Annales d'Ethiopie

Reflections on the Power Elite of the Wärä Seh Mäsfenate (1786-

1853).
Shiferaw Bekele

Citer ce document / Cite this document :

Bekele Shiferaw. Reflections on the Power Elite of the Wärä Seh Mäsfenate (1786-1853).. In: Annales d'Ethiopie. Volume 15,
année 1990. pp. 157-179;

doi : https://doi.org/10.3406/ethio.1990.951

https://www.persee.fr/doc/ethio_0066-2127_1990_num_15_1_951

Fichier pdf généré le 14/05/2018

 REFLECTIONS ON THE POWER ELITE OF THE
WÂRÀ SEH MÂSFENATE

(1786-1853)

by

SHIFERAW BEKELE

This article deals with a small social collectivity located at the

top of the social and political pyramid of the Zâmànd Masafent.
The stratum was made up of the men and women who ruled the
country. Because they owed their position in the society to the office
they occupied, a brief description of the state system of the era
precedes a discussion of their characteristics and role.

The article will have achieved its purpose if its social canvas
is taken by the student of Ethiopian history as a pointer to future
lines of research.

The Zamanà Masafent has been studied by several scholars

from one angle or another. Some of these studies constitute
important contributions to Ethiopian studies. Nevertheless, these
works are more in the nature of surveys or indepth studies of a
concrete subject matter than extended investigations of the period
as a whole. For example, political history has not been the subject
of an extensive research.

All the surveys however share certain basic interpretations

and assumptions though they differ from one another in emphasis.
Perhaps, the most pivotal interpretation is the one that concludes
with the "disintegration of the state" theory when every princle
"did that which was right in his own eyes" in the words of the Bib e
from which the very name of the period is said to have been taken

— 157 ~-
 Trimingham characterized Zâmand Masafent as an era which
witnessed "the triumph of regionalism"1 or "the disintegration
of the empire"2.
sovereigns" 3. This The
thesis
lords
is extremely
of the various
important
regions
firstly
were
because
"regional
it is
the popular view of the period. Secondly, Abir accepted it in its
entirety and developed it into a major theory in his book. 4 In one
way or another, even the finest historians of the 19th c. (Rubenson,
Crummey and Caulk) have adopted this same theory.

Crummey has captured the essence of this interpretation by

the expression "collapse of the Ethiopian state" for the period.5
Rubenson was no less emphatic when he wrote that the Zâmand
Masafent "was characterized by a complete breakdown of imperial
authority and by increasing regionalism."6 This regionalism led
to constant civil wars which in turn created anarchy and considerable
lawlessness. Thus, there is a consensus among the specialists on this
thesis. This should not, however, lead us to the conclusion that
Crummey and Rubenson did not represent an improvement on
Abir and Abir on Trimingham. They did. But on essentials, the
"disintegration" thesis ran all the way through.

In all this disintegration and chaos and anarchy, the Yaju lords
maintained "predominance" over the other lords. Abir organized
three chapters of his book round what he called "Galla
predominance."7 Crummey and Rubenson fully followed in his footsteps.*
But what exactly this "predominance" or "hegemony" involved
has not, to my knowledge, been precisely defined by any of these
historians. By implication, one can deduce that this hegemony did
not tamper with the regional powers of the lords and that it rather
meant influence than actual authority.

This summary is no doubt a simplification. But it is hoped

that it has represented the central point of the thesis. If this is so,
then one can contend that the theory stands in need of révision
because it does not by any means reflect all the reality.

As I have tried to demonstrate elsewhere at considerable length 9,

Zâmdnâ Masafent was rather an era in which the Wârâ Séh lords
or masafent ruled the country in the name of the Solomonic dynasty.
To be sure, these masafent did not exercise absolute authority
outside of the provinces directly under their rule. These provinces
extended from Bâgémder across the Câçâho Pass to the Amhara
provinces of Mâqét, Wadla, Dâlanta and Dawent. Their authority
over the lords of the rest of the regions-Gojam, Wârâ Himâno

— 158 —
Wâllo, Semén, Tegray upto Hamasén, Wag and the like — was
rather restricted to the power of making them pay regular tributes,
be in attendance in their court for a given period of the year,
participate in their wars and exercise right of appeal in matters of justice.
They also had the right of confirmation in their offices when new
lords came to power in one way or another. At times, they even
went to the extent of chaining some of their recalcitrant lords.
The relationship between the Wârâ Séh rulers and the regional
governors was one between suzerain and vassal or tributary lord.
Two traditional Ethiopian scholars — Aldqa Tayâ Gâbrâ
Maryam and Ato Tâklâ Sadiq Mâkuriya — use the term ?°ft<RC10
(mesfenwa) for the state. Mesfenna means rule by a mdsfen.
Perhaps, the best rendering of this term into English is masfenate.
Labelling the state mdsfenate, following Tayâ and Tâklâ Sadiq, has
a number of points to it. Firstly, it precisely expresses the nature
of the state. Secondly, it suggests that the state of the Zdmdnd
Masafent has some difference from the monarchical state.
It was Ali I who succeeded in establishing this system following
his ascent to power in 1786 10. His seizure of power was a
culmination of a process that was set off by Mikael Sehul in early 1768
when he was appointed ras bitwâdâd. n Mikael right away arrogated
power to himself and proceeded to supress the factions in Gondar
who had been such a source of instability to the throne.12 This
seizure of power by openly controlling the throne did not go
unopposed. But Mikael got Iyo'as killed. This act, in combination
with other events, led to a series of civil wars which raged for two
years. In the end, Mika'el was defeated in 1771 and removed from
office.
Following his removal, the monarchy tried, even if timidly,
to reassert some of its former powers. 13 Though it did not fully
succeed, equilibrium of power was established between the throne
and regional lords for over a decade. The king continued to exercise
power. 14 Even if this was a power much reduced than before, it
was a power nonetheless because the king had under his command
several regiments, maintained court, exercised some amount of
sum-ser and continued to govern directly a number of districts
around Gondar. Although it was more nominal than actual, all the
lords pledged their allegiance to him.15

Thus, the years from 1771 to 1784 constituted a sort of

interregnum or a transition age between two eras — the era of the absolute
monarchy and the era of rule by the Wârâ Séh Mâsfenate. Tâklâ

— 159 —
 Giyorgis, whose first reign was from 1779 to 1784, tried to assert
all over again some, if not all, of the powers of the monarchy. "
Unfortunately for him and for the dynasty as a whole, he ended
up losing everything. A coalition of lords, among whom numbered
Ali I, defeated him in early 1784 at a place called Afârâ Wanat
and replaced him with Iyasu III who was a full puppet. Thus, this
battle is regarded as a turning point by traditional historians because
it marked the dispossession of all the authority of the kingship.
For this reason, they nicknamed Tâklâ Giyorgis fesamé or tâfsdméta
mdngest. This is usually translated as end of state/government.
Perhaps, the best rendering would rather be end of the monarchy.

It was two years later in 1786 that Ali became ras bitwdddd.
He then set about obtaining the recognition of his seizure of the
highest office by the other lords. At the same time, he made it
clear that he was ruling in the name of the king. The title of ras
bitwdddd came to acquire definitely the meaning of re'esa makwanent
or re'esd mdsafent or their equivalent which could be rendered in
English as head of the governers or lords. His old position in the
court of the asé of the most senior officer was now elevated to that
of a suzerain or overlord of all the lords of the Gonderine kingdom.
This suzerainity in effect replaced the monarchy.

