World Development Vol. 38, No. 9, pp.

1251–1262, 2010
Ó 2010 Elsevier Ltd. All rights reserved
0305-750X/$ - see front matter
www.elsevier.com/locate/worlddev
doi:10.1016/j.worlddev.2010.02.007

“Ancient and Backward or Long-Lived and Sustainable?” The Role

of the Past in Debates Concerning Rural Livelihoods and
Resource Conservation in Eastern Africa
DARYL STUMP *
University of York, UK
Summary. — Attempts by external agencies to intervene in the operation of local resource exploitation strategies frequently include ref-
erence to historical arguments. These vary in accuracy and sophistication but are nevertheless rhetorically useful since discussions of
economic or environmental sustainability or degradation are substantially strengthened by historical comparisons and precedents.
Focussing on examples of indigenous intensive agriculture in eastern Africa, this paper agues that relevant evidence of this sort is often
unavailable or far from unambiguous. It is therefore necessary to be critical of the ways in which perceptions of the past are invoked
within these discourses, and to be aware of the strengths and weaknesses of historical arguments in this regard.
Ó 2010 Elsevier Ltd. All rights reserved.

Key words — historical ecology, archeology, intensive agriculture, indigenous knowledge, east Africa

1. INTRODUCTION systems in the past. The histories of specific economies or envi-

Many debates concerning rural development and conserva-
tion in Africa and elsewhere have focussed not on the present
or future—as one might expect—but on the past. Conserva-
tionists may present landscapes as virgin territory unaffected
by local populations (for reviews and critiques of which see,
for example, Anderson & Grove, 1987; Fairhead & Leach,
1996); proponents of the extension or reapplication of indige-
nous knowledge tend to emphasize the longevity and environ-
mental sustainability of local cultivation techniques (e.g.,
Alteiri & Koohafkan, 2008; Harrop, 2007); supporters of
modernization may highlight examples of local environmental
degradation or stress the inadequacies of indigenous tech-
niques in the face of perceived social, economic, or environ-
mental crises (see, e.g., Rocheleau, Steinberg, & Benjamin,
1995; Lambin et al., 2001); and proposed interventions may
rely on models of ecological or social change that are implic-
itly historical or which are themselves extrapolated from his-
torical case-studies (see, e.g., Boserup, 1981; Siedenburg,
2006). As much of this research attests there is nothing new
about the invocation of the past in such debates, but historical
arguments remain of pivotal importance and indeed are now
discussed from a range of disciplinary standpoints including
ethnobiology/ethnobotany (e.g., Sillitoe, 2006), historical ecol-
ogy (e.g., Balée, 2006), indigenous 1knowledge (e.g., Briggs,
2005), resilience theory (e.g., Constanza et al., 2007), applied
archeology (e.g., Erickson, 2006), and through the consider-
ation of the role to be played by property rights legislation;
the latter encompassing historical perspectives owing to the
need to identify where and by whom particular technologies,
lands or genetic resources were first employed (e.g., Brush,
2007; Sillar, 2005).
This referencing of the past within such discussions is argu-
ably inevitable, since any appeal to the twin concepts of sus-
tainability and conservation invites comparisons between the
contemporary and historical situations on the grounds that
both refer to the need to balance short-term gains with long-
term resource maintenance. Since the future is unknown, the
sustainability of a system must be predicted either through
experimentation or by reference to the behavior of similar
1251
ronments may therefore be employed as potential sources of
positive or negative precedents on the grounds that if sustain-
ability can be demonstrated in the past, then this at least sug-
gests the possibility of future sustainability. Conversely, if a
mode of resource use can be shown to have caused environ-
mental depletion in the past, or can be likened to a past com-
munity that collapsed or suffered as a result of similar
practices, then this would seem to indicate that the system in
question is unsustainable. This is putting the argument
crudely, perhaps, but examples of these positions persist,
and will be illustrated here through the general case-study of
indigenous irrigation- and terrace-using agronomies in eastern
Africa.
Given that few if any environments can be conclusively de-
scribed as unmodified by previous periods of human exploita-
tion, other economic and land-use practices from Africa or
elsewhere could be employed to examine this relationship be-
tween sustainability debates and historical knowledge (e.g.,
Fairhead & Leach, 1996; Shetler, 2007). Nevertheless, this pa-
per concentrates on east African indigenous irrigation and ter-
race systems for several reasons. First, a number of these
agronomies have themselves been cited as potential paradigms
of sustainable development, but have also been denigrated as
examples of economies on the brink of collapse through re-
source degradation. They thus serve to illustrate the ways in
which historical arguments are marshalled to justify interven-
tion in the name of economic or environmental stewardship.
Second, related models of increasing agricultural intensifica-
tion proposed by Malthus (1985 (1798)) and Boserup (1965)
have proved extremely influential to developmental thinking
* This paper builds on work undertaken as part of a studentship funded by
the AHRB, though the issues discussed have been explored further as an
aspect the EU-funded Historical Ecologies of East African Landscapes
project (HEEAL). The discussion has benefited from comments and cri-
tiques by numerous people, for which particular thanks are due to Bill
Adams, Maureen Bennell, Paul Lane and Federica Sulas. The insightful
comments made by four anonymous reviewers are also gratefully ackno-
wledged. Final revision accepted: January 25, 2010.
1252 WORLD DEVELOPMENT
and have been applied to these agricultural systems. However,
the appropriateness of these models to specific African rural
economies has been questioned by what are in effect historical
critiques (e.g., Börjeson, 2007; Conelly, 1994) and, indeed, the
schema presented by Boserup is itself partly defended by re-
course to historical and archeological case-studies (Boserup,
1981; see also Netting, 1993). Third, although no similar pro-
jects have been undertaken in Africa, attempts to employ
archeological data to define and rehabilitate abandoned indig-
enous resource strategies elsewhere in the world have under-
standably focussed on structures with very high
archeological visibility such as agricultural terraces and irriga-
tion features (e.g., Erickson, 1998 (1992); Kendall, 2005). The
experiences of these projects thus aid in an appraisal of the fea-
sibility of this approach in eastern Africa, but are primarily
referenced here to highlight to a non-archeological audience
how these historical interpretations are constructed, and the
implications of this as regards the use of historical narratives
in assessments of sustainability. Fourth, since several of these
agronomies have their origins in the precolonial period but
persist in modified forms up to the present, they offer oppor-
tunities to explore the strengths and weaknesses of paleo-envi-
ronmental, archeological, historical and observation-based
sources of evidence. Indeed, lack of historical detail regarding
the operation of these systems well into the 20th century
means they also act as useful examples of how archeological
techniques may inform reconstructions of the relatively recent
past, and highlight the degree to which archeological interpre-
tations rely on analogies drawn from historical and observa-
tional sources. Finally, given that the histories of these
African agronomies remain poorly understood, they also serve
to emphasize how historical arguments are employed even in
the absence of good historical data.
It should perhaps be stated at the outset, however, that the
focus of this paper is not on the deliberate misrepresentation
of local history by those seeking to justify intervention or to
secure funds (see, e.g., Ferguson 1990). Rather, the chief con-
cern is the use of implicit or explicit historical arguments with-
in conservationist and developmental narratives pertaining to
these agronomies, with such arguments ranging from asser-
tions that aspects of local resource-use should be encouraged
or exported, to conclusions that particular practices have
had detrimental consequences in the past. Although the
emphasis here is on the precolonial past this issue applies
equally to studies of late-20th century change, since in all such
arguments the relationship between historical data and pre-
dicted future response is an act of interpretation and is there-
fore never unproblematic. Studies employing very different
sources or evidence from a variety of disciplinary backgrounds
may therefore reach very similar conclusions when attempting
to translate this data into policy objectives. In short, while this
paper makes no pretence to being a comprehensive review of
either the historical or developmental literature pertaining to
these economies, the brief overview offered here is presented
to stress the need to be critical of the ways in which percep-
tions of the past are applied within broader debates.
