African Studies Program, University of Wisconsin-Madison

Regents of the University of Wisconsin System

The Dynamics of Change in Pre-Colonial East African Societies

Author(s): Abdul H. Sheriff
Source: African Economic History Review, Vol. 1, No. 2 (Autumn, 1974), pp. 7-14
Published by: African Studies Program, University of Wisconsin-Madison
Stable URL: http://www.jstor.org/stable/3601096
Accessed: 29-10-2015 16:25 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

African Studies Program, University of Wisconsin-Madison and Regents of the University of Wisconsin System
are collaborating with JSTOR to digitize, preserve and extend access to African Economic History Review.

http://www.jstor.org

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 ARTICLES

THE DYNAMICSOF CHANGEIN PRE-COLONIALEAST AFRICANSOCIETIES*

A curious convergence in the conception of pre-colonial African societies

has been occuring between people whose pre-conceptions and pre-occupations are
highly divergent. On the one hand there is the "Colonial" school which views
pre-colonial African societies as Primitive. Even though the term is now being
replaced by the less objectionable but no less sterile term "traditional", the
conception of pre-colonial, and indeed of much of modern Africa, even among many
of the modern-day "Africanists", is one of a "tribal" stage of development in
which economic and social forces as exemplified by agriculture, is stagnant at
a very low level of development. On the other hand there is the "nationalist"
school which argues that pre-colonial African society was characterized by
equality, and by "the egalitarian nature of land of tenure system"; in short,
pre-colonial African societies were classless.2 As will be obvious, these are
not explanations at all but merely assertions of the inherent characteristics
of the African genius which prevailed during the whole of the long and complex
history of pre-colonial Africa and came to grief only with colonialism of the
late 19th century. Ironically, the implications of this idyllic picture of the
African past lead the nationalists to the same conclusions as the colonial school.
As Rweymamu (p. 3) and Hymer (p. 33) point out, pre-colonial equality was accom-
panied by a low level of material production and a low degree of specialization
and exchange. Secondly, by applying this description to the whole long history
of pre-colonial Africa the nationalists are unwittingly subscribing to the no-
tion that pre-colonial Africa was also stagnant. Thus equality, poverty and
stagnancy, with varying emphasis, seem to characterize the conception of pre-
colonial Africa of both the colonial as well as the nationalist schools.

One thing that both these schools seem to deny to pre-colonial Africa is
the internal dynamic of socio-economic development of African societies which
does indeed imply the development of social differentiation and inequalities,
and the development of productive forces of the societies. Historical evidence
for these developments in pre-colonial East Africa are by no means lacking.
There are many examples of agricultural intensification going beyond the alleg-
edly universal "subsistence shifting cultivation" which is supposed to charac-
terize much of tropical Africa.3

There are also many examples of division of labour, of non-agricultural

specialization and exchange over large areas of not only the rare commodities
such as iron and salt, but also of surplus foodstuffs on a regular basis.

Recent studies of pre-colonial history of East Africa have shown that while
"primitive communism", in which there was no exploitation of man by man and in
which, therefore, there was no need for a state to protect the interests of the
ruling class, undoubtedly existed in certain parts of East Africa; in many cases
African societies had passed well beyond the classless "tribal" stage and had
established the state which implies the existence of classes in the society.
What were the factors that brought about these socio-economic and political
changes before the era of long distance trade?

*This paper was presented at the University of East Africa Social Science
Conference in DecAer of 1973. The author hopes to provoke discussion, and
would like to correspond with others interested in the topic.

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 The first factor that obviously influences the development of societies is
the environment. It sets limits to human endeavours, but it must be emphasized
that these limits are highly elastic, determined by the level of technology at-
tained by a given people which, in turn, may be determined by other social con-
siderations. With a low level of technology man largely adapts to the environ-
ment, extracting wild foods by hunting and gathering. With the development of
technology man can manipulate the environment to produce food through pastoralism
or agriculture. The environmental factor, therefore, does not "determine" the
development of societies as the geographical determinists believed. However,
we should not under-estimate the influence that the environment exercises on a
particular society at a certain level of technology, on the type of food-producing
system they can practice, on the density of population that the land can support,
and on the size of the surplus that can be produced to provide the motive force
for the further development of the productive forces.