The years from 1786 to the beginning of the 19thc. have not
been studied at all by modern historians. At this stage of our
research, the following two points can be made. Firstly, these were
years in which Ali I (1786 - 1788) and his brother and successor,
Aligaz (1788-1803) were fighting hard to keep themselves in power18.
They were constantly challenged by the lords of the various
provinces. Secondly, the major wars were fought either for the office of
the ras bitwdddd or in defiance to the suzerainity of Ali I or Aligaz. 19

The greatest achievement of Gugsa (1803-1825) who was the

nephew of Ali I and who succeeded Aligaz was the consolidation
of the Mâsfenate. He succeeded in reducing to vassalage all the lords
from Gojam and Wâllo to Hamasen. He brought about several
years of peace and stability which are remembered even today. 20
However, he failed to attain legitimacy for his house.

Thus, his death gave rise to a succession strife because a number

of lords wanted to take his place. The strife lasted from 1826 to 1831.
It took such a long period mainly because his son and successor,
Yimam, died only three years later in 1828. Even if Yimam had

— 160 —
managed to subdue all the rebel lords, his premature death gave
them another opportunity to bid for the ras-ship all over again.
His brother and successor, Maryyé, was an able general and a
capable diplomat. He brought to heel one after another the lords
who had challenged him. He died unfortunately in the course of
defeating the last of the principal rebel lords, Subagadis, in 1831.
He was succeded by his brother Dori who, however, ruled for only
three months.

Ali II succeeded him in the same year in 1831 without much

of a problem. Ali, who was the nephew of his three predecessors,
was a minority. The Masfenate was run by a council of regency of
which his mother Mânân was a prominent member. When he
reached of age by the second half of the 30s, he took over the overlord-
ship. His rule lasted until 26 June 1853 which marked his total defeat
and exit from history. In this day, the battle of Aysal was fought
between him and Kassa, the latter Tewodros. By ruling for no less
than 22 years he rivalled hi suncle Gugsa who had reigned for as long.
Though Ali II was not as remarkable a leader as Gugsa, he had
considerable achievements to his credit. He had managed to keep
all the major lords under his suzerainity upto the end. His reign
was definitely punctuated by recurrent rebellions; but it is to his
credit that no major civil war was fought apart of course from the
battle of Dâbrâ Tabor of 1842 and the war of 1852/3. Unfortunately,
he was succeeded by no less a figure than Tewodros in contrast with
which his mediocre personality dimmed a great deal.

Kassa's victory over him made his rule the last of the dynasty,
a house which stayed in power from 1786 to 1853 -for 67 years. This
state had a number of characteristics which set it apart from the
previous and latter Solomonic states.

Much more than the absolute monarchy, the Masfenate had

the attributes of "a loose confederation of regional princedoms." 21
Indeed, the various regional ruling houses and the administrative
structures which they headed exhibited in all but name the
characteristics of vassal states. They retained tremendous local autonomy
in return for a few obligations outlined at the beginning of this
article.

Thus, the Zâmânâ Mâsajent was not the era of independent

princes who constantly fought against each other. In the words
of the Biblical verse from which the term is said to have been taken,
"In those days there was no king in Israel: every man did that which

— 161 — .:
was right in his own eyes."22 Though it witnessed considerable
anarchy and civil wars, the Ethiopian Zàmânâ Mdsafent was not
by any means like the one described in this verse notwithstanding
popular notions. Perhaps a more accurate way to render this term
is by taking it as a counterpart to the term Zâmdnà Mdngest. Zdmdnd
Mdngest is the period of rule of a king and by the same token Zdmdnd
Mdsafent constituted a period of rule of the [Wârâ Séh] mdsafent.

II

A discussion of the state would not be complete without a

discussion of the men & women who ran it. Not much social history
has been written so far. Whenever historians discuss the ruling class
especially those of the Zdmdnd Mdsafent, they tend to focus on
their divisive characteristics. It would probably be of as much value
to see if they had unifying features and centripetal traits. To identify
their self-interests and their driving force would open the key to
a true understanding of their nature.

Perhaps a useful concept to be deployed here is power elite. By

power elite we mean the highest ranking segment of the Mdkwanent
class. These were princes, provincial governors, high court officials
and the high ecclesiastical dignitaries (like the eçdgé, Neburd ed etc).
The concept also includes the women who belonged to the families,
of these men. Neither the term mdkwanent nor mdsafent could be
used for this social collectivity as the former covers the entire ruling
class & the latter excludes several power holders. For instance the
aqabé sa1 at and provincial governors who did not have royal blood
could not be included within the mdsafent category; and yet, they
held some of the highest offices in the state system. This working
definition will hopefully justify its use in this essay.

All the top governors and their principal courtiers of the

masfenate constituted one power elite. They were all interrelated
either by blood or marriage. They shared common aspirations,
value systems, perceptions and self-interests.

Ali I himself was the son of Séru Gwangul of the Wârâ Séh
while through his mother he was related to Ras Faris of Lasta. 23.
When he died in June 1788 he was buried in Lalibela24 because
Lasta was as much his "country" as Yâju. His sister Kâfay was
married to Mârso Bârénto, chief of a Wâllo Oromo clan. 2S.
This would entitle her sons, Alula and Gugsa to claim Yâju, Wâllo

— 162 —
and Lasta as their patrimony. Gugsa's daughters were married to
Maru of Dambeya, to Ddjac Wâldâ Rufaél of Meça (Gojam), to
Wag sum Kenfu and to Dajac Haylâ Maryam of Semen. The latter
had three children from Hirut Gugsa-Mârso, Betul and Yâwbdar.
Yâwbdar was married to Subagadis and her two young brothers
stayed with her in Tegray after their father died 26. Later, they
came to their cousin's (Ali II) court at Dâbrâ Tabor where they
lived most of their lives. Ali II himself was the son of Mânân,
member of the Mamâdoc family of Wârâ Himâno and Wâlo 27.

He was married to Hirut, the daughter of Webé. His sister

was married to Bern Gosu. Dâjazmac Kenfu, his other major
tributary lord, was the nephew of Dajazmac Maru, son-in-law of
Gugsa. That very troublesome vassal, Dâjazmac Faris Aligaz, who
was for many years governor of Yaju and parts of Lasta, was his
cousin being the son of Aligaz, the brother of Ali I. Faris's brother,
Beru Aligaz of the battle of Dâbrâ Tabor fame, was an important
dajazmac who occupied governorships of different AmharaDâjazmac'
districts
like Dâmbeya. The lords of Wârâ Himâno and Wâllo,
Baser & Amâdé, respectively, played central roles in the politics of
the 1820s and 30s. They were the brothers of Mânân.

All the provincial ruling 'houses shared two characteristics :

inter-relationship and continuity which incidentally has always
been the trait of power elites throughout history. The Mamâdoô
family was inter-related by blood & marriage with practically every
province in the Mâsfenate 28. The case of the Wârâ Séh rulers
is already wellknown. There was no ruling house not related to
them 29. Taytu would be a typical case in point in that family.
She was as much a member of the Wârâ Séh as of the Semen house;
her mother was from Gojam and her aunt was married to Subagadis.
Yâju, Lasta, Wâlo, Bâgémder, Meça (Gojam), Semén were all her
"countries". In her later life, there was practically no family of
importance to which she was not related either by blood or marriage.