2. DEGRADATION, SUSTAINABILITY, RESILIENCE,
AND THE RHETORICAL USE OF THE PAST
The precolonial irrigation- and terrace-using agronomies of
eastern Africa defy simple models of their origins based on
environmental determinism or population-push intensification
having developed in a range of ecological and social environ-
ments at various times over at least the last 600 years. Re-
cently referred to as “islands” of intensive agriculture
(Widgren & Sutton, 2004), these systems include extensive
areas of agricultural terraces in the Harar and Burji-Konso
areas of Ethiopia, in the Kigezi district of Uganda, in the Pare
mountains of Tanzania and on the islands of Ukara, Rushinga
and Mafangano in Victoria Nyanza, while extended networks
of irrigation and drainage features include those at Konso in
Ethiopia, at Sonjo, Pare, Meru, Kilimanjaro and Usambara
in Tanzania, and at Marakwet and Pokot in the Kerio Valley,
Kenya (for gazetteers see Anderson & Adams, 1988; Grove &
Sutton, 1989), see Figure 1. As noted, the ability of these
agronomies to support large rural populations and the use
of apparently low-cost soil and water conservation techniques
have led to the suggestion that they might act as paradigms for
small-scale, low external input, sustainable resource use as
promoted by the concept of “development from below” (Stöhr
& Taylor, 1981) and by the United Nation’s Agenda 21 (UN,
1992, reaffirmed 2002). There is also the perceived added
advantage that several of these areas employ “zero” tillage
agriculture as endorsed by contemporary conservationist pol-
icies, and exist within—and thus may have helped maintain—
the Eastern Afromontane “biodiversity hotspot” (Brooks
et al., 2002). The irrigation system at Marakwet, for example,
has been described as potentially paradigmatic on the basis of
its “non-bureaucratic management” (Ssennyonga,1983, p. 96;
see also Moore, 1983, p. 132), while the use of stone terraces
and water harvesting structures at Konso were described as
one of a select few “lessons from the past” by a UN scheme
for land conservation and rehabilitation in African (FAO,
1990). 2 Similarly, the precolonial irrigation structures in
North Pare have been the focus of a series of rehabilitation
programmes from the early 1990s, first as a joint project by
the Dutch and German development agencies SNV 3 and
GTZ 4 (Sheridan, 2002) and now as part of ongoing initiatives
by the local agencies TIP 5 and PADEP 6. Similar projects have
been attempted or advocated for South Pare (Yoshida, 1985),
Usambara (Kaswamila & Masuruli, 2004; Sokoni & She-
chambo, 2005) and Kilimanjaro (Grove, 1993; Lein, 1998) in
Tanzania. In addition, the existence of forested areas within
or alongside the agricultural landscapes in both Pare and
Usambara has been cited as evidence of the existence and effi-
cacy of “indigenous conservation” (Gillson, Sheridan, & Broc-
kington, 2003; Stocking & Perkin, 1992).
Whether implicitly or explicitly, these endorsements of local
practices employ perceptions of environmental and economic
history despite the fact that in the vast majority of cases the
histories of these areas remain fragmentary at best. As was
touched upon above, this apparent paradox results from a ser-
ies of interrelated arguments that require or could be substan-
tially strengthened by the inclusion of an historical dimension.
These take the form of direct and sometimes simplistic com-
parisons between the history of indigenous cultivation and
the history of previous interventions, or between the history
of African agriculture and the agricultural history of the devel-
oped world. Arguing for the potential sustainability of local
practices by contrasting these with the detrimental social, eco-
nomic or environmental consequences of previous attempts at
modernization can certainly be seen in this light (e.g., and dis-
cussions see Adams & Anderson, 1988, pp. 529–530; Adams &
Grove, 1984; Chambers, 1973), as can research which demon-
strates that the Malthusian crises predicted by earlier assess-
ments of local agriculture have yet to take place (e.g.,
Carswell, 2007; Rocheleau et al., 1995; Tiffen, Mortimore, &
Gichuki, 1994). Studies that question whether local environ-
ments are as pristine or as historically stable as imagined with-
in degradation narratives would also fall into this category
 “ANCIENT AND BACKWARD OR LONG-LIVED AND SUSTAINABLE?”
(e.g., Boerma, 2006; Gillson et al., 2003; Leach & Mearns,
1996).
In terms of the agronomies discussed here, however, the
prevalence of the use of historical arguments in discussions
of their future can also be attributed to high levels of land-
scape modification. While it may be superficially plausible to
present open grasslands and forested areas as “unoccupied,”
“virgin” or “natural”—that is as being without history—the
same clearly cannot be said for large areas of agricultural ter-
racing or for fields served by complex systems of irrigation
channels. The crude rhetorical device of denying a local his-
tory of resource use to justify the colonial appropriation of
land or to validate external intervention (e.g., Anderson &
Grove, 1987; Shetler, 2007) was thus never applied to these
areas, since the sheer physical presence of highly modified
landscapes appears to attest to a long history of cultivation.
At their simplest, then, the use of historical arguments by pro-
ponents of indigenous knowledge might merely represent the
perception of a direct relationship between longevity of occu-
pation and the sustainability of local techniques; an approach
that is apparently reflected in the UN’s stance that indigenous
groups “have developed over many generations a holistic tra-
ditional scientific knowledge of their lands, natural resources
and environment” (UN, 1992, p. 26.1).
Leaving aside for the moment the problematic historical
assumptions contained within such statements, it is notewor-
thy that this line of argumentation is essentially the same as
that employed in colonial period accounts and policies which
acknowledged the potential longevity of indigenous agrono-
mies but which equated antiquity with a lack of recent techni-
cal progression. Note, for example, that attempts to improve
and rationalize the 250 km of main irrigation channels in Mar-
akwet, Kenya, from the 1930s onwards included overt refer-
ences to a perceived history, such as those by a Provisional
Commissioner in 1933 who reported that “so far as I can judge
the Marakwet have not advanced one whit [. . .] since the time
when their more energetic and enterprising ancestors dug the
furrows that are in use today” (cited by Adams, 1996, p.
160; see also Anderson, 2002, p. 272 in reference to similar
comments regarding Baringo, Kenya). At this highly rhetori-
cal level, then, allusion to the past is used to demonstrate a
lack of modernity and hence a need for modernization. Yet
in very general terms, the concept of indigenous knowledge
as a developmental trope may work at a similar level of
abstraction by viewing any evidence of prolonged occupation
as proof of sustainability.
It is, of course, something of a caricature to suggest that all
colonial period narratives saw these agronomies as ancient
and backward or that all proponents of indigenous knowledge
view them as long-lived and sustainable. Indeed, the potential
benefits of aspects of indigenous soil conservation techniques
were seriously explored by British colonial authorities in West,
East and Southern Africa in the 1920s and 1930s, albeit with
differing levels of commitment and with various concessions
to local economics and politics, particular as regards the inter-
ests of white settlers (for discussion of which see Anderson,
1984; Beinart, 1984; Richards, 1985). This dichotomy may
thus be something of a straw man. Nevertheless, the point to
be emphasized here is not simply that these competing narra-
tives produce opposing conclusions, but rather that they do so
by reference to unsubstantiated perceptions of the past. More-
over, there are clear echoes of this approach in the application
of neo-Malthusian and Boserupian models within develop-
mental narratives, not least because both of these related
schemes appear to provide an unambiguous trajectory for
agricultural change and a definition of the conditions under
1253
which this change will occur. In consequence—and without
wishing to include a lengthy outline or critique of these models
here—there is a danger that they will be used not just predica-
tively but as substitutes for historical data on the grounds that
the reactions of any given community to specific stimuli are as-
sumed to be well understood (a shortcut to historical interpre-
tation that has also proved tempting to historians and
archeologists, see Morrison, 1996).
While it is recognized that these models lend themselves to
the production of crisis narratives and that such narratives
may in some cases be consciously employed to raise awareness
and funds (e.g., Rocheleau et al., 1995), to do so is to risk hav-
ing one’s broader case undermined by historical critiques. Ho-
ben (1996), for example, highlights the prevalence of neo-
Malthusian arguments in debates surrounding large-scale ter-
race building projects in Ethiopia from the 1980s onward, but
also points out that these narratives are supported by reference
to a vague and unsubstantiated past, itself “a period seldom
defined more precisely than “before the present century””
(ibid., p. 196). Note too that although advocates of indigenous
knowledge are apt to view terraces and irrigation features as
examples of indigenous soil and water conservation (e.g., Reij,
Scoones, & Toulmin, 1996) and thus as “investments” in
“landesque capital” (e.g Widgren, 2007, pp. 62–63), the mod-
els proposed by Malthus (1985) and Boserup (1965) both
anticipate that a community will only develop labor intensive
technologies when prompted by stress, and both expect this to
happen only at a relatively late stage in the development of
intensive agriculture. The same kinds of agricultural features
are thus viewed as evidenced of either good or bad land hus-
bandry: as either positive or negative precedents. Deciding be-
tween these two extremes for specific instances therefore
requires detailed historical evidence.