The second factor in the development of societies is population whose dis-

tribution and size is certainly influenced by the natural environment. The
Malthusiasts, however, believe that the natural environment determines the supply
of food which is, therefore, inelastic, and which, in turn, governs the rate of
population growth. This approach does recognize the possibility of technological
change which may increase the supply of food, but it considers technical inven-
tions and imitations extraneous and fortuitous factors. Thus population growth
is seen as the dependent variable, determined by preceding changes in agricultural
productivity.

Boserup6 on the other hand, represents the anti-Malthusian approach which

considers population growth as the independent variable which, in its turn, is a
major factor determining agricultural, social and political developments. She
argues that "under the pressure of increasing population there has been a shift
...from more extensive to more intensive System of land use." (p. 16) However,
this shift is not possible without the essential changes in agricultural techniques.
Under the most extensive "forest-fallow" system of land use, the land is cultivated
for 1 - 2 years and is allowed to regain its fertility under a fallow of 20 - 25
years. Thus the cultivator needs at least ten times as much land as he actually
cultivates during any particular year, and population density may thus be as low as
ten per square mile. Under these conditions the cultivator needs only the ax and
fire to clear the forest, the ash from burnt vegetation for additional fertilization,
and a digging stick to punch holes for planting. However, with increasing popula-
tion, when the fallow is shortened to 6-10 years, the land gets grassier and a hoe is
needed to prepare the land; and because the land accumulates less vegetation during
the shortened fallow, and since it may be cultivated for a longer period, it may
become necessary to collect a large amount from the surrounding bush to burn on the
cultivation plot in order to obtain enough ash for fertilization. However since
the reason for shortening the fallow was land shortage, this system of fertiliza-
tion could not have been too stable, since it undermines the regeneration of the
surrounding land. It has been reported only for the Bemba in N. Zambia7. When a
further increase in population calls for further intensification of land use, the
fallow may be reduced to as little as 1-2 years which does not allow the bush to
regenerate. The land, therefore, becomes more grassy. Fire is no longer as help-
ful in burning the increasingly fire-resistant grasses, and the hoe is no longer
adequate to remove their roots. Ox-drawn ploughs thus become essential for the
land which can no longer recover its fertility naturally under fallow. The in-
creasingly grassy vegetation may attract pastoralists into such areas, and their
presence may thus facilitate the transition to this system of land use. However,

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
if the cultivator is unable to use the plough owing to lack of animals or other
technical handicaps, he may make the transition directly to annual or multi-
cropping with a short fallow or no fallow at all. This prevents the establish-
ment of wild grasses and, therefore, the hoe remains adequate. The land, how-
ever, requires a massive amount of fertilization by man with green and animal
manures and composts, The system may also call for irrigation and terracing to
conserve soil and to permit year-round cultivation.

Boserup argues that when a change in the land use system is imminent as a
result of increasing population density, the fields first chosen for more frequent
cropping would be among those relatively well suited for the next step. Moreover,
a further increase in population density may bring about further intensification of
land use in the most favorable areas rather than an orderly transformation of the
whole region from the oldest, most extensive system to the second stage of develop-
ment. Land use systems, therefore, can co-exist within the same region, and even
within the same village (pp. 58-9).

It is obvious from the description of the evolution of land use system that
productivity per unit of land is increased considerably with intensification of
agriculture. However, Boserup argues that in pre-industrial societies it is
unlikely that this productivity fully compensates for additional inputs of human
labour. In other words, there are diminishing returns to labour with agricultural
intensification. The cultivator tries to make up for his decIning productivity by
increasing the number of working hours from as little as an average of 3 hours under
the forest-fallow system to a full working day under annual or multi-cropping sys-
tems. Therefore, "the cultivator would find it profitable to shift to a more in-
tensive system of land use only when a certain density of population has been
reached," if the necessary technical knowledge exists (p. 41). An obvious corol-
lary of this conclusion is that if population pressure declines, or if the cultiva-
tors migrate to sparsely populated regions, there is a corresponding technical
regression in agricultural methods which increase labor productivity even if pro-
ductivity of the land declines at the same time (p. 62).