The Semén ruling house of Gâbré was the same. Nor were
the Mikael Sihul, Subagadis & Wàldâ Selasé families any different.
In short, the rulers constituted a small elite. Everybody knew
everybody else. Ali II knew & was related to, in one way or another, with
all the important families of his realm.

At a lower level, the story was somewhat similar. Tigrean

matchlockmen were employed everywhere. Other professional
soldiers moved round the country in search of better pay 30.

— 163 —
It would be erroneous to characterize these soldiers as "mercenaries"*
Surely, they were not viewed as such by their countrymen nor by
their employers. The priestly class circulated freely. They travelled
throughout the country to attend famous schools and went wherever
the pay was good 31. Therefore, it was not only the cream that
constituted, so to speak, one national elite. Some of the strata below
them as well were nationally intergrated social groups.

The national elite also exhibited remarkable continuity. The

ruling houses of all the provinces survived as a ruling house from
around the middle of the 18th c. right through to the middle of.
the 20thc — an impressive continuity by any standard. The so-
called "founders" of these provincial dynasties were in actual fact
men who had already belonged to Mdkwanent families with a long
historical continuity. Haylu Sr. who is regarded as the founder of
the Gojam ruling dynasty belonged to the rarefied reaches of high
society — his father, Yosédeq, being one of the important lords
of the Gonderine Kingdom and his mother, a princess. Prince
Ras Haylu Bâlâw of the post- 1941 era could be regarded as the last
prince of the house, a house which governed that province on-and-
off for around two centuries.

The Semen dynasty was no different. Ras Gâbré was the founder
of this house. He was a governor in a line of governors going back
many decades 32. He ruled in the last years of the 18th and the
first decade of the 19th c. His son (Dâjazmtâ Haylâ Maryam) and
grandson (Dajtâ Webé) succeeded him. Members of this family
continued to play a central role in the politics of Semen upto the
invasion of the country by Fascist Italy. Ayaléw Beru was the last
member of the house to govern the province.

The Mamâdoc family governed Wârâ Himâno and Wâlo from

the second quarter of the 18th c. right through to the revolution.
The last governor, Crown Prince Asfa Wâsân was a descendant
through his mother in a direct line. Even lesser families like that
of the Wagsumate exhibited a perfectly similar feature 33 . Perhaps,
the sole exception were the Tigrean families of Mikael Sehul, Wâldâ
Selasé Keflâ Iyâsus and Subagadis Wâldu because none of them
succeeded in monoplising power in their region for an extended period
of time. An yet, they did succeed in maintaining themselves to the
last as members of the makwanent class. Like the others, the three
founders belonged to the families of governors. Again like the others,
they were widely intermarried which was epitomized in Kassa-
Yohannes who could claim descent from all the three lords.

— 164 —
 This continuity was however obscured by the fact that
succession was not always from father to elder son. It was not uncommon
for collaterals to succeed. But all told, power remained in the firm
hands of the house.

Marriage which was normally intended for political purposes

was the factor that constantly sustained this internal link within
the power elite. Scholars have repeatedly underlined the political
motives of marriage which, they point out, did not create any long
lasting alliance or loyalty 34. This may be true though we tend
to notice those marriages that had failed rather than those that had
succeeded in their apparent motive. In any case political motivation
ensured that anyone who was politically important was drawn into
the elite by marriage whatever his background. The tradition also
explains why marriages in the ruling class tend to be so farflung.

In the process, an inter-related national elite was constantly

maintained which shared common ways of life, value systems and
self-perceptions. This seems to have been always the case since at
any rate the 14thc upto 1974.

"Qwarânoô"
One can oftake
the era
two ofcases
Mentwab
just to
(1730s-1770s)
illustrate the
and point:
the Sâwan
the
Mdkwanent of the Menelik era. The extended family of Mentwab
was intermarried and inter-related with all the principal provincial
families from Gojam to Tegray. Her own daughters were married
to the lords of Gojam, Bâgémder, Tegray while her son was married
to the family of a Wâlo Oromo chief 35. The case of the era of
Menelik and Haile Selassie was exactly the same with only one
difference — the inter-marriages in the latter era were on a much
bigger scale.

Thus, there was a similarity in terms of the composition of the

power elite. But there was one development peculiar to the Zâmânâ
Mâsafent. In the earlier and latter ages, the national elite was
Orthodox Christian. In other words, religious homogeneity was
maintained. In this era, however, the national elite was made up
of two religious groups: Christian and Moslem.

The ruling houses of Wâlo and Wârâ Himâno professed Islam

while the Wârâ Séh had been Moslem before. Notwithstanding
their devotion to Christianity they continued to bear Moslem names
which showed that they did not give much respect to the sensibilities
of the Christian lords.

— 165 —
 It was not a question of sensibility only. In the Christian state,
the central institutions had all along been staffed by Christians as
a matter of principle., There is a fundamental reason for this. The
monarchical state, as I try to demonstrate elsewhere, was a
theocracy 36. The monarch, the Itégé, the principal courtiers, the
governors and other members of the makwanent class had dual
functions: secular and religious. To be sure, the religious functions
were ordinarily carried out by a substitute. But a Moslem could
not fulfill them even by substitutes. For instance, the governor of
Wârâ Himâno was supposed to be in charge of the overall
administration of the famous Tâdebabâ Maryam and Atronsâ Maryam
churches located in Amara Saynt.

In this era, however, the central institutions that had survived

were filled in by Moslems or former Moslems. Indeed, those Moslem
or ex-Moslem families were inter-married with the important
Christian families. In spite of this however, the Christian lords had all
along remained resentful. They constantly and invariably referred
to the Wârâ Séh as "Galla" or "Moslem". On the other hand, they
did not refer as "Galla" to the Christian families of patent Oromo
origin (e.g. the Gojam ruling house of Zâwdé and Haylu Sr.) 37.
Abir takes the name-calling at its face value and formulates the
theory that the struggles of the period were basically ethnic - i.e.
Oromo vs. Amhara/Tegray where as D. Crummey carefully
considers the sources and dismisses Abir's thesis as rather
superficial 38. He concludes that "the application of the term Galla..,
is less an index of objective descent and degree of non-semitic blood
than of disapprobation, which frequently arose from grounds far
removed from the ethnic" 19.

• True, the term "Galla" had acquired by the 1760s pejorative

connotations as Crummey rightly pointed out 40. However, the
question
"Amara"that
denoted
should
to first
the people.
be asked
Firstly,
is whatboth
the of
terms
them"Galla"
referredand
to
the two ethnic groups. Secondly, they referred to religious affiliation.
"Amara" denoted "Christian" in contrast to the term "Galla"
which meant "pagan" or at times "Moslem" 41. So when an
Oromo referred to himself in the 19thc as "Amara" he was not
necessarily trying to hide his Oromo identity because he was ashamed
of it as external observers mistook it. He was rather informing his
interviewer that he was a Christian.
It was very likely in the same sense that the lords and
chroniclers who opposed the Wârâ Séh called them "Galla" while they

— 166 —
did not call the Zâwdé family "Galla". The latter family had already
been fully Christianised and their court was staffed by Christian
courtiers and retainers. In the case of the Wârâ Séh, however, their
principal counsellors were the Wâllo and Wârâ Himâno lords who
remained Moslem and they themselves were frequently ambivalent
towards Islam in spite of the fact that they built churches. At any
rate, they could not be excused for one thing: giving high titles
and court positions to Moslems (e.g. Dajazmacs Baser & Amâdé).