In essence, therefore, the position here is that current defi-
ciencies in historical data make invocations of the past in sup-
port of particular policies or paradigms somewhat rhetorical,
yet this is rarely the impression received from the existing lit-
erature. What, then, is the way forward? One approach, of
course, would be to initiate targeted research that aims to ad-
dress historical questions raised by developmental or conser-
vationist debates. Demands for work of this kind range
from the very specific to the extremely complex: an example
of the former including the call by Gillson et al. (2003, p.
384) for archeological and paleo-environmental research to
test hypotheses regarding local forest management in Pare,
Tanzania, while the latter is best represented by the ambition
among resilience theorists to produce long-term historical
models which recognize the level of interaction between social
and ecological systems and that can account for non-equlibrial
dynamics and for social and environmental disjunctions (e.g.,
Holling, Berkes, & Folke, 1998; Redman & Kinzig, 2003;
Constanza et al., 2007).
However, even advocates of resilience theory acknowledge
that this is more an aspiration than an immediate goal, and
note that at present the problem of how to account for cultur-
ally specific and historically contingent human actions means
that these tend to be equated with anomalies highlighted by
the modelling process (Dearing, 2007, p. 51; Oldfield, 2007).
While these and other writers thus argue that well understood
historical case-studies will form a pivotal role in testing resil-
ience models, they accept too that in some instances this nec-
essary historical detail may be indiscernible at the scale
required (as has been suggested by Brockington and Home-
wood (2001, p. 475) regarding the role of pastoralist produc-
tion in shaping the ecology of Mkomazi Game Reserve,
Tanzania), and that subsequent changes to the social, political
1254 WORLD DEVELOPMENT
or ecological environment may be sufficient to make compar-
isons between the historical and modern situation problematic
(a view espoused by McCann (1990) in reference to a brief per-
iod of agricultural intensification in early 20th century Ethio-
pia). Widgren’s (1999, p. 6) initial stated aim of attempting to
define a “common explanatory framework” for why some
African intensive agronomies appear resilient to ecological,
political, or economic shifts while others are abandoned is
thus extremely ambitious, and is certainly complicated by lack
of good historical data for many of the potential case-studies
(Widgren, 2004, p. 11). Until such time as these histories are
better understood, therefore, it will be necessary to remain
wary of any claim that the history of a particular African
agronomy can act as a source of positive or negative prece-
dents, and indeed even the apparently innocuous practice of
including an historical background to provide pre-disturbance
baselines needs to be treated with caution.
3. THE PAST AS BACKGROUND
It is a common if not ubiquitous practice to include a sum-
mary of the history of an area as an introduction to a given
case-study or environmental impact assessment. On the face
of it such summaries are benign, but become problematic
where poorly understood or ambiguous histories are presented
as well-established facts. Furthermore, whether or not these
sections are intended to be read as offering accurate socio-eco-
nomic or ecological baselines for the analyses that follow, they
nevertheless risk being seen as providing precedents and are
therefore employed either implicitly or explicitly in support
of the report’s findings.
In what is ostensibly an examination of factors affecting that
adoption of “improved” traditional soil conservation tech-
niques in the North Pare and West Usambara mountains,
for example, Mbaga-Semgalawe and Folmer (2000, pp. 222–
223) present a short history of agriculture in the highlands
of northeastern Tanzania; an area that encompasses Kilimanj-
aro, Meru, Arusha, North and South Pare, and the East and
West Usambaras. To do so is inherently to generalize a com-
plex and highly varied history. Indeed, despite advocating the
use of what they describe as traditional soil and water conser-
vation methods (including furrow irrigation, terracing, zero-
grazing, minimum tillage, intercropping and mulching) Mbag-
a-Semgalawe and Folmer imply that all of these techniques
date to the early 20th century (2000, p. 322); assert that “In
the beginning of the 18th century most of the northeastern
mountains of Tanzania were covered in natural forests”
(ibid.); state that fallowing became impossible because “By
1936 all arable land was under cultivation” (ibid.); and claim
that as a result “most of the soil cover was removed, rivers
and springs dried up and land productivity started to decrease
due to soil degradation” (ibid.). Every one of these statements
is contentious, not least because they deny the existence of a
complex precolonial history of cultivation in these areas. Un-
der such a reading local practices are therefore neither ancient
and backward nor long-lived and sustainable: they are recent,
crisis driven and, moreover, unable to cope with a worsening
situation without external intervention. The intention in offer-
ing this summary is thus presumably to establish the historical
background to the problems they seek to explore and redress,
yet the result is to weaken their case through the inclusion of
detail that is at best uncertain and at worst inaccurate. At the
risk of doing the same by countering this position with an
overly simplistic summary, it is instructive to outline briefly
the extent of some of these errors.
From an historical standpoint, the uncorroborated and
unusually precise dates cited for several processes are cause
for concern. To take Kilimanjaro as an example, although
land shortages and soil erosion were both sources of unease
by the 1930s, fallowing, swidden cultivation, the use of soil
conservation techniques and the clearance of formerly unculti-
vated land for agriculture, were all apparently practiced at this
time (Curry, 1939; Swynnerton, 1949). Similarly, far from
being seen as solely a problem stemming from the local Afri-
can population, anxieties about increasing pressures on land
and water resources were generally expressed in reference to
competition between African farmers in the highlands and
European owned farms and industry in the foothills and
surrounding plains (Kanthack, 1938, p. 16; Lein, 1998; Swynn-
erton, 1949, p. 122). Moreover, in contrast to Mbaga-Semgal-
awe and Folmer’s suggestion that indigenous agricultural
technologies such as irrigation postdate the foundation of
colonial rule, various early European observers not only
reported the existence and potential antiquity of the network
of water channels on Kilimanjaro, but also commented
favourable on its extent and technical assurance (e.g.,
Johnston, 1886, p. 120; New, 1873, p. 370; Swynnerton,
1949, p. 124). Indeed, despite citing Kimambo (1991) in refer-
ence to protests against the colonial imposition of bench ter-
racing in Pare, this summary ignores Kimambo’s (1991, pp.
20, 22) references to precolonial terracing and irrigation in this
area; the latter practice undertaken from at least the 18th cen-
tury and possibly for up to two centuries earlier (see also
Håkansson, 1995, pp. 303–304; Kimambo, 1969, p. 80;
Sheridan, 2002). Parallel cases can be made for the technolog-
ically similar and possibly related irrigation systems employed
in several of the precolonial polities on the southern slopes of
Kilimanjaro (Devenne, 2006; Masao, 1974; Stahl, 1964, pp.
26, 211–212, 301; Tagseth, 2006, p. 495), on Mount Meru
(Spear, 1997, pp. 17, 25), and within and around the Usambar-
a Mountains (Feierman, 1990, p. 65; Koponen, 1988, pp. 232–
233 citing Holst, 1893).
Taken together, then, the use of soil and water conservation
techniques has a far longer history in highland northern Tan-
zania than the century or so implied by Mbaga-Semgalawe
and Folmer (2000). Yet, as noted previously, this historical
evidence presently lacks the precision to allow simplistic corre-
lations between apparent longevity of cultivation and agricul-
tural sustainability. Indeed, in the case of North Pare, the
suggestion that soil and water conservation techniques exist
alongside examples of indigenous forest conservation has been
challenged by studies which query the assumption that extant
sacred groves represent relics of precolonial forests, chiefly on
the basis of records of colonial tree-planting schemes and by
reference to early European observations of a largely treeless
landscape (Gillson et al., 2003; Håkansson, 1995, citing Bau-
mann, 1891). The area’s reputation as a centre for iron pro-
duction in the late precolonial period further suggests at
least the possibility that deforestation resulted from local
extraction of fuel wood (Gillson et al., 2003, pp. 373–376;
Håkansson, 2007, citing Kersten, 1969, II, p. 19; Meyer,
1893, p. 223; Sheridan, 2002, p. 83, citing von der Decken,
1978 (1871), p. 17), and invites questions regarding the possi-
bility that this process was combined with the clearance of for-
ests for agriculture.
Criticizing individual authors within the developmental lit-
erature for a lack of historical precision is, of course, unfair,
and it is perhaps worth noting that an historian is unlikely
to be censured for failing to provide recommendations
for the future of the area under study. Nevertheless, at some
level the historical detail included in reports of this type
 “ANCIENT AND BACKWARD OR LONG-LIVED AND SUSTAINABLE?”
influences the conclusions drawn by either the authors or read-
ers; most obviously in this case by the suggestion of an ongo-
ing process of extreme environmental degradation over a very
wide area and dating back some 70 years. Although still far
from fully understood, a more nuanced understanding of the
history of these areas indicates—perhaps inevitably—that
the past is considerably more complex, and thus does not al-
low the formation of simplistic equations of cause and effect,
problem and solution.
