Ethnic Federalism in a Dominant Party State: The Ethiopian Experience 1991-2000 Lovise Aalen r 2002: 2

Sources of “evidence”
The analysis of the Ethiopian federal system relies on three different kinds of
sources: qualitative interviews, documents and academic literature. Most of the
literature and parts of the documents were collected in Norway (University of
Bergen Library and Chr. Michelsen Institute Library) and Sweden (Nordic
Africa Institute in Uppsala). Nearly all the interviews and the majority of the

---

C M I
27
documents were collected in Ethiopia, primarily in the period from May to
September 2000. My first direct exposure to politics in Ethiopia, however,
started right before the major data collection period. During the first three
weeks of my first stay in Ethiopia (2.-21. May 2000) I was an observer from
the Norwegian Institute of Human Rights for the regional and national
elections in Tigray (Aalen 2000a). This gave me background knowledge on the
political situation in the country, introduced me to the skills of data collection
through interviews and gave me many useful contacts both in Tigray and Addis
Ababa. During the major collection period, I spent about one and a half month
in Addis Ababa, one month in Tigray, primarily in the regional capital
Mekelle, and one month in Southern Nations, Nationalities and People’s
Regional State, SNNPRS, basically in the regional capital Awassa. After this
period, more information was gathered during a month’s stay in Addis Ababa
in November 2000 as a participant at the 14
th
International Conference on
Ethiopian Studies (Aalen 2000b) and during a second month’s stay as an
observer for the Norwegian Institute of Human Rights during the local
elections in Addis Ababa in February-March 2001 (Pausewang and Aalen
2001). These new visits gave me opportunities to fill in gaps in the data already
collected, to maintain contacts and to be updated on recent developments.
In the following, each category of sources for the study of the Ethiopian
federation will be described. I have tried to make explicit the way the data was
presented to me as a researcher and how it was collected and interpreted. In
this way, I hope to enhance what Yin (1994) called “construct validity” and
avoid biased conclusions.
Interviews
The interviews include talks with seventy-one people in the federal capital
Addis Ababa, in Tigray and the SNNPRS, each lasting on average about an
hour. The total number of interviews might sound large, but the information
value of each interview varies greatly. Some of the interviewees provided
substantial and essential information, while others were either unwilling or
unable to give useful information. The interviewees were not used as
respondents, giving direct information about themselves, but as informants,
providing information and opinion about pre-determined issues: the Ethiopian
federal system and federal-regional relations. Informants, however, may have
an interest in manipulating the reality and present it as it best serves their
interests. When different informants give contradictory versions of events or
processes, I am forced to interpret what is the most likely to have happened. In
such situations, it is particularly important to be aware of my own impact on
the interpretation. Although I have tried to make all sides heard, it is
nevertheless impossible to be an objective observer. Where such instances
occur, I have tried to make my own viewpoint known.
The interviews could be characterised as “open-ended” (Yin 1994: 84),
where the informants were asked about both opinions and facts.
Although no
fixed sets of questions were presented, the interviews were “focused” and semi-
structured in a sense that they centred on pre-determined issues defined in an

---

C M I
28
interview guide.
9
The actual formulation of the questions was adopted to suit
each particular interview. As I obtained more knowledge of the functioning of
the Ethiopian federation during my stay, I added more questions on specific
issues, like for instance the informal and uninstitutionalised aspects of the
political system.
The interviews are crucial sources for the understanding of the de facto
functioning of the federal system and are crucial in comprehending the
informal political structures and politics that take place outside the institutional
frames, particularly linked to the party system. The interviews will be used
both directly and indirectly in the analysis. I have no intention of presenting the
interview material as a whole, but will let the structure of the analysis
determine where the material will be referred to, directly as quotes, or indirectly
as references. But there are situations where the interview material will “talk
for itself” and will be cited directly and in whole, as for instance when different
political parties give their views on the Ethiopian federalism, the normative
base for the federal system.
I rely on hand-written notes, despite the fact that the use of tape recorder
would give more accurate information. This is time saving, but most important
of all, to make notes does not make the informant as suspicious and
uncomfortable as the use of a tape recorder might do. All names are kept
confidential due to the sensitivity of the matters discussed, except those of
people in public positions who have expressed that they are not afraid of
exposing themselves.