But this should not obscure the fact that the ruling houses of
all the provinces of the Mâsfenate constituted one inter-related and
inter-twined elite whose values & aspirations were predicated on
ruling the country. True, they differed in their aspirations from the
power elite of the previous and latter monarchical states. In the
latter case, the royal court was the epicenter of the social group.
They gravitated towards the royal family. The degree of success
of individuals and families as well as of factions was directly
proportional to the degree of proximity to the royal house. On the other
hand, the power elite of the Zâmând Màsafent did not have any
royal epicenter. Nor were many of them prepared to rotate round
the Wârâ Séh court at Dâbrâ Tabor. Notwithstanding this
difference, however, they politicked and fought to have a place or a
say in the centre.

Scholars have often ignored this trait and rather focused on

the negative characteristics of the power elite — namely, fractious-
ness, a tendency to unruliness and a strong centrifugal predilection.
In fact, the latter aspect is made the central motif of Ethiopian
history at any rate for the last 600 years. As Crummey neatly
summarised this view:

Abyssinian political history has a major motif of

centrifugal tendencies, which are associated a great
deal more with the drives of the nobility towards
maximum local autonomy than with the ethnic or
regional identities to which many previous
commentators have attributed them. These centrifugal forces
triumphed from the 1770s to the 1850s in the period
known as the Zamana Masafent 42.

Another great scholar, M. Perham, also puts the tug-of-war

between the top màkawanent and the monarchy at the very centre
of the country's political history when she wrote that there was

— 167 —
"permanent tension" between these two elements43. This is
such a widely held theory that today it has acquired the status of a
truism in Ethiopian historiography. The Zâmânâ Mâsafent is taken
as its culmination and as its final proof.

And yet the true nature of the elite was far from the picture
that this theory gives us. The top mdkwanent constituted an
intrinsic part of the centre and occupied a number of the central
institutions of the state in addition to or because of their governorship.
A much more accurate way to characterize this class would be
to state that they were at one and the same time a centripetal and
a centrifugal force with, however, the centripetal tendency being the
predominant trait,4* Contrary to the universal assumption, the
Zàmànâ Mâsafent showed clearly that the centripetal tendency
was the over-riding trait of this class rather than the centrifugal
factor.

The campaigns Gugsa conducted in the first and second

decades of the 19thc were to bring about the submission of the
various lords and the consolidation of the Màsfenate. Rases Gâbré,
Mâred, Wâldâ Selassé — all fought to mount the alga occupied by
Gugsa. Most of the major battles of the late Zâmânâ Mâsafent
(1820-1853) were fought over the office of the ras bitwâdâd — i.e.
for seizure of state power. The same thing could be said for the wars
of the 1780s and 1790s. Armies marched and couter-marched in
order to place some one or other on the alga of the ras. Thus,
the major wars of the period were fought to seize state power.

But this does not mean that there were no centrifugal elements.
There were. The most well-known or notorious cases of. the latter
Zâmânâ Mâsafent were Beru Gosu of Gojam and Faris Aligaz of
Yaju and Lasta. These men continually rebelled. There were others
as well even if not so wellknown. Therefore, Ali II spent much of
his reign moving from Gojam to Lasta; from Lasta to Yaju in order
to suppress one rebellion or another. But these rebellions could not
be regarded as the fundamental motif of the era.

The political crises that brought out the true character of the
power elite was succession strife. They would break' into factions
and then form coalitions. The whole struggle between coalitions
revolved round the question of whose candidate sat on the throne
and not at all on centrifugal questions. Though there were variations
in detail, the same fundamental logic operated during the Màsfenate
as I have tried to demonstrate elsewhere.4S i :

— 168 —
 Thus, contrary to what is widely assumed, the main drive of
the power elite was towards maximum proximity to the court rather
than towards maximum local autonomy. The theory of eternal
conflict between the king of kings and his governors should therefore
have to be reformulated. What should perhaps be posed is the
question of what the dynamics of factional politics in court were
and why it led to rebellion.
The power elite tended to rotate round the monarchy rather
than to pull away from it fundamentally because the state system
had built-in mechanisms that maintained this tendency. These
mechanisms could be broken into two components : the double role
of provincial governors; and the system of military organization.
These institutional devices were reinforced by the Solomonic
ideology which was embodied within Orthodox Christianity. Since
Solomonism also embodied the myth of national origin and was
the basis of identity its espousal was mandatory for any member
of the ruling class.46 A total acceptance of its tenets meant that
the country had to be ruled by a member of the Solomonic dynasty.
All of these elements were built into the psyche of governors and
other members of the elite. It is within the framework of this
"mentalité" that the institutional mechanisms have to be viewed
The first mechanism would tie provincial governors to central
institutions. A provincial governor had a number of duties which
could be split into two sets. The first set of functions emanated
from his provincial charge. He was obviously responsible for the
administration of his province. He was at the same time commander
ofthearmyofhis province as well as the highest judge. He was also
responsible for the overall administration of the churches of his
province. The second bundle of duties were discharged at the centre
in the royal court. Basically, he would be the counsellor of the
monarch. He was under obligation to maintain a residence near
the royal court. He would stay in his province for a part of the year
and during the remaining segment he would be in attendance on the
king. He would sit on counsels presided by the king to deliberate
on state matters,47 as well as on imperial tribunals where he would
act as a sort of "jury". Even when there was no counsel or tribunal,
he would spend much of his time in the court simply attending on
the king.
There were also high offices in the state system which in effect
were two offices. The most prominent was the office of the Ras
Bitwâdâd from the 16th to mid- 18th centuries. This officer was at
one and the same time a principal courtier and a prominent governor.

— 169 —
As a courtier one of his tasks was to deputize for the king of kings
when the latter went on campaigns. As a prominent governor, he
would command perhaps the, or one of the, biggest provincial
armies. There were also other officers like the royal fitawrari. This
officer was a governor while at the same time being the commander
of the advance guard of the royal army. There were several other
officers of this type — like Semen agafari, Azal of Yibaba and the
Kântïba of Dambeya during the Gonderine period who had so to
speak double offices rolled into one (49). All these officials could
carry out such scarcely compatible roles because of the institution
of meslâné or endàrasé, or belatén géta. This official was a proxy
who discharged all the provincial duties of a governor while the
latter was away in the court.
Thus, a governor embodied in his person two scarcely
compatible interests and charges:- provincial and central. Since his self
interests were better protected by his success in court politics than
by centrifugal aspirations, he would give priority more often than
not to national over provincial tendencies. If he rebelled, it was
when he had failed in court politics and when rebellion remained
the only option to regain his lost position (48).
The system of military organisation was perhaps the other key
mechanism which checked centrifugal tendencies. I try to outline this
institution elsewhere(so). It would suffice to recapitulate the main
points here.
Military organization was one of the principal institutions built
into the structure of land tenure. Throughout all the agrarian
provinces, there were types of land tenure which require the obligation
of full-time or part-time military service in return for the ownership
of the land. The men who had this kind of land ranged from the
nobility to ordinary peasants. The governor of a district would know
the number of lands in this category. Therefore, he would be able
to know how many soldiers could be mobilised from that territorial
unit.
In order to illustrate this point we will take the gendabdl force
as a case in point. Tradition attributes the foundation of this corps
to Nâgasi, the governor of parts of Sâwa towards the end of the
17thc 51. It served mainly as a trains corps 52. Tradition also
maintains that it was organised along the same lines as the old corps,
namely Kokdb, Afro Aygdba, Aqétzdr, qurban and many others. (53)
Gendabdl soldiers were initially given land at the rate of one gala
of land per soldier in several districts of northern Sâwa. This land
was called Gendabdl mdrét (land).