Boserup goes further and attempts to show the relationship between the land
use system, and therefore population density, and land tenure system. She argues
that under the most extensive land use system when the cultivator may shift every
few years, it is not ownership but the right to use land that is crucial. Already,
at that stage, that right is confined to members of the tribe and is denied to
"strangers", captives, etc., who, however, may be "adopted" by the tribe as full
members. Thus the tribe as a whole retains ownership of the land which is alloca-
ted to the members for use.

Moreover, even at the more extensive "land rotation" stage ("shifting culti-
vation" is a misleading term), the cultivator retains his rights to the land under
fallow unless he fails to take up his option to re-occupy that land at the end of
the fallow period which may be as much as 25 years. With increasing population,
good plots of land become scarce and begin to be more intensively cultivated un-
til, under the annual or multi-cropping systems, the cultivator occupies the land
more or less permanently and acquires property rights. He may continue to exer-
cise these rights even under feudalism in which the nomadic conquerers may exact
a tribute in kind and labor without necessarily appropriating the land (though the
latter also occurred in certain parts) (pp. 79-84).

The final significant part of Boserup's equation attempts to trace the rela-
tionship between population growth and the development of economic, social and

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 10

political organizations. She argues that under the more extensive land use system
the sparse population, concentrated in small, impermanent settlements, can produce
food surpluses but lack of market may prevent its production. Therefore only
rudimental division of labor is possible, and the production of tools etc. is
little developed. Under such conditions an elaborate social organization is
unlikely to develop, though a chief with some authority may exist, receiving
conventional and ceremonial gifts from tribesmen. With the intensification of
agriculture and more settled population greater division of labor and market may
develop where surpluses of food may be exchanged for non-agricultural goods.
Full-time artisans and traders emerge ultimately, and under permanent cultiva-
tion the region may also be urbanized. Thus "tribal organization gives way to
more differentiated forms of social organization." Specifically, as agriculture
is intensified and forests and bush degenerate into grasslands, nomads may appear
who may conquer the agriculturists and establish their feudal overlordship, levy-
ing their tribute in kind and in labor (pp. 70-1, 82).

This is an extremely attractive theory which seems to identify the dynamics.

of economic, social and political change in all pre-industrial societies. Some
social scientists have attempted to apply the first portion of the equation and
have suggested that "population growth has been a major cause of agricultural
change in Tanzania" while ignoring the rest of the equation. It is not diffi-
cult to find some examples of land use systems which seem to correspond to popu-
lation densities. The Tunduru system in S. Tanzania, which includes cultivation for
three years and a fallow of at least twenty years, supported a population density
of about ten persons per square mile.9 On the other hand, the more intensive
Makonde system, also in S. Tanzania, in which the land was cultivated for three
years and rested for six-nine years, could support a population density of 30
(Allan, 214-16). Ukara, a small island in Lake Victoria, seems to offer an
ideal test case in which the very dense population appears to have evolved
perhaps the most intensive system of land use in sub-Saharan Africa. The roughly
30 square miles have supported a population of about 16,000, giving a population
density of over 500 per square mile. To cope with such a dense population the
islanders have, over the past couple of centuries, evolved a land use system
which includes stall-feeding of cattle, application of massive amounts of animal
and green manures, crop rotation, pit-cultivation on steeper hills similar to
the matengo system, terracing, erosion control and irrigation, all intended to
permit an intensive permanent cultivation system. The productivity of the land
has been vastly increased, yield per acre being three times that on the mainland
under more extensive land use systems. This has been brought about by a massive
input of labor at diminishing returns. Thus, whereas the Sukuma on the mainland
need to work an average of only about six hours a day, the Kara regularly work
for 12 hours. However, Kara productivity per hour is only about a third that on
the mainland.10

While these cases seem to confirm this part of the equation, there are other
examples which raise grave doubts about its applicability. In a study on rural
Hauseland Polly Hill shows that although land is not scarce, the bulk of the crop
is produced on intensively cultivated, manured farmlands.1' A recent attempt to
operationalize this equation in field research has come to the conclusion that
"agricultural systems in the Uluguru mountains are not, on the whole, congruent
with population density".12 Obviously the partial equation that seems to hold in
some cases but not in others cannot provide a firm basis for speculation on the
dynamics of agricultural change. Failure to apply this portion of the equation
may be related to the lack of recognition that though compartmentalization may be
an academic convenience, life is essentially indivisible. Boserup's theory was

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 11

not confined to the population/land use equation alone but attempted to deal with
the whole society. Its validity or otherwise, therefore, cannot be checked by
considering only a segment which cannot stand on its own.

