapped or permanently ill persons in the household etc.). Therefore, if necessary we will also pay attention to these “non-social” aspects of vulnerability.
Social vulnerability can be defined, in a first step, as the specific social inequality in the context of a disaster (be it technological or “natural”).1 This conceptualisation is surely in line with the origin of the discourse in empirical studies on disastrous famines (O’Keefe et al. 1976; Susman et al. 1983) and is fostered by today’s prevalent approach in research practice—which entails an operationalisation by means of indicators and indices in order to “measure” vulnerability (examples are given in: Blaikie et al. 1994, 9, 13, 132–4; King and Mac Gregor 2000; Buckle et al. 2000; Tapsell et al. 2002; Cutter et al. 2003, 246–9, 252; for an overview: Tapsell et al. 2005, 11–7). However, so-called “demographic” or “taxonomic” approaches ignore the situativeness of vulnerability (Wisner 2004, 184–8). The underlying hypothesis of such studies is the existence of a strong positive correlation between socio-economic status and vulnerability or, to put it with Blaikie et al. (1994, 9): “as a rule the poor suffer more from hazards than the rich”. It needs to be stressed that most “classical” vulnerability indicators (age, income, formal qualification, gender, race etc.) are basically indicators of social inequality in general and therefore of social vulnerability with respect to hazardous events in the life-course other than only those caused by “nature”2.
Such an approach of strictly “measuring” vulnerability has both strengths and weaknesses (e.g. Adger et al. 2004; Kasperson and Kasperson 2001). Surely a central advantage relates to the implications for policy: It puts the issue of natural hazards and vulnerability on the public agenda or into the “heart of government thinking” (Benson 2004, 159). Additionally, indicators and indices are transferable to other contexts and allow for cross-regional or cross-national comparison. Moreover, they can be fed into complex, even interdisciplinary models in order to explain flood impact. Not surprisingly, the weaknesses are strongly related to the aforementioned points. When applying indicators and indices which were developed in one cultural context into another one, it is not only the question of whether the respective data are available but, much more important, whether seemingly identical variables measure “the same”.3 A good example in this context refers to tenure: While in some cultures renting a flat is considered as a sign of lower social status, in others (e.g. in Switzerland or in Germany) this causal relationship is not as strong as might be predicted—rental housing is widespread also among middle- and partly even upper classes. Hence, home-ownership does not mean the same in different cultural backgrounds. It is therefore necessary to develop a context-sensitive concept and respective indicators of social vulnerability—this is what we mean by the “situativeness” of vulnerability. Otherwise, researchers run the risk of stereotyped approaches (Handmer 2003, 57), in the end of which they rather approve their own prejudices instead of critically assessing the concepts applied and data analysed.
In our point of view, a worthwhile working definition was developed by Blaikie and his colleagues. By vulnerability they mean “the characteristics of a person or group in terms of their capacity to anticipate, cope with, resist, and recover from the impact of a natural hazard” (Blaikie et al. 1994, 9). This definition highlights both the social and temporal dimensions of a disaster. Instead of emphasising characteristics of the natural or technological hazard itself or the exposure (structures, buildings etc.) to the hazard, it focuses on the question of how communities and social groups are able to deal with the impact of a natural hazard. Hence, it is not so much the susceptibility of entire communities or certain groups to a specific hazard that is of interest but the coping capacity, hence active behaviour, in a very general sense (Green 2003).4 Moreover, this definition takes into account the long-term character of a disaster and the significance of human behaviour in the different phases of such an event.
Although this definition also has some shortcomings (as discussed in Part A, Chapter 5.1), we will apply it because of its genuine sociological character. But in order to make clear that we will not be interested in atomised individuals but rather in people who in mutual social relationships create intersubjective sense, trust, knowledge and interpretations, there is a further concept that deserves our attention: social capital.