— 170 —
 The amount of land classified under gendâbal as well as under
gâbar, sâmon and other categories was published by Mahtâmâ Selasé
in the Zekrâ NâgârSA. The figures were presumably obtained
from the archives of the Ministry of Pen. Most probably, the latter
had collected data from various districts. This makes them fairly
reliable and close estimates. Even if they pertained to the first three
decades of the 20thc, they are just as good to demonstrate the
mechanics of the system of military organization and land tenure
in the 17thc as much as that of the early 20 the. Though changes
in the system could not be ruled out, the basic structure seems
to have remained static over these centuries.
Mahtâmâ Selasé has also included in his compilation data on
the number of soldiers transferred from the gendâbal into the
Imperial Body Guard as a trains corps. In the table below we will
correlate the data from randomly selected five districts.

Gendâbal Màrét and Soldiers

Data from Randomly Selected Five Waradas of Northern £âwa

No of Gendâbal
District solidiers
No.
to of
thetransferred
gendâbal
IBG*
Mârét

Sânkora 334 268

Qenbebit 432 334
Ketem Wâyra 350 350
Asagert 369 364
Tara 110 58
Total 1595 1374

Source: 2Jekra Nagar,

*IBG:
p. 116
pp.
Imperial
(for
148,column
149,
Body152
Guard
3). (for column 2) and

There was a total of 1595 gendâbal mârét in the five randomly

selected Wârâdas. The unit used was gala. Assuming that all these
mâréts were owned by peasants, the state could raise 1595 soldiers
from these districts because the arrangement was one soldier per
gala. But the state transferred to the Imperial Body Guard only
1374 soldiers. To account for the difference (221) one has to do a
further investigation.
In any case, the figures clearly prove the point that there were
peasants in the countryside whose obligation for the ownership of
a specified land was military service. There is considerable evidence

— 171 —
to show that the system was the same throughout the old Christian
provinces including- Wâlo, Wârâ Hirnano and Yaju55. In short,
the Ethiopian army drew much of its man power from the landed
peasantry.
.The state had mechanisms of ensuring that the men fulfilled
their obligations in return for their ownership of the land. The
responsibility of supervision fell on the Ziqa sum and malkana in
that order. In cases of default, the matter would be referred to the
district and/or provincial governor. The punishment would range
from a small fine to outright confiscation of the concerned land56.
Normally, the confiscated land would be given to someone else who
had given military service. Rebellion or any other form of lese
majesty would be subjected to the same penalty - i.e. confiscation.57
To be sure, there were soldiers who were regularly given
provisions (known as qâlab) to support them. They were known by
various names, but we would subsume them under the term qaldbana
by which they were called at least in some parts of the country58.
They were dependent on the governor for their livelihood to whom
they tended to be loyal unlike the landed army which was
independent of him. But they constituted a small segment of the army
of any province. On the other hand, it was not possible for a governor
to raise a big qàlâbâna force simply because he could not find the
"funds" (grains, butter, honey, cattle, sheep etc.) quite easily.5*
Thus,. the backbone of the army remained the landed peasant.
To have a bigger army a lord would therefore need to bring
under his controll or governorship a wider territory. Ambitious
governors aspired to be appointed over bigger and bigger provinces.
In addition to benefits like enlarged tributes, a large province would
at the same time equal a large army which proportionately
augmented the political leverage of the governor in the royal court. This was
particularly felt in times of political crises like succession. Much
of the court politicking quite naturally focused on the appointment
for the governorship of such a province.60
This appointment would not however guarantee supremacy for
the incumbent. The court had devices for preventing lords from
acquiring the loyalty of their provincial armies. Travellers from
Alvarez to the 20thc. have unanimously observed the frequent
reshuffle of governors called sum-ler. One objective of this tradition
was to forestall the possibility of the provincial army developing
loyalty to and institutional ties with its commander. If the governor
called upon the army to march against the monarch, the soldiers
definitely ran the risk of being dispossessed of their land in case

— 172 —
the rebellion failed. Therefore, they would not for instance readily
follow a governor who decided to break away from the state61.
A teritorially based army which does not live in barracks is
structurally capable of failing to show up on mobilisation day in cases of
rebellion. The soldiers do not simply come. And if their land is
confiscated by the rebel lord, they would appeal to the king of kings.
They would on the other hand fight in cases where there was a
possibility of their lord and his faction succeeding & putting their
own man on the throne. Therefore, an army could possibly be
raised for a civil war but not as easily for centrifugal objectives.
This situation was modified not before but after 1769. The
success of Mikael Sehul in openly seizing state power created the
circumstance for governors to form institutional ties with the
provincial army. After 1769, they mobilized the provincial army by
their awaj for any war. Defeat of the governor in battle did not lead
to confiscation of land of the soldiers by the winners. The power
of the ras bitwàdad to hear appeals was restricted to cases of capital
punishment. For cases like land confiscation the provincial governor
was the highest judge. Thus, a person was extremely likely to be
penalised for failing to go on a war of rebellion rather than as
heretofore for participating in one.62 This new structure seems to have
developed in the 1770s and became fully operational from the 1780s
to 1855 which enabled the various provincial governors to raise
huge armies and engage in constant civil war.
One of the . common mechanisms by governors to develop
armies loyal to themselves was to dissolve or restructure the old
regiments and corps and organise in their place new corps and give
them land. We have evidence to show that Ali I organized a new
force and gave them land called fAA, H*1^/* a»&ft (Yd-Ali Zcimaç
Mârét) ; so did Kolasé Amâdé of Wârâ Himâno and Wâllo who
instituted Yà-galla maret <tp\ tw&ty. We have already referred to
Nâgasi and his gendcibdl corps. §wm Agamé Wâldu and his son,
Subagads took a similar measure though it is very difficult to say
much because the tradition is rather elliptical.63
The very name of the land distributed by Ali I is highly
suggestive of the fact that the new corps were tied to the courts of the
regional lords. It would have been a grave breach of tradition if a
lord could do the same things in times when the monarch was strong.
This structural factor prevented to a very large extent the Wârâ
Séh rulers from enforcing their full authority on their governors. It
forced them to become accommodating suzerains rather than absolute
sovereigns. But even in this kind of system governors could not laun-