4. THE PAST AS NEGATIVE PRECEDENT
If caution therefore needs to be exercised in the formation of
simple formulae regarding systemic sustainability, then the
opposite—equating abandonment with systemic failure—
should also be avoided. Writing from an archeological per-
spective Barker and Gilbertson (2000, p. 4) question the
assumption that “people sowed the seeds of their own destruc-
tion [. . .] by developing irrigation systems that caused saliniza-
tion, or by stripping the landscape for fuelwood, or by
allowing their livestock to over-graze the vegetation,” and
go on to observe that “In general, the debate has been charac-
terized more by confident assertions than well-founded argu-
ment” (ibid.).
In terms of east African agriculture such assertions inevita-
bly focus on the site of Engaruka, this being the largest irri-
gated and terraced agronomy to have been comprehensively
abandoned prior to colonial contact, with calibrated radiocar-
bon dates indicating that the area was farmed between the
14th and 18th centuries AD (Laulumaa, pers. comm.; Sutton,
1998). In the introductory section to an environmental history
that explores changing attitudes to intensive cultivation and
forest management in the Usambara Mountains during the
20th century, Conte (2004, p. 25) thus briefly refers to this site,
concluding that:
The more sophisticated and specialized the system of terraces
and furrows became [. . .] the greater the potential for soil
exhaustion, and the more precarious its owner’s subsistence
situation. Ultimately, the Engaruka works proved unsustain-
able and their failure forced its inhabitants to migrate.
Although it should be stressed that this interpretation is not
untenable and is made with reference to the archeological sur-
veys and discussions of the site conducted by Sutton (1984,
1989), Conte’s conclusion nevertheless serves as an example
of an historical case-study simplified to the point of inaccu-
racy, and has echoes of Koponen’s (1988, p. 383) assessment
that the agronomy “ended in a complete cul-de-sac”; the
implication within the wider context of Koponen’s analysis
being that Engaruka offers a rare example of the local mis-
management of resources in the precolonial period. In fact,
whilst anthropogenic resource depletion remains a possibil-
ity—specifically as regards a hypothesis of diminishing river
levels resulting from river catchment deforestation (Sutton,
1998, pp. 24, 34)—the assertion that locally induced resource
degradation led to forced desertion must be regarded as pre-
mature. Soil exhaustion or the accumulation of salts or alkalis
as the result of prolonged periods of irrigation now seem
doubtful as causes of abandonment (Stump, 2003, 2006), with
the most likely explanation being declines in river flows as is
illustrated by the presence of irrigation features fed from what
are now permanently dry river beds (Sutton, 1998, p. 11;
Stump, 2006). In contrast to the view that the visible remains
of this irrigated landscape resulted from the short-sighted
1255
extraction of available resources, it is now clear that the agron-
omy developed gradually over a period of perhaps four centu-
ries (Stump, 2006) and that it persisted during a comparatively
dry period that spanned the early 16th to mid-17th centuries
(Westerberg et al., 2007).
Obviously, these authors cannot be criticized for not refer-
encing this more recent work, but from the perspective of the
current discussion it is noteworthy that the summaries offered
by both Conte and Koponen remove Sutton’s numerous cave-
ats regarding uncertainty and the need to test various possible
hypotheses. Doing so allows the presentation of definitive con-
clusions that appear to add both temporal depth and balance
to their respective positions; a point that is particularly relevant
to Koponen’s discussion given that his stated aim is to explore
the veracity of Kjekshus’s (1977) thesis that precolonial com-
munities in what is now Tanzania were capable of production
far above subsistence level and were generally well adapted to
their environments. Once again, then, it is necessary to be crit-
ical of the ways in which invocations or perceptions of the past
are employed within broader discussions and to be aware of the
processes through which these pasts are constructed.
Indeed, it is interesting to note that Engaruka is even men-
tioned in reference to overstocking degradation narratives at
the Mkomazi Game Reserve, Tanzania, with Brockington
and Homewood (2001, p. 459) citing correspondence between
a game warden and a local Member of Parliament dated 1966
in which it is stated that “In a short time this area will become
like the desert of Engaruka [. . .]. We will shortly face the pros-
pect of the River Umba drying up because the trees around the
river have been finished off and pushed over by cows.” Clearly
in this instance Engaruka’s past was not being cited as part of
a case-specific background history, but it was nevertheless in-
voked as a precedent of what may occur if processes perceived
as damaging to the Mkomazi area were allowed to continue
unchecked. Regardless of the historical accuracy of this corre-
lation between abandonment and mismanagement, the relict
field system at Engaruka remains an unusually visible appar-
ent “lesson from the past.” Watson (2007, pp. 220–222), how-
ever, cites a similar example of an alleged negative precedent
at a landscape scale, though in this instance the use of images
of severe soil erosion alongside descriptions of overexploita-
tion in an NGO’s promotional literature regarding Konso,
Ethiopia, would appear to be at best careless and a worst a
deliberate misrepresentation. As Watson points out, the de-
graded landscape lies outside the area currently farmed, and
is mentioned within the oral tradition of one linage of ritual
leaders as having been eroded at the time this linage was estab-
lished at some as yet undefined time in the distant past (ibid.,
85, p. 221). Given that local soil and water conservation tech-
niques at Konso have also been cited as a positive precedent
by the FAO (1990), this example illustrates how organizations
with similar objectives might select opposing historical narra-
tives to support the case being made.
Over the course of the 20th century this tacking between
admiration and denegation of local cultivation is perhaps most
obvious in reference to the agronomies on Kilimanjaro, Tan-
zania, and within the Kerio valley, Kenya; repeated external
intervention in the latter having been justified during the colo-
nial period by the apparently sincere beliefs of a series of Agri-
cultural Officers who saw the indigenous irrigation system as
technically impressive, but as wasteful of water, a contributor
to soil erosion, and as having undergone a period of decline
since its precolonial heyday (Adams, 1996). Indeed, it is inter-
esting to note that the poor record of these colonial interven-
tions and that of the Kerio Valley Development Authority
(from 1979 onwards) were subsequently viewed as negative
1256 WORLD DEVELOPMENT
precedents in their own right by the various historians, anthro-
pologists, geographers, archeologists and development special-
ists who contributed to Kipkorir, Soper, and Ssennyonga
(1983).
5. THE PAST AS POSITIVE PRECEDENT
As already discussed, several of the intensive agricultural
systems in eastern Africa with precolonial origins have been ci-
ted as potential paradigms of sustainable indigenous knowl-
edge despite the absence of long-term historical,
archeological or paleo-ecological data. Indeed, for the major-
ity of these agronomies the history of exploitation is uncertain
even for the late 19th and early 20th centuries, with reliable
dates for inception only available for the comparatively late
examples established during the 19th century at Arusha, Tan-
zania (Spear, 1997) and at Baringo, Kenya (Anderson, 2002).
At present, then, although historical research has been able to
date specific elements of these agronomies such as individual
irrigation schemes (e.g., Kodalo, 2000; Tagseth, 2006), this
in itself only indicates that the technology is long-lived, and
does not necessarily demonstrate continuity of function. The
irrigation system that later supported the cultivation of coffee
as a cash crop on the southern slopes of Kilimanjaro, for
example, originally developed at least two centuries earlier
as a means to supply domestic water and to irrigate finger mil-
let (Devenne, 2006; Stump & Tagseth, 2009; cf. Burra & van
den Heuvel, 1987). Holling’s notion of resilience as the capac-
ity to “absorb disturbance and reorganize while undergoing
change so as to still retain essentially the same function, struc-
ture, identity, and feedbacks” (Walker, Holling, Carpenter, &
Kinzig, 2004), thus faces the problem that defining these
dynamics may require an essentially subjective decision as to
what level of alteration constitutes the creation of a fundamen-
tally different system to the one that preceded it. A similar con-
cern arises from those readings of local history that see
refutations of earlier negative assessments as demonstrating
sustainability (see Figure 1).