Boserup's theory is essentially an argument in linear causation:

Population Agricultural Sedantarisation Division of Development of

growth Intensif ication labor and -1+ social and politics
g nteification _ Individualisationorganizations
of property rights exchange organizations

No doubt examples can be found which would appear to confirm this model of change
in which we should expect the more extensive agricultural system to go along with
an underdeveloped economic system and a decentralised political system; and, con-
versely, an intensive agricultural system to correlate with a developed economic
system and a centralised political organization. But here again we have examples
which do not conform to this equation. Richards comments that the complexity of "...
Bemba political organization forms an interesting contrast to the simplicity of
the people's economic system" (p. 25). On the other hand Ukara, which evolved the
most sophisticated agricultural system in sub-Saharan Africa, had "no sultans or
Village headmen" before the advent of the Europeans. The Island was "administered
on democrafic lines with a council of elders composed of the heads of enlarged
families."

There are two major problems with Boserup's theory which limit its applicability.
The first is the fact that though she presents a whole system to explain change in
pre-industrial societies, that system was evolved piece-meal and in a linear fashion,
concentrating attention on the relation between population growth and agricultural
change and then trying to trace the consequences of that change. Her conception is
not of a dynamic system in which a change at one point reverberates throughout the
system. Specifically, changes in the political system, some of which may be induced
by extraneous factors, could bring about changes in agriculture which do not owe
their origin to any increase in population. As Hymer points out, land, even where
it is plentiful in relation to population density, can be made scarce if certain
groups gain the power of excluding others from its use (p. 34), and Boserup herself
cites such a possibility under feudalism (p. 84).

This points to the second major problem in Boserup's theory: her failure to
recognize the crucial roles of (a) the surplus, and (b) the surplus-appropriating
ruling class in the transformation of agricultural systems. As Boserup herself
recognizes, the production of food surpluses is potentially large even under the
most extensive forest-fallow system though they may not be produced for lack of
demand (p. 70). In a system in which the cultivator needs to work an average of
only about three hours a day to produce adequate foodstuffs for subsistence, that
surplus is used in the form of longer hours of leisure. A portion of the surplus,
however, is likely to be produced as an insurance against famine. Subsistence
cultivators, therefore, tend to cultivate an area at least large enough to ensure
food supply in a season of poor yields. Therefore, in the average year we should
expect the production of the normal surplus which, if not needed, after all, may
be consumed in beer parties or for prestige, or may be distributed as gifts to
relatives. (Allan, p. 38) However, for any development to occur there is a need
for the production of a more regular surplus to feed those involved in the
produc-
tion of tools and services, i,e., to facilitate the social division of labor and

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 12

specialization in non-agricultural activities. Thus, the production of a surplus

becomes essential for economic developmeht in general, and an increase in popula-
tion will not be allowed, if possible, to eat into this surplus before agricultural
change occurs to raise the level of productivity.

Hand in hand with the social division of labor occurs social differentiation
which heaks the egalitarian constitution of the "tribal" or "primitive Ujamaa"
stage. As Engles points out, "the exercise of a social function was everywhere
the basis of political supremacy."15 Control over an essential commodity or function
permits a section of the community to extract a larger return for its labor. In
many pre-colonial African societies control over iron-working, so essential for
efficient agriculture, provided the leverage by which iron smith clans elevated
themselves into ruling clans and established the state to safeguard their inter-
ests and to levy taxes on the rest of the society (Kimambo and Temu, pp. 15-17).
Elsewhere control over cattle which is a more convenient form of storing wealth
than vegetable foodstuffs and whose manure is so important for agricultural in-
tensification, may have provided the basis for the establishment of a stratified
political organization in which the cattlemen were the rulers, who extracted
tribute in kind or in labor.16 Such surplus-appropriating ruling classes have
an active interest in pushing through technical changes to increase agricultural
productivity when increasing population threatens to eat into their share of the
surplus (taxes and tribute), or merely to increase the size of the surplus that
they can appropriate. In both cases agricultural change will occur long before
population growth reaches the ceiling of the carrying capacity of the land at
subsistence levels. The size and reliability of the surplus, of course, will
influence the level of development of the society. If the surplus is likely to
fail often as a result of an unfavorable environment, the level of economic
development and non-agricultural specialization is likely to be low, and the ruling class
is likely to be unstable and weak. At the same time, the size of the surplus is
likely to be influenced by the size of the population and its level of technology.