— 173 —
ch a sustained campaign to break away from the Centre. Ordinarily
when they rebelled they endeavoured to league with other lords for
the simple reason that the military balance of power would be tilted
against them when they fought individually as their suzerain could
summon the other provincial armies to his succour. The very
structure of the state system dictated that major wars in this era had to
be fought for seizure and control of state power and not, as is
universally assumed, for centrifugal motives.
On could counter this argument by pointing to the numerous
rebellions that wracked the country over these centuries. Fortunate
ly detailed empirical studies have been made on seftnat-onQ aspec-
of rebellion-for the 19th and early 20th centuries 64. These studiest
demonstrate that the big leftas were men who belonged to office
holding families but who had been removed from or denied office.
In becoming a lefta, they aspired to occupy the offices they were
denied or even higher ones.
Thus, leftnat was an avenue of last resort to be integrated by
the Centre. The major provincial rebellions of the first half of the
18thc. also fall within the same framework. Lords launched their
rebellions against the monarch by no means to break away from
Gondar. On the contrary, it was to Gondar that they marched
with the declared objective of replacing the monarch by a man of
their choice — or put another way, to seize state power. 65. This can
be explained by the fact that rebellions were taken as a device to
retain a threatened office or as a mechanism for upward mobility 66.
As such, therefore, governors individually were a potential source
of destabilization to the throne. This latent trait rather gives the
class a split personality to be at one and the same time an intrinsic
part of, and a potential threat to, the royal court 67. But, it should
be emphasized that lords could be a threat only to a particular king
and not to the institution of kingship as a whole.
It is therefore highly debatable to present the recurrent political
crises in the country's political history in terms of monarchy vs.
nobility, centripetal vs. centrifugal forces. The question had to be
reformulated in a different manner: what was the basic cause that
ultimately led political factions in the imperial court to resolve
their conflicts by resort to arms rather than by peaceful means?
This question in turn raises the question of establishing the dynamics
of factional and coalition politics and how these groups struggled
particularly during times of succession strife. Identifying the self-
interests of the power elite, their aspirations and value systems,
in short, the elements that "make them tick" would contribute a
great deal towards answering these research problems.

— 174 —
 NOTES
'
1. J.S. Trimingham, Islam in Ethiopia (London: Frank Cass and Co. Ltd, 1965), p. 104.
2. Ibid., p. 105. • ■
3. Ibid., p. 107. * •
4. M. Abir, Ethiopia: the Era of Princes (London: Longmans, 1968).
5 . D.Masafent,"
Crummey,International
"Society andJournal
Ethnicity
of African
in the Historical
Politics of Studies,
ChristianVIII,
Ethiopia
2 (1975),
during
p. 270.
the Zamana
6. Sven Rubenson, "Ethiopia and the Horn," in J. E. Flint (éd.), The Cambridge History o
Africa: from C. 1790 to C. 1870, Vol. V (Cambridge: Cambridge University Press, 1976), p. 57
*
• 7. Abir, chapters 2, 5 and 6. ,

'

,
8. D. Crummey, Priests and Politicians: Protestant and Catholic Missions in Orthodox Ethiopia
1830-1868 (London: Clarendon Press, 1972), pp. 29, 92; S. Rubenson, The Survival of
Ethiopian Independence (London: Heinemann in association with Addis Ababa University Press,
1 their
rep. 1978),
synonyms
p. 35.inThese
their various
two historians
writingsapply
for the
the Wârâ
term Séh
"hegemony"
Mâsafent.and "predominance" or
9* Shiferaw Bekele, "The State in the Zamana Masafent (1786-1853): an Essay in Reinterpreta
tion", in Taddesse Beyene, R. Pankhurst and Shiferaw Bekele (eds.), Kassa andKassa: Paper
on the Times, Lives and Images of Tewodros II and Yohannes IV (1855-1889) (Addis Ababa
forthcoming). • ' . . ,' ,
10. Alaqa Tayâ Gâbrâ Maryam, History of the People ofEthiopia trans. Grover Hudson and Tekeste
Negash (Uppsala: Centre for Multiethnic Research, Uppsala University, 1987), p. 112; Tâklâ
' 4th
Sadiq
rep.Mâkuriya,
by the Ministry
Yé-Etyopiya
of Education,
Tarik: Ka1961,
A%èE.C.),
Lebnàpp.
Dengel
313,eskà'Açé
314 et passim.
Tewodros (Addis Ababa:
11. Traditional historiography places Ali's seizure of power in early 1784 following the victory
over Tâklâ Giyorgis at Afârâ Wanat by a coalition of lords in which AH was a prominent
member. See Zeneb, Yâ-Téwodros Tarik (Princeton: ed. and pub. by E. Littmann, 1902),
pp. 2-4; Bayru Tafia (trans., ed and annot.) Asma Giorghis and His Work '.History of the Galla
and the Kingdom of Sawa (Stuttgart: Steiner-Verlag-Wiesbaden-GmbH, 1987), pp. 434, 624.
Afârâ Wanat was significant in the career of Ali because he was able to move up to the position
of DâjazmaS of Bâgémder in the subsequent gum- ger.H. Weld Blundell, (trans, and éd.),
The Royal Chronicle of Abyssinia, 1769-1840 (Cambridge: Cambridge University Press, 1922),
p. 342; D. Crummey, Çâôâho and the Politics of the Northern Wâllo-Bâgémder Border,"
Journal of Ethiopian Studies, XIII, 1 (January, 1975), p. 7. But he did not seize supreme power.
In the list of the new appointees given by the chronicler, he was the third coming after "Kanaz-
mach Tsadalu [who was appointed] Bitwadded and Ras Khaylu [who was appointed] Dajaz
mach of Gojam among the Agaw..." Weld Blundell, p. 342. Ali used the subsequent two years
to consolidate his position and groom himself for supreme power. It was in 1786 that King
Tâklâ
meder."Giorghis "promoted Ras Ali to be Betwadded with the title of Dajazmach of Bega-
Ibid.,
- 1786 as the starting
p. 365.
yearCrummey,
of the Wârà
"ÇâÇâho...",
Séh Masfenate
p. 7.notwithstanding
Therefore, it is proper
the views
to of
taketraditional
the year
historians. -, ,
12. I Guidi, Annales Regum Iyasu II et Iyo'as (Roma, 1912), p. 237.
13. Ibid., pp. 235-237; J. Bruce, Travels to Discover the Source of the Nile in the Years 1768, 1769
1770, 1771, 1772 and 1773, Vol. II (Edinburgh, 1790), pp. 682-686.
14. This conclusion is based on the first-hand accounts of the fall of Go§u, the ras Bitwadâd folio»
wing Mikael's defeat at Sarbakussa in 1771. Bruce, Vol. IV, pp. 245 — - 47.
15. Ibid., p. 247; Weld Blundell, pp. 321 ff. The chronicler gives an account of the activities of
Tâklâ Haymanot which clearly showed t hat the king held power. '
.