This point too has been touched upon above in relation to
the tendency to compare the apparent longevity of local adap-
tations to the perceived failures of earlier attempts at interven-
tion; a relationship succinctly summarized by Adams and
Anderson (1988, p. 533) who note that “certain indigenous
sites [. . .] offer a model that has far outlasted any modern
development.” This relationship of course applies equally to
large-scale damming, drainage or relocation projects, but in
terms of the current discussion is most relevant to attempts
to intercede in the management of local economies, especially
where external involvement was partially justified by reference
to perceived evidence of locally induced environmental degra-
dation (e.g., Adams, 1996; Kanthack, 1938). Research that
empirically qualifies these earlier narratives or which high-
lights the political motivation or methodological assumptions
behind their conclusions may, therefore, overturn previous
assessments regarding systemic vulnerability. The east African
historical studies that come closest to demonstrating positive
presidents could thus be characterized as critiques of the appli-
cation of neo-Malthusian degradation narratives to particular
agronomies (Carswell, 2007; Tiffen et al., 1994), though similar
research that questions the applicability of Boserup’s (1965,
1981) model of agricultural change (e.g., Börjeson, 2007;
Conelly, 1994) are also of relevance given the former influence
of this model as regards technical modernization and the per-
ceived need for tenure reforms to promote private ownership
(Tiffen, 1996 citing Boserup, 1970). By demonstrating that ear-
lier predictions of immanent systemic collapse dating back
over half a century have as yet proven unfounded, such studies
suggests that the featured agronomies may act as positive
precedents of locally developed resource maintenance regimes
in the face of increasing population. Thus, comparisons be-
tween surveys carried out by colonial Agricultural Officers
and the recent re-examination of these same transects in Kige-
zi District, Uganda, show that fallow lengths—Boserup’s
ready-reckoner of the degree of intensification (1965, p. 28,
1981, p. 19)—have actually been increasing over the course
of the 20th century (Carswell, 2002). Governmental and
NGO assessments dating to as recently as 1994 which forecast
catastrophic soil exhaustion due to “poor farming practices”
(Uganda Government, 1994, cited by Carswell, 2002, p. 132)
therefore seem equally questionable.
This having been said, Carswell is wary of equating this
reappraisal with an assessment of future sustainability, since
these new data merely refute the time scales and carrying
capacities mentioned in earlier forecasts and do not necessarily
disprove the overall thesis that continued pressure will one day
cause severe problems. This is the conclusion drawn by Broc-
kington and Homewood (2001, p. 457) with reference to the
study by Tiffen et al. (1994) which demonstrates a five-fold in-
crease in population alongside evidence of decreasing erosion
levels in the Machakos area of Kenya, despite predictions of
irrevocable soil damage from the 1930s onwards. Moreover,
in the course of this argument it is further emphasized that his-
torical analyses of this sort are rarely in a position to compare
like with like; a point illustrated for the Kigezi and Machakos
studies through the recognition of contemporary economic
options such as out-migration and non-farm incomes that
were less readily available at the time of the initial forecasts
(Carswell, 2002, p. 138; Murton, 1999; see also Conelly,
1994). Indeed, these and other factors are contributing in
many small-scale rural societies to the erosion of customary
hierarchies and to the falling credibility of supernatural reper-
cussions for unauthorized resource use (e.g., Grove, 1993;
Sheridan, 2002), both of which mean that previously effective
techniques may prove difficult to replicate in the future. One
can criticize the political manipulation of degradation narra-
tives, therefore, but this does not in itself demonstrate the sus-
tainability of local technologies as regards possible future
threats, of which the most obvious are still population in-
creases and climate change. In short, just as demonstrating
that earlier predictions of environmental collapse have yet to
occur does not mean this situation will be avoided indefinitely,
so recognizing the political co-option of a particular narrative
cannot be viewed as proof that these conclusions are necessar-
ily unsound (Brockington & Homewood, 2001, pp. 449, 457).
Positive precedents are not only harder to identify within the
historic record, therefore, but they also do not automatically
follow from the rejection of former negative assessments.
6. APPLIED ARCHEOLOGY AND THE
ETHNOGRAPHIC PAST
The problem of how to translate past performance into pre-
dictions of future success represents a serious stumbling block
for any attempt to invoke historical data in support of con-
temporary policies, and is certainly not avoided by drawing
the more general conclusion that the level of dynamism and
complexity displayed within African agricultural history sug-
gests individuals will inevitably find workable solutions to cur-
rent or future problems provided they have sufficient freedom
to continue to experiment with and refine a range of
 “ANCIENT AND BACKWARD OR LONG-LIVED AND SUSTAINABLE?” 1257

Konso

K E N Y A
West Pokot
U G A N D A
Marakwet
Kigezi Elgeyo Baringo

Mfangano
Rushinga

Machakos
Ukara
ru
Engaruka
Me

Kilimanjaro
T A N Z A N I A Arusha Taita
N. Pare
Iraqw’ar Da’wa
S. Pare
Usambara

KEY: Furrow irrigation , Terracing , Terracing and irrigation ,

Drainage or flood farming , Drainage or flood farming with terracing
Figure 1. Intensive agricultural systems with precolonial origins in eastern Africa.

techniques, technologies and crops (Niemeijer, 1996, pp. 102–
103). Indeed, although Niemeijer (1996, p. 87) advocates what
he describes as “a new development paradigm” built around
“a diachronic approach that is firmly based on an understand-
ing of the past,” such a conclusion is so wide-ranging that it
could be made in the absence of historical data in that it is
effectively a statement that individuals in Africa generally cope
to some extent. The position thus says nothing about sustain-
ability or about relative standards of living, health or life
expectancy in the developed and developing worlds, and says
still less about these issues in the past.
Despite this fact, Niemeijer also endorses a more targeted
approach that takes account of particularistic and culturally
specific factors over the long-term such as long fallow regimes
and the maintenance of delayed reciprocal relationships with
neighboring communities. Such an stance has clear affinities
with the emphasis among historical ecologists on the history
of interaction between humans and environments within spe-
cific landscapes, and shows further similarities with historical
ecology as regards the recognition that former or ongoing re-
source-use practices can have both positive and negative con-
sequences for modern communities, for example, by leading to
increases as well as decreases in the nutrient levels of cultivated
soils (Niemeijer, 1996, p. 104; cf. Balée, 2006, pp. 84–86).
Moreover, by suggesting that this research should focus not
just on the technologies and techniques employed today but
on the processes by which these strategies were developed
through time, Niemeijer recognizes the significance of change
and adaptation as regards the identification of potential posi-
tive or negative precedents; a contention that receives support
from the example of how an appreciation of the gradual devel-
opment of the agronomy at Engaruka casts doubt on the
assumption that its abandonment resulted from the short-
sighted over exploitation of resources.
Niemeijer is therefore no doubt right to highlight to a
developmental audience that some of its number may have
underestimated the degree of complexity and dynamism evi-
denced in the historical record of Africa’s past, and in doing
so critics attempts to restore an imagined precolonial
stability and simultaneously counters the accusation that
1258 WORLD DEVELOPMENT
proponents of indigenous knowledge take an ahistorical and
potentially romanticized view of non-western, small-scale
communities as homogenous, technologically static and
intellectually conservative (see also Briggs, 2005, 107–109).
However, by emphasizing the degree and rate of change
within African economies and the complexity of intra- and
inter-societal relationships Niemeijer also highlights the dif-
ficulty in gaining access to the level of historical detail re-
quired under this new paradigm. Historians, archeologists
and paleo-ecologists would certainly question the assertion
that “environmental change is relatively easy to document
through the study of sediments, lake levels, pollen analysis,
and so on” (Niemeijer, 1996, p. 94) in that it would seem to
take an optimistic view of the temporal resolution of paleo-
environmental reconstructions (Butzer, 1996, p. 144), ignores
the problem that these proxies may record very localized ef-
fects (ibid.), and downplays the extent to which different
communities see and react to the environment in different
ways. Similarly, while he is broadly correct in acknowledg-
ing the difficulty of using archeological techniques to discern
social, political and economic factors, the conclusion that
“this poses a clear problem in terms of prehistory [but. . .]
is much less of an issue for the recent past” (ibid.) over-
looks the problem that documentary or oral historical
sources may provide very limited data regarding the physi-
cal or technical aspects of an agronomy and its develop-
ment.
The very robust and detailed case-studies of past practices
required by historical ecologists, resilience theorists and by
Niemeijer’s “new paradigm” are thus likely to consist of
archeological, historical and paleo-environmental evidence
combined with observational data drawn from research into
current social and ecological systems. In fact, the practice of
employing modern ethnographic observations or evidence
from relatively well-defined historical case-studies to flesh-
out deficiencies in historical and archeological data is an ex-
tremely common interpretive device and is arguably an
essential feature of all historical reconstructions. There is a
danger, therefore, that in the absence of an explicit exposi-
tion of sources and inferences, historical interpretations of
this kind will be employed in misleading and circular argu-
ments whereby ethnographic analogies are used to produce
historical case-studies which are then cited within the ethno-
graphic and developmental literature as evidence of the con-
tinuity of local practices (Stahl, 2001, p. 25). Stahl refers to
the histories thus produced as “ethnographic pasts” (2001,
p. 22 following Chance, 1996) and notes that these risk rein-
forcing the stereotype that African communities are histori-
cally static. Yet in the context of the current discussion
there is an additional concern that these circular arguments
will give the false impression that indigenous resource-use
strategies are long-lived and hence both environmentally
and economically sustainable. Indeed, although there has
yet to be an attempt to revive elements of a wholly aban-
doned African agronomy to act as a positive precedent of
local agricultural knowledge, the two projects that have at-
tempted this in Andean Peru both emphasize this effect.