It is obvious from the above discussion that Boserup's formula is hardly a

formula of development of the productive forces of the society. According to her
theory of agricultural and other changes under population pressure, we should ex-
pect the change to occur when the upper limit of the carrying capacity of the land
is reached. And yet, this means that as that limit is approached the standard of
living of the people is likely to decline, and every stage of economic and socio-
political development is likely to be a period of grave crisis, hardly opportune
moments for major socio-economic change to a higher level which would require a
larger surplus to support new techniques and new specialists. Agricultural change
under population pressure implies a decline of the surplus. Unless these classes
have already become entirely unproductive parasites, i.e., they no longer discharge
their social functions, the increase in the amount of the surplus produced is
likely to lead to a further development of the productive forces of the society.

In view of this explanation of social change in pre-industrial societies, we

can now explain the lack of congruence between population densities and land use
systems, and between land use systemi and social and political systems. The former
is easier to explain in that the social scientists concerned simply failed to take
into consideration the social context in which these changes occurred. As to the
contrast between Bemba political organization and economic system, the explaunation
lies in the success of the Bemba ruling class in forcing the production of surplus
which is potentially large in an area where the cultivators need to work only 3-4
hours a day to meet their subsistence needs. (Richards, p. 394)

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 13

Ukara, on the other hand, seems to provide a case in which a technological

barrier appears to have prevented the development of productive forces to produce
adequate surplus for the support of non-agricultural specialist classes. The Kara
agricultural system has impressed every observer for having employed almost every
technique conceivable under pre-industrial conditions to increase productivity of
the land, and agronomists appear to be at a loss to suggest any further improve-
ment except the use of chemical fertilizers. The Kara cannot produce more cattle
manure because of the shortage of fodder on the Island. Moreover, if make as
well as kind of tool makes a difference to the efficiency of agriculture, the Kara
may have suffered from shortage of iron on the island, for they appear to have
used wooden hoes apparently well into the 20th century. The Kara have also reached
the limit of the amount of labor they can invest for they already work an average
of 12 hours a day. Thus a technological barrier appears to have been reached.
According to Ludwig, Ukara was "unusually densely settled" as early as the 17th
century (in Ruthenberg p. 91). An oral source records their outmigration on a
regular basis during the 19th century to Ukerewe where they worked as herdsmen
and agricultural labourers in return for a head of cattle or milk, or "a handful
of millet" which they sent home while continuing to work. This strongly suggests
that at least by the 19th century Kara population growth had indeed begun to reach
the limit of the carrying capacity of the island, and the amount of surplus pro-
duced was too small to support many non-agricultural specialists or a ruling
class. There was, apparently, no division of labor between agriculture and
fishing, and there are no references to any kind of productive non-agricultural
specialists. There is an isolated reference to some "rain-magicians" but, as
has already been mentioned, there were no chiefs or headmen before the advent of
Europeans.

There are two possible interpretations of this state of affairs. The first
would be closer to the first part of Boserup's theory of agricultural intensifi-
cation under population pressure, the changes occurring precisely when the upper
limit of the carrying capacity of the land is reached. This means that there was
never significant production of surplus to support any specialist classes which
of course, is contrary to Boserup's theory taken as a whole. In other words,
Ukara has, for a long time, stagnated in terms of the development of productive
forces, always fighting "a valiant rearguard action, with all the weapons at their
disposal" to maintain bare subsistence standards despite remarkable improvement
in agricultural techniques.
The other possibility is that the Island has undergone a regression in
socio-political organization. It is possible that before the island was so heavily
populated the area was capable of producing a surplus perhaps similar to that in
Sukumaland under extensive agricultural systems which may have been adequate to
support Sukuma-type non-agricultural specialist clases. As population increased
and the amount of land was limited, agriculture was gradually intensified until
the technological barrier was reached, at which point any further increase in the
population had to be fed out of the surplus hitherto appropriated by the non-
agricultural specialist classes. These classes could therefore have been starved
out of existence under these conditions. Unfortunately we lack historical evidence
to determine which of the possible interpretations is correct since this technolo-
gical barrier was apparently reached a long time ago, and no detailed historical
research has yet been undertaken on the island.
Abdul H. Sheriff
University of Dar es Salaam
for 1974-75: 19B University Houses
Madison, Wisconsin 53705

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions
 14

1. P.F.M. McLoughlin, African Food Production Systems, Baltimore, 1970, pp. 16-17.

2. J.K Nyerere, Freedom and Unity, London, 1967, pp. 166, 170. J.F. Rweyman,
"Tanzania: Economy and Society in the Pre-Independence Period," p. 2.
S.H. Hymer, "Economic Forms in Pre-Colonial Ghana", J. Econ. Hist.,
XXXI, 1 (March 1970), p. 33.

3. n.P. Miracle, Agriculture in the Congo Basin, Madison, 1967, p. 33.

4. M.P. Miracle, "Plateau Tonga Entrepreneurs in Historical Inter-Regional Trade",

Rhodes-Livingstone Journal, 26 (1959), 34-50. "Aboriginal Trade Among
the Senga and Nsonga of N. Rhodesia," Ethnology, 1 (1962), 212-222.

5. R. Mukherjee, The Problem of Uganda: A Study in Acculturation, Berlin, 1956,

Ch. III. I.N. Kimambo and A.J. Temu (eds.), A History of Tanzania, Nai-
robi, 1969, 14-33. J. Halpern, "Traditional Economy in West Africa,"
Africana Bulletin, 7 (1967), p. 101.

6. E. Boserup, The Conditions of Agricultural growth: The Economics of Agrarian

change under Population Pressure, London, 1965.

7. Miracle, 1970, p. 40. A.I. Richards, Land, Labor and Diet in Northern Rhodesia,
London, 1933, p. 288.

8. J. Iliffe, Agricultural change in modern Tanganyika. Nairobi, 1971, p. 7.

9. W. Allan, The African Husbandman, Edinburgh, 1965, pp. 212-213.

10. H. Ruthenberg, (ed.), Smallholder Farming and Smallholder Development in Tan-
zania, Munchen, 1968, pp. 72, 89-133. D.W. Malcolm, "Ukara: A Report on
Economic Conditions in the Island of Ukara with special reference to soil
erosion and other matters influencing production," Typescript, Mwanza, 1934,
N.W. Rounce and D. Thornton, Ukara Island and the Agricultural Practices
of the Wakama, Nairobi, 1944.

11. P. Hill, Rural House: A Village and a setting, London, 1972, p. 21.
12. B.A. Datoo, "Relationship between population density and agricultural systems
in the tropics: a pilot investigation in the Uluguru mountains, Tanzania,"
typescript, p. 61.

13. Ukerewe District Book, extracts from R.C. Thurnwald, Black and White in East
Africa, 1935, and from Mwanza District Book, Microfilm.
14. Mukherjee, pp. 49-51. A.C. Coulson, "A Simplified Political Economy of Tan-
zania," p. 3.

15. F. Engels, Anti-Duhring, Moscow, 1969, 215.

16. Mukherjee, Ch.III. I.K. Katoke, The Making of the Karagwe Kingdom, Nairobi,
1970, p. 20.

17. D.W. Malcolm, Sukumaland, London, 1953, pp. 136-7. "The Coming of the Wasi-
langa chiefs," 7-8. Ukereve District Book. Boserup, 27, says the make of
took makes no difference.

This content downloaded from 132.203.227.61 on Thu, 29 Oct 2015 16:25:13 UTC
All use subject to JSTOR Terms and Conditions