16. This generalization is based on accounts of the activities of King Tâklâ Haymanot in Ibid. <
passim.Tàk\â Haymanot managed to get the removal of Go§u, the overbearing and despotic ras
bitwadâd, the successor of Mikael, whom he tried to replace in all his powers. Bruce, Vol. IV

— 175 —
 ■ pp. 245-247. He stayed on the throne upto 1777. He was succeeded by Solomon whose attempt
to assert his authority ended in his own deposition by the lords in 1779. Tâklâ Giorgis
succeeded. He set out on several expeditions at the head of an army he fully commanded. But
his defeat in 1784 seems to have given the last blow to whatever authority the Kings used to
xercise since 1771. These three kings commanded regiments, exercised considerable §«»*-§«•
rights, kept court and were shown much difference by the lords and the people. See Weld
Blxxdell, passim; Bayru, p. 434. Asmé states that Tâklâ Giorgis's "position was strengthened
by the support of the Tegroch and the Gojamoch." Ibid., p. 435. Iyasu III (1784-1786), the
successor of Tâklâ Giorgis, was a puppet king.
17. Tâklâ Giorgis indeed took energetic measures. Though the chronicle is not very helpful in
reconstructing a detailed account of his reign, it enables the reader to see that he exercised
considerable power and authority. He could not be defined as a puppet in the true sense of
the word. See note 16 above. See also 'Ibid., pp. 232-313. /
18. Aligaz's period of rule was punctuated by big civil wars He did not exercise authority con-
. tinuously through the years. Rases Asrat and Wâldâ Gabriel had scored a victory over him
at one time, pushed him out of Bâgémder which they ruled for sometime. ' See Bayru, pp.
436-439. . • " . \ . •
'

19. • Crummey,, "Çavaho...", p# 7; Abir, p. 31. * .

,
20. Elders interviewed in Dabra Tabor and its environs in the summer of 1985 clearly remembered
the peace and "prosperity" which marked the reign of Gugsa. I am indebted to Agegnehu
x Gebeyehu, graduate of the Department of History (A.A.U.) for letting me see the notes he
took during these interviews which he conducted. ■■>. « . .
21. Taddesse Tamrat, Church and State in Ethiopia, 1270-1527 (Oxford: Clarendon Press, 1972),
p. 96. Taddesse used the expression in the context of his description
> of the structure of the
'
early Solomonic state (1270-1527).
'
22. Book of Judges, chapter 21, verse 25, King James edition of the Bible as cited in Abir.
23. Gâbrâ Selasé, Tarikà Zàman Zâ-Dagmawi Menelik, Negusà-Nàgast Zà-ltyopya (Addis Ababa,
f 1959talks
he E.C.),
of Séru
p. 118.
Gwangul
Alâqa Tayâ
and his
(p. marriage
82) uses exactly
to Gâlàbu,
the same
"the sister
sentence
of Ras
as Gâbrâ
Faris,Selasé
the balabat
when
of Lasta and Sâlâwa." -,.'.' ,

.
24. Crummey, "(?â$âho...," p. 7. " • . - \
* •
25. Gâbrâ Selaseé p. 118; Tayâ/ p. 82.
Gâbrâ*
26. Selasé/ p. 117. ' • • ' ' ■ ! - .
27. Ibid., p. 120. ' • . - " ■ < » .
28.' • See the genealogy of the Mamâdoc family in Zefgaw Asfera, "Some Aspects of Historical
• - pp.
Development
70-72. 'in Amhara, Walo ca. 1700-1815" (B.A. Thesis submitted to' ' HSIU,- 1973),
'-/'-,
29. For an overview of the family history of the Wârâ Séh, see D. Crummey, "Family and Property
amongst the Amhara Nobility," Journal of African History, XXIV (1983), pp. 207-220.
30. Abir, p. 29. . "
31. See the profiles of noted clerical scholars of the past few centuries in Academy of Ethiopian
Languages, Yâ-Gé'ez Qene'yat Yà'-senâTebab Qeres: Nebabu Kânâ Terguaméw (Addis Ababa:
Ministry of Culture and Sports, 1980 E.C.), pp. 433-455. • x '
32. See the genealogy in Gâbrâ Selassé, pp. 116 ff.
i • "' ■< - • , '
33. James McCann. From Poverty to Famine in Northeast Ethiopia: A Rural History, 1900-1935
(Philadelphia: University of Pennsylvania Press, 1987), pp. 103-4, 110-122.

34.' political
See, amongAspect
others,
of the
Bairu
Menelik
Tafla,Period,"
"Marriage
Journal
as a Political
of Ethiopian
Device:
Studies,
an Appraisal
X, 1 (1972),
of the
pp. Socio-
13-21.

, — 176 —
35. See Tâklâ £adiq Mâkuriya, Asé Téwodros éna Yd-Etiyopia Andenât (Addis Ababa: Kuraz
Publishing Agency, 1981, E.C.), pp. 37-38.
36. Shiferaw Bekele, "Prayers and Rulers in Ethiopian History: Notes for a Structural History
of the Solomonic State" (forthcoming) • \ ■
37. Crummey, "Society and Ethnicity...", pp. 266-278.' . ' " '.
38. Ibid.; Abir, chapters 2, 5, 6.
39. Crummey, "Society and Ethnicity...", p. 278.
40. Ibid. ', ,, ■ i. *

'
41. Isenberg who collected information for his dictionary in the Zâmàna Mâsafent describes
"Amhara", inter-alia, as follows: "The name is used by common people to signify a Christian, ,
in contrast with Mohammedans and Heathens." C. W. Isenberg, Dictionary of the Amharic
Language (London, 1841), p. 115 b. When he was compiling his Amharic-Italian and Italian-
Amharic dictionary at ths end of the 19th C, I. Guidi was assisted by a well-known Ethiopian
scholar,Alâqa Keflâ Giorghis. Guidi defines Amhara as "Cristiano (opp. a GzX\^)."Vocabolario
' Amarico-ltaliano (Roma, 1901), p. 421a. Bastman, who in the like manner, was assisted by
the famous Alâqa Kidanâ Wâld and his student Dâsta Tâklâ Wâld definesAmhara as"Chrétien
par opposition avec les Gallas autrefois paisns," J. Baetman, Dictionnaire Amarigna-Français
(Dire-Dawa, 1929), p. 556. '
42. D. Crummey, "Abyssinian Feudalism", Past and Present, 89 (Nov., 1980), pp. 124-5; see
also idem, "Society and Ethnicity...", p. 269/ . •
43. Perham, The Government of Ethiopia (London: Faber & Faber Ltd., 2nd ed.1, 1969), pp. 267-8.
44/ Shiferaw Bekele, "Kassa and Kassa: The State of their Historiography," in Taddesse Beyene,
R. Pankhurst and Shiferaw Bekele (eds.), Kassa and Kassa: Papers on the Times, Lives and
Images of Tewodros II and Yohannes IV (1868-1889) (Addis Ababa, forthcoming), sections
h, in, v.
45. See note 9 above.*
46. Much has been written on this aspect of the Solomonic ideology. For the earliest and still one
' of the best analyses, see D. Levine, Greater Ethiopia: The Evolution of a Multiethnic Society
. : (Chicago: University of Chicago Press, 1974),- pp. 92-112.
47. . Historians have not so far recognized the institution of royal council. The scholar who give»
" an
it some Institutional
thought Analysis
is L. Berry.
of Kingship
See L. Berry,
in the"The
Christian
Solomonic
Kingdom
Monarchy
of Ethiopia"
at Gondar,
(an unpublished
1630-1755:
Ph. D. dissertation submitted to Boston University, 1976), pp. 242 ff. On the other hand»
travellers do mention a "council." Of all the observers, Bruce gives a detailed description of
this institution including sitting arrangement and protocol of deliberation. He emphasize»
the fact that it was a viable and permanent central institution. See Bruce, Vol. Ill, pp. 266-270,
275, 280. The order of precedence within the counsel that he gives tallies with minor variation»
with the order given in the Chronicles as well as in the Ser'ata Mangest. I. Guidi (trans.) Ann-
. 1i ales lohannis I, Iyasu I,Bakaffa,pp. 152,313,314. Bairu Tafla and H. Scholler (ed. and trans.)
"Ser'ata Mangest: An Early Ethiopian Constitution", (Mimeo, Addis Ababa, 1974. Available
in the Institute of Ethiopia Studies.) Perhaps, an important feature of this council as an
institution was that it was not clearly defined in terms of membership and above all attendance.
48. This generalization is based on accounts of rebsllion given in Berry, pp. 281,332. Berry*»
' , accounts are based primarily on chronicles. - -,
49. . For the office of ras bitwadàdsee Ibid., pp. 304,308. For the other offices see I. Guidi (trans.),
Annales Regum Iyasu II et Iyo'as, pp. 184-5 (French) et passim. If we take the era of Menelik,
this element becomes abundantly clear. Ras Mekonnen acted as a kind of "foreign minister'*
in addition to his governorship. Fitawrari Habte Giorghis carried several court functions on
top of his duties as a governor of a huge province right upto his death in 1926. Cases of thi»
: kind abound when one closely investigates the nature of court offices.
50. Shiferaw Bekele, '«Kassa and Kassa: The State of their Historiography", Section VII.