Thus, in outlining the methodology and results of an at-
tempt to re-employ Inca terraces and irrigation structures,
Kendall (2005, p. 211) notes that the experience of operating
aspects of this agronomy itself provided insights into histor-
ical value, observing that “the information gained in the
present [. . .] also became a feedback to help in the interpre-
tation of the archeological data.” Similarly, Erickson’s (e.g.,
1998 (1992)) description of the early stages of his raised field
rehabilitation project in Peru further illustrates this point,
since the first attempts to reconstruct and cultivate these
abandoned features were undertaken as an exercise in exper-
imental archeology designed to provide precisely the sort of
social and technical detail that could not be discerned
through the archeological data alone.
By attempting to re-run elements of now abandoned agron-
omies these applied archeology projects clearly treat the past
as in some way paradigmatic, but they are also quite open
about the potential benefits of incorporating more recent inno-
vations, and about the need to include techniques for which
there may be little or no historical or archeological informa-
tion. In these instances this missing historical data includes
not just social information such as how labor was organized,
but also prosaic technical details such as crop rotations, fal-
lowing cycles, irrigation schedules and fertilization regimes.
Such projects thus clearly represent an amalgam of historical
and contemporary sources of data, but given the near ubiquity
of historical arguments, models and allusions within attempts
to define potential replicable modes of local resource exploita-
tion, this could be said to be true of many indigenous knowl-
edge projects.
7. CONCLUSION: ANCIENT AND BACKWARD OR
LONG LIVED AND SUSTAINABLE?
In Africa, as elsewhere, perceptions of the past play a sig-
nificant role in debates concerning the sustainable manage-
ment of resources since any assessment of sustainability
requires authoritative statements regarding the long-term
consequences of a particular strategy. Conceptions of Afri-
can agriculture thus rely on an historical dimension that is
all too often ignored, imprecisely understood, or presumed
to be somehow self-evident. Yet ignorance of this historical
detail has not stopped the production of narratives which
present African environments as “pristine,” “virgin” or
“natural,” even where these areas have been employed for
either expansive pastoralism or recurrent arable cultivation.
Conversely, in those areas where the level of landscape
modification clearly indicates that they have been subject
to prolonged periods of permanent or recurrent cultivation,
the application of conflicting models of agricultural change
has led to contrasting conclusions regarding their future sus-
tainability. The extant examples of east African indigenous
intensive agriculture have thus been both denigrated and
promoted, with proponents of external intervention empha-
sizing their antiquity in order to justify attempted improve-
ments whereas advocates of indigenous knowledge have seen
their apparent longevity as evidence of their economic and
environmental sustainability. It is clearly misleading, how-
ever, to claim that local resource exploitation strategies
are either unsustainable or environmentally appropriate in
the absence of historical information concerning changes
to the environment, climate or techniques of cultivation
through time. Lack of historical depth and precision thus
represents a substantial obstacle to contemporary debates
that reference local economic practices, regardless of
whether the discussants are proponents or detractors in re-
spect of the value of indigenous knowledge.
Attempts to cite the history of African agriculture in defense
of developmental interventions should therefore be viewed
critically, principally on the grounds that this history is not
sufficiently well understood at present to draw definitive con-
clusions, and is far too diverse both spatially and temporally
to enable the formulation of simple “lessons from the past.”
On the basis of current data, for example, the history of
 “ANCIENT AND BACKWARD OR LONG-LIVED AND SUSTAINABLE?”
Marakwet, Kenya, indicates that the techniques employed
have allowed intensive agricultural production for at least four
centuries and, as such, it might be tempting to conclude that
aspects of the agronomy could be replicated elsewhere
(Kipkorir et al., 1983, particularly Ssennyonga, 1983). Yet
similar techniques in Arusha, Tanzania, are less than two cen-
turies old and appear to be showing signs of strain insofar as
the system requires increasing levels of labor-inputs to main-
tain soil fertility: the 19th century practice of rotating fields be-
tween crop production and cattle pasture having been replaced
by the stallage of cattle with its attendant needs for the trans-
portation of fodder and manure (Spear, 1997, p. 53). In
contrast, the long predicted Malthusian collapse of the agri-
cultural systems in Kigezi and Machakos have yet to occur
(Carswell, 2007; Tiffen et al. 1994), though it is also notewor-
thy in the context of the current discussion that the use of his-
torical arguments of this sort and the potential to predict
future responses or to extrapolate to other areas have also
been critiqued (Siedenburg, 2006).
At present, therefore, the state of historical knowledge
regarding indigenous African agriculture is ambiguous in
terms of development planning unless one takes the level
of dynamism evident within these agronomies as an
endorsement of the view referred to by Sillitoe (1998, p.
225) as “extreme empowerment,” that is that local farmers
have long demonstrated an ability to experiment with and
make informed decisions about the usefulness of available
technology and should therefore be given access to these re-
sources and allowed to design their own solutions (ibid. cit-
ing Chambers et al., 1989; Scoones and Thompson, 1994,
and others). As noted previously in relation to the discus-
sion presented by Niemeijer (1996), such arguments are
attractive in that they acknowledge the long history of dy-
namic and adaptive behavior displayed by local farmers,
but the position is so vague as to be hardly an historical
argument at all, and is really saying nothing more than
African communities—in common with all societies world-
wide—have largely muddled through to date. This is not
to say that they can do so indefinitely (Brockington &
Homewood, 2001). Moreover, the level of spatial and tem-
poral variation and the extent of recent and on-going
changes within these agronomies serve to illustrate that con-
tinuity of occupation cannot be equated simplistically with
continuity of practice, and argues against the conclusion
that the failure of large-scale interventions can be countered
by promoting the “tried and tested” techniques employed by
local communities (Adams & Anderson, 1988, pp. 532–535).
Seen in this light the lesson to be taken from the history of
local agricultural adaptations could be simply that agrono-
mies, like ecosystems generally, are only ever provisionally
stable (Behnke, Scoones, & Kerven, 1993), and that success-
ful strategies rely on the convergence of largely non-replica-
ble historical contingencies. Indeed, this is largely the
conclusion drawn by McCann (1990, p. 133) who regards
the brief intensification of sorghum production that took
place in the Mazega lowlands of northwest Ethiopia in
the early 20th century as an “historical accident” and that
“the Mazega’s agricultural revolution is an anti-model since
its success derived from a confluence of historical factors
[. . .] rather than a set of synchronic factors which can be
replicated”; the contingent dynamics in this instance being
the conjunction of labor and fertile soils in a depopulated
area with access to markets that remained unregulated dur-
ing the period in which colonial power was being consoli-
dated in Ethiopia, Eritrea and Anglo-Egyptian Sudan.
Such a conclusion is mirrored by Anderson’s (2002) apprai-
1259
sal of the development of the Chamus irrigation system near
Lake Baringo, Kenya, between approximately 1840 and
1920, though the re-establishment of small-scale irrigation
at Baringo since the 1960s (Adams & Anderson, 1988, pp.
530–531; Anderson, 2002) demonstrates that these conver-
gences of resources and opportunities can occur at different
times for very different reasons.
Yet just as “historical accidents” can create opportunities
they can also create restrictions which may lead to the
“premature” curtailment of potentially successful economic
strategies, as may well be the case in the Dime area of Ethiopia
where terraces abandoned following conflicts with the Ethio-
pian empire at the turn of the 20th century were subsequently
employed for slash-and-burn agriculture whereas those in
nearby Konso continued to be cultivated intensively (Amborn,
1989, p. 82n). Given that oral historical and genealogical evi-
dence indicate that the agronomy at Konso may have sup-
ported a comparatively high rural population for over five
centuries (Amborn, 1989, p. 73) the possibility that the tech-
niques formerly employed at Dime could be sustainably re-
established deserves to be explored. Equally, the recognition
that the abandonment of extensive areas of agricultural fea-
tures at Dime may have had external causes demonstrates
the simplicity of the formula that sees abandoned systems as
failed systems.
One can rightly criticize governments and development
agencies for attempting to reduce complex societies and eco-
systems with equally complex histories to simple cause and
effect; problem and solution (Rocheleau et al., 1995). Never-
theless, it is unrealistic to expect such organizations to wait
for historians to resolve these historical questions, and only
an extreme optimist would wait for them to design future
solutions. Yet it is equally unrealistic to expect historians,
archeologists and paleo-ecologists not to comment on the
false assumptions and conclusions cited within these devel-
opmental narratives, just as it is inevitable that historians
and development professionals will continue to cite the dis-
credited premises of earlier paradigms in defence of new ap-
proaches. Of course, it needs to be acknowledged that some
of these questions may never be resolved, but other hypoth-
eses can and should be tested. Referring to several of the
unresolved historical questions touched upon above in refer-
ence to indigenous conservation, for example, Gillson et al.
(2003, p. 384) note that “Only archeological investigation of
the soil profiles of several sacred forests, radiocarbon dating
of charcoal from precolonial iron-smelting sites and analysis
of fossil pollen can conclusively document how and when
the vegetation of North Pare, Mkomazi and Tsavo
changed.” Numerous similar enquiries demanding a range
of historical, archeological and paleo-environmental meth-
ods alongside the direct observation of contemporary prac-
tices could also be cited, and include repeat photography
(whether from the ground, air or from satellites) to docu-
ment landscape change (e.g., Boerma, 2006); the measure-
ment of carbon13 isotopic levels in soils to reconstruct
forest vegetation histories (e.g., Eshetu & Högberg, 2000);
or simply the examination of archeological stratigraphy to
investigate the historical development of agricultural systems
(e.g., Stump, 2006). Since the current interest in the use of
indigenous knowledge in development planning shows no
sign of abating in the immediate future, this combination
of contemporary and historical approaches should produce
a more complete understanding of the long-term history of
specific examples of African resource exploitation strategies,
and in doing so may provide information essential to ade-
quately qualify these debates.
1260 WORLD DEVELOPMENT
NOTES
1. The term ‘indigenous’ is problematic from an historical point of view
in that it might be taken to imply a longevity and continuity of occupation
that may be inaccurate, inappropriate, contested or impossible to prove
(Ingold, 2000, chapter 8; Kuper, 2003; Nygren, 1999). The phrase
‘indigenous knowledge’ and its numerous variants are in common usage
however, and is thus also employed here.
2. Although none of these east African intensive agronomies are included
within the pilot study, they would appear to be strong candidates for
inclusion on the FAO’s list of Globally Important Agricultural Heritage
Systems (GIAHS), a report on the identification and protection of which
recommends an ‘historical and archaeological description of the system or
site’ (FAO, 2004, Annex 17) and notes that ‘A clear understanding of their
history and current trends [. . .] is essential to identify pathways of
evolution that will maintain their resilience and options for the future’
(FAO, 2004, p. 1, ibid.).
REFERENCES
Adams, W. M. (1996). Irrigation, erosion and famine: Visions of
environmental change in Marakwet, Kenya. In M. Leach, & R.
Mearns (Eds.), The Lie of the Land: Challenging received wisdom on the
African environment (pp. 155–167). London: James Currey.
Anderson, W. M., & Adams, D. M. (1988). Irrigation before development:
indigenous and induced change in agricultural water management in
east Africa. African Affairs, 87, 519–535.
Adams, W. M., & Grove, A. T. (Eds.) (1984). Irrigation in tropical Africa –
Problems and problem-solving. Cambridge: African Studies Centre.
Alteiri, M. A., & Koohafkan, P. (2008). Enduring farms: Climate change,
smallholders and traditional farming communities. Penang: Third World
Network.
Amborn, H. (1989). Agricultural intensification in the Burji-Konso cluster
of south western Ethiopia. In J. E. G. Sutton (Ed.). History of African
Agricultural technology and field systems (Vol. 24, pp. 71–83). Azania.
Anderson, D. M. (1984). Depression, dust-bowl, demography and
drought: The colonial state and soil conservation in east Africa during
the 1930s. African Affairs, 83, 196–210.
Anderson, D. M. (2002). Eroding the commons: The politics of ecology in
Baringo Kenya 1890–1963. Oxford: James Currey.
Anderson, D., & Grove, R. (1987). Introduction: The scramble for Eden:
Past, present and future in African conservation. In D. Anderson, & R.
Grove (Eds.), Conservation in Africa: People, policies and practice
(pp. 1–12). Cambridge: Cambridge University Press.
Balée, W. (2006). The research program of historical ecology. Annual
Review of Anthropology, 35, 75–98.
Barker, G., & Gilbertson, D. (2000). Living at the Margin: Themes in the
archaeology of drylands. In G. Barker, & D. Gilbertson (Eds.), The
archaeology of drylands: living at the Margin (pp. 1–17). London:
Routledge.
Behnke, R. H., Scoones, I., & Kerven, C. (Eds.) (1993). range ecology and
disequilibrium: New models of natural variability and pastoral adapta-
tion in African Savannahs. London: Overseas Development Institute.
Beinart, W. (1984). Soil erosion, conservation, and ideas about develop-
ment: A Southern African exploration, 1900–1960. Journal of Southern
African Studies, 11, 52–83.
Boerma, P. (2006). Assessing forest cover change in Eritrea – A historical
perspective. Mountain Research and Development, 26(1), 41–47.
Börjeson, L. (2007). Boserup backwards? Agricultural intensification as
‘its own driving force’ in The Mbulu Highlands, Tanzania. Geografiska
Annaler: Series B, Human Geography, 89(3), 249–267.
Boserup, E. (1965). The conditions of agricultural growth: The economics of
Agrarian change under population pressure. London: Allen and Unwin.
Boserup, E. (1981). Population and technology change: Study of long-term
trends. Oxford: Blackwell.
Briggs, J. (2005). The use of indigenous knowledge in development:
Problems and challenges. Progress in Development Studies, 5(2),
99–114.
Brockington, D., & Homewood, K. (2001). Degradation debates data
deficiencies: The Mkomazi Game Reserve, Tanzania. Africa, 71(3),
449–480.
3. Originally Stichting Nederlandse Vrijwilligers but now known simply
as SNV Nederlandse Ontwikkelingsorganisatie or SNV Netherlands
Development Organisation.
4. Deutsche Gesellschaft für Technische Zusammenarbeit GmbH.
5. Originally the successor project to the SNV and GTZ programme
from which it took the name ‘Traditional Irrigation Project’, and now
known as the Traditional Irrigation and Environmental Development
Organisation.
6. Participatory Agricultural Development and Empowerment Project:
a rural development initiative funded by the Tanzania Ministry of
Agriculture and Food Security, the World Bank and other international
donors.
Brooks, T. M., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A.
B., Rylands, A. B., Konstant, W. R., et al. (2002). Habitat loss and
extinction in the hotspots of biodiversity. Conservation Biology, 16,
909–923.
Brush, S. B. (2007). Farmers’ rights and protection of traditional
agricultural knowledge. World Development, 35(9), 1499–1514.
Burra, R., & van den Heuvel, K. (1987). Traditional Irrigation in Tanzania.
Historical analysis of irrigation tradition and government intervention
(Vol. 1). Dar es Salaam: SNV-TIP.
Butzer, K. W. (1996). Ecology in the long view: Settlement histories,
agrosystemic strategies, and ecological performance. Journal of Field
Archaeology, 23(2), 141–150.
Chambers, R. J. H. (1973). The Perkerra irrigation scheme: A contrasting
case. In R. Chambers, & J. Moris (Eds.), Mwea: An irrigated rice
settlement in Kenya (pp. 344–364). Munich: Weltforum Verlag.
Carswell, G. (2002). Farmers and fallowing: Agricultural change in Kigezi
District, Uganda. Geographical Journal, 168(2), 130–140.
Carswell, G. (2007). Cultivating success in Uganda: Kigezi farmers and
colonial policies. Nairobi: BIEA; Oxford: James Currey; Athens, Ohio:
Ohio University Press.
Conelly, W. T. (1994). Population pressure, labour availability, and
agricultural disintensification: The decline of farming on Rushinga
Island, Kenya. Human Ecology, 22(2), 145–170.
Constanza, R., Graumlich, L., Steffen, W., Crumley, C., Dearing, J.,
Hibbard, K., et al. (2007). Sustainability or collapse: What can we
learn from integrating the history of humans and the rest of nature?.
Ambio, 36(7), 522–527.
Conte, C. A. (2004). Highland sanctuary: Environmental history in
Tanzania’s Usambara Mountains. Athens, Ohio: Ohio University Press.
Curry, J. R. (1939). Elusine cultivation by the Wachagga on Kilimanjaro.
The East African Agricultural Journal, 5(2), 386–390.
Dearing, J. A. (2007). Integration of World and Earth Systems: Heritage
and Foresight. In A. Hornborg, & C. Crumley (Eds.), The world system
and the earth system. Global socioenvironmental change and sustain-
ability since the neolithic (pp. 38–55). Walnut Creek: Left Coast Press.
Devenne, F. (2006). On ‘The Mountain of Waters’, the furrows are
running dry: Rethinking Mount Kilimanjaro’s irrigation system. In F.
Bart, M. J. Mbonile, & F. Devenne (Eds.), Kilimanjaro. Mountain,
Memory, Modernity (pp. 201–214). Dar es Salaam: Mkuki na Nyota
Publishers Ltd..
Erickson, C. L. (1998 [1992]). Applied archaeology and rural develop-
ment: archaeology’s potential contribution to the future. In M. B.
Whiteford, & S. Whiteford (Eds.), Crossing currents: Continuity and
change in Latin America (pp. 34–45). Upper Saddle River: Prentice-
Hall. Reprinted from Journal of the Steward Anthropological Society,
20(1–2), 1–16.
Erickson, C. L. (2006). Intensification, political economy, and the
farming community: In defence of a bottom-up perspective of the
past. In J. Marcus, & C. Stanish (Eds.), Agricultural Strategies
(pp. 334–363). Los Angeles: Cotsen Institute of Archaeology,
University of California.
 “ANCIENT AND BACKWARD OR LONG-LIVED AND SUSTAINABLE?”
Eshetu, Z., & Högberg, P. (2000). Reconstruction of a forest site history in
the Ethiopian Highlands based on 13-C natural abundance in soils.
Ambio, 29(2), 83–89.
Fairhead, J., & Leach, M. (1996). Rethinking the forest-savannah mosaic:
Colonial science and its relics in west Africa. In M. Leach, & R.
Mearns (Eds.), The Lie of the Land: Challenging received wisdom on the
African environment (pp. 105–121). London: James Currey.
FAO [Food and Agriculture Organization of the United Nations]. (1990).
The conservation and rehabilitation of African lands: an international
scheme. Rome: FAO. Available online at: <http://www.fao.org/
docrep/z5700e/z5700e00.HTM> (accessed in December 2008).
FAO [Food and Agriculture Organization of the United Nations]. (2004).
Globally-important ingenious agricultural heritage systems GIAHS
project. Report of the second international workshop and steering
committee meeting, Rome, 7–9 June 2004. Rome: FAO. Available
online at: <http://www.fao.org/nr/giahs/meetings-events/second-inter-
national-workshop-and-steering-committee-meeting/en/> (accessed in
March 2010).
Feierman, S. (1990). Peasant intellectuals, anthropology and history in
Tanzania. Madison: University of Wisconsin Press.
Ferguson, J. (1990). The anti-politics machine: ‘Development’, depolitici-
zation and bureaucratic power in Lesotho. Cambridge: Cambridge
University Press.
Gillson, L., Sheridan, M., & Brockington, D. (2003). Representing
environments in flux: Case studies from east Africa. Area, 35(4),
371–389.
Grove, A. (1993). Water use by the Chagga on Kilimanjaro. African
Affairs, 92, 431–448.
Grove, A. T., & Sutton, J. E. G. (1989). Agricultural terracing south of the
Sahara. In J. E. G. Sutton (Ed.). History of African agricultural
technology and field systems (Vol. 24, pp. 114–122). Azania.
Håkansson, N. T. (1995). Irrigation, population pressure, and exchange in
precolonial Pare, Tanzania. Research in Economic Anthropology, 16,
297–323.
Håkansson, N. T. (2007). The decentralized landscape: Regional wealth
and the expansion of production in Northern Tanzania before the eve
of colonialism. In L. Cliggett, & C. Pool (Eds.), Economies and the
transformation of landscape (pp. 239–265). Walnut Creek: AltaMira
Press.
Harrop, S. R. (2007). Traditional agricultural landscapes as protected
areas in international law and policy. Agriculture, Ecosystems and
Environment, 121, 296–307.
Hoben, A. (1996). Paradigms and politics in Ethiopia. In M. Leach, & R.
Mearns (Eds.), The Lie of the Land: Challenging received wisdom on the
African environment (pp. 186–208). Oxford: James Currey.
Holling, C. S., Berkes, F., & Folke, C. (1998). Science, sustainability and
resource management. In F. Berkes, & C. Folke (Eds.), Linking social
and ecological systems (pp. 242–263). Cambridge: Cambridge Univer-
sity Press.
Ingold, T. (2000). The perception of the environment: Essays on livelihood,
dwelling and skill. London: Routledge.
Johnston, H. H. (1886). The Kili-manjaro expedition; a record of scientific
exploration in Eastern Equatorial Africa. London: Keegan P., Trench
and Co.
Kanthack, F. E. (1938). Report on the control of the natural waters of
Tanganyika and the framework of the water law on which such control
should be based. Dar es Salaam: Government Printer.
Kaswamila, A. L., & Masuruli, B. M. (2004). The role of traditional
irrigation systems in poverty alleviation in semi-arid areas: The case of
Chamanzi in Lushoto District, Tanzania. Dar es Salaam: Mkuki Na
Nyota.
Kendall, A. (2005). Applied archaeology: Revitalising indigenous agricul-
tural technology within an Andean community. Public Archaeology, 4,
205–221.
Kimambo, I. N. (1969). A political history of the Pare of Tanzania, c. 1500–
1900. Nairobi: East African Publishing House.
Kimambo, I. N. (1991). Penetration and protest in Tanzania: The impact
of the world economy on the Pare, 1860–1960. London: James
Currey.
Kipkorir, B. E., Soper, R. C., & Ssennyonga, J. W. (Eds.) (1983). Kerio
valley: Past, present and future. Nairobi: University of Nairobi Press,
Institute of African Studies.
Kjekshus, H. (1977). Ecology control and economic development in east
African history: The case of Tanganyika. Berkeley: University of
California Press.
1261
Kodalo, J. T. (2000). An archaeological reconnaissance of the Marakwet
irrigation furrows: A case study from the Embotut and Arror Rivers,
Kerio Valley, Kenya. Nyame Akuma, 53, 12–21.
Koponen, J. (1988). People and Production in Late Pre-Colonial Tanzania.
History and Structures. Helsinki: Finnish Society for Development
Studies; Finnish Anthropology Society; Finnish Historical Society;
Scandinavian Institute of African Studies.
Kuper, A. (2003). The return of the native. Current Anthropology, 44(3),
389–402.
Lambin, E. F. et al. (2001). The causes of land-use and land-cover change:
Moving beyond the myths. Global Environmental Change, 11, 261–269.
Leach, M., & Mearns, R. (1996). Challenging received wisdom in Africa.
In M. Leach, & R. Mearns (Eds.), The Lie of the Land: Challenging
received wisdom on the African environment (pp. 1–33). Oxford: James
Currey.
Lein, H. (1998). Traditional versus modern water management systems in
Pangani river basin, Tanzania. In L. de Haan, & P. Blaikie (Eds.),
Looking at Maps in the Dark. Directions for geographical research in
land management and sustainable development in rural and urban
environments of the Third World (pp. 52–64). Utrecht: Royal Dutch
Geographical Society & Faculty for Environmental Sciences, Univer-
sity of Amsterdam.
Malthus, T. R. (1985) [originally published anonymously in 1798]. An
essay on the principle of population, and a summary view of the principle
of population. Edited with an Introduction by A. Flew. London:
Penguin.
Masao, F. T. (1974). The irrigation system in Uchagga: An Ethno-
historical Approach. Tanzania Notes and Records, 75, 1–8.
Mbaga-Semgalawe, Z., & Folmer, H. (2000). Household adoption
behaviour of improved soil conservation: The case of the North Pare
and West Usambara Mountains of Tanzania. Land Use Policy, 17(4),
321–336.
McCann, J. C. (1990). A Dura revolution and frontier agriculture in
Northwest Ethiopia, 1898–1930. Journal of African History, 31,
121–134.
Moore, H. (1983). Anthropology and development: Some illustrations
from the Marakwet of Kenya. In B. E. Kiporir, R. C. Soper, & J. W.
Ssennyonga (Eds.), Kerio valley: Past, present and future (pp. 132–138).
Nairobi: University of Nairobi Press.
Morrison, K. (1996). Typological schemes and agricultural change:
Beyond Boserup in precolonial South India. Current Anthropology,
37, 583–608.
Murton, J. (1999). Population growth and poverty in Machakos, Kenya.
