Comparative Research Design - The Politics of Water Institutional Reform in Neopatrimonial States - page 48

they are accountable to, despite the fact that I repeatedly explained my intentions. In part, the
reaction was open hindrance of my research. This was not the case with other local officials
and the farmers, to whom we usually introduced ourselves without reference to the donor
organization. 31
For the case studies, participatory tools as developed in development research and social
anthropology were used. They are most appropriate with regard to the research interest of this
study:
“Ethnographic research methods remain essential for investigating the dynamics of political processes at the local
level, particularly where we are dealing with (...) informal aspects of power relations in which the way people
understand the situations they face and the options open to them must be central to the analysis” (Gledhill 2000:
7f).
As the time constraints of the field research periods (see above Table 2) did not allow for ex
tensive ethnographic fieldwork, elements of approaches developed in development coopera
tion were employed. They allow the application of qualitative methods in a narrow time frame.
Such approaches are the Rapid Rural Appraisal (RRA), the Participatory Rural Appraisal
(PRA), or the Participatory Learning and Action (PLA). They include semi structured and
open interviews, informal conversations, observations, and group discussions. They are typical
ly conducted in an interdisciplinary team and involve the local population. However, RRA and
PRA approaches have been criticized by social anthropologists for falling short of reaching
their own objectives. Typical allegations are that their time frames are too tight; that despite
their claimed openness they are implicitly based on Western models of thinking and partici
pation; that they cannot guarantee the inclusion of marginalized groups; that the socio cultural
setting is not considered enough; and that the people conducting the research are not trained
well enough (Krummacher 2004; Schönhuth et al. 1998). On the one hand, this critique is
warranted in many respects. On the other hand, these approaches allow for the limited use
of ethnographic methods under time constraints, which is a virtue of these approaches. As the
research question was guided by the above mentioned assumptions (chapter 4.5) and was thus
not entirely undefined (which allowed me to focus my research from the beginning on), this
course of action seemed to be justified. As PRA and related approaches are not explicitly de
veloped for doing research on local institutions, but rather on local conditions, I also took the
guidelines for the analysis of local institutions and livelihoods developed by the Rural Devel
opment Division of FAO into account (Messer, Townsley 2003).
The following methodological tools were used in the case studies: Participant observa
tions, semi structured as well as open interviews, informal conversations, group discussions,
and transect walks. The four former methods were already described above. Group discussions
and transect walks will now be described briefly. There are very different conceptions of group
discussions. I refer to the idea that in group discussions underlying assumptions are articulated
that might not be expressed in individual interviews but that apparent during discussion with
others (Bohnsack 1999: 123 128; Dreher, Dreher 1991). Transect walk is a method of inter
viewing a person or a group informally while observing the person/group during a walk along
an identified transect across the research area (mostly a settlement). It is particularly used to
address issues that have a spatial or physical dimension, e.g. poverty in a village or land man
agement. During the transect walk, all of the aspects of interest for the research are observed
in order to get an overall idea of the problem and its context. In addition, it is expected that
31 In general, before every interview, both at the local as well as at the national level, I had to take the particular
situation into account to decied whether it would be more beneficial to introduce myself as a PhD student, a member
of an international research project, or with reference to a national academic institute or a donor organization.