— 177 —
51. Bayru, pp. 502-504; Berhanu Abebe, Evolution de la Propriété foncière au Choa (Ethiopie)
du Règne de Menelik à la Constitution de 1931 (Paris, 1971), p. 154.
52. Ibid.x for a good definition of gendabâl based on a synthesis of the sources, see J. Mantel»
Niecko, The Role of Land tenure in the System of Ethiopian Imperial Government in Modern
Times Trans. Krzysztof Adam Bobinski, Warsaw, 1980), pp. 88-90.
53. Bayru, pp. 138-41.
54. Mahtâmâ Selassé Wâldâ Mâsqâl, Zekrâ Nâgàr (Addis Ababa, 1942 E.C.), pp. 143-154. .
- See N. Pearce, The Life and Times of Nathaniel Pearce Written by Himself during a Residence
55» in Abyssinia from the Years 1810 to 1819; Together with Mr. Coffin's Account of his Visit to
Gondar, Vol. II ed. by J. J. Hall (London: Colburn and Bentley, 1831), pp. 220 — 226, 264
for an account of an incident between a lord and the abun over some districts which were
-. under the jurisdiction of the latter. The lord wanted to get these districts for himself because
"the soldiers belonging to those districts... are more than three thousand, and some of them
• my best men, the loss of whom will weaken my army..." (p. 220); see also MacCann, p. 161,
"j for a description of how the system works in Northeast Ethiopia; Assefa Maru, "Land Tenure
in Shadaho-Maqet Warada from 1632 - 1974" (B.A. Thesis Submitted to AAU, 1989), pp.
10-38. Mantel-Niecko, passim. * . - . ...
Oral informant: Ato Bâqâlâ Jimma, age 68, interviewed in Addis Ababa; W/o Tâwabâfi Ayâlâ,
56. age 65 at the time of interview in 1979 E.C. in Addis Ababa, informed me thatlong before
the Italian occupation the governor of their province in Northern u.câwa ruled that her
grandfather's gendabâl Màrét be confiscated because he had failed to participate in a campaign.
He was therefore reduced to the status of a gdbar and tisàha (i.e. a tenant). I am grateful to
Ato Dârâjâ Gâbré of the ILS, AAU, for introducing me to this woman. See also Gâbrâ Wâlâ
Engda Wârq, Yà-Ityopya Maretna Geber Sem (Addis Ababa, 1948 E.C), passim.
57. Abâbâ Fisseha, a graduate of the Department of History, at present an academic
staff of AAU, has collected tradition from Tehulâdâré which show large scale confiscation
of land following the defeat of Negus Mikael in October, 1916.
58. Fussela, Yate Tewodros Tarik (Roma, 1959), p. 73; Tâsâma Habtâ Mikael, Kâsate Berhàn
Jàsà'ma Yd Amaratta Mdzgnbà Qalat (Addis Ababa, 1951 E.C), p. 349a gives the definition
of qâlâbâna. Dâstâ Tâklâ Wâld, Addis Ya Amarehna Mdzgâbà Qalat (Addis Ababa, 1962
E. C), p. 1241b gives the term <&<nfl A which is roughly equivalent to ♦Afi*? There could
.

also be other terms synonymous to it.

59.' The lords of the Zâmânâ Mâsafent seemed to have found it difficult to find the necessary
resources to support a big qàlàbâha force. In any case, this is the impression that one gets
from reading about actual starvation breaking out in the camps of none other than Ali II,
Webe and Bern GoSu. For Ali & Bern, see Plowden, pp. 421-3, 228; and for Webé, Combes et
, Tamisier, I,. p. 239. ' '
60. There are well-known cases in this regard.In 1767,following the death of Wâldâ Le'ul a huge
crisis arose as a result of the scramble for the office of the ras bitwàdâd. In 1908, when Yelma
Mâkonnen, the governor of Harar, died Balia had to bribe & do what all hs could to get the
governorship of Harar. There was a great deal of politicking over Harar from the death of
Makonnen in 1906 upto the coup d'état of 1916.
61. It is not quite easy to document this factor. Ths most well-known case is the abortive rebellion
of Bal 2a in 1928 which was easily forestalled because the army abandoned him.
62. These conclusions are drawn from reading the unpublished chronicle of Gojam, Tâklâ Iyasus,
"Yâ-Gojam Tarik" (IES MS no. 254), passim and Asmé's history. See also Gâbrâ Wâld, p.53.
63. Ibid.. ' .
64. D. Crummey, "Banditry and Resistance: Noble and Peasant in Nineteenth Century Ethiopia,"
in S. Rubenson (éd.), Proceedings of the Seventh International Conference of Ethiopian Studies
(Addis Ababa, Uppsala, East Lansing, 1984), pp. 263-277. R. Caulk, "Bad Men of the Borders:
Shum and Shefta in Northern Ethiopia in the Nineteenth Century," Proceedings of the Second
Annual Seminar of the Department ofHistory ofAAU, Vol I, (Addis Ababa, 1984) pp. 39-104;
— 178 —
 T. Fernyhôugh, "Social Mobility and Dissident Elites in Northern Ethiopia: the Role of
Banditry, 1900-69", in D. Crummey éd., Banditry, Rebellion and Social Protest in Africa
(London: J. Currey and Heinemann, 1986), pp. 151-172.
65. See note 48 above.
66. See H. Erlich, "Tigre in Modern History", in S. Rubenson (éd.). Proceedings of the 1th
International, p. 328 where he makes a similar conclusion on the basis of archival research
regarding the collaboration of Tigrean lords with Italians in the 1930s: "Collaboration with Asmara
against Addis Ababa did not stem from a separatist instinct or a modern sense of Tigrean
nationalism. Rather, it was always a means of obtaining promotion at home within the
framework of traditional Ethiopian power game."
67. There seems to have existed a fairly similar structure in other African polities which suggests
that a comparative study of Ethiopia with the rest of Africa will offer a great deal of insight
into the history of this country. See, for example, J.H.M. Beattie, "Bunyoro: an African
Feudality?" Journal of African History, V, 1 (1964), pp. 27, 34.

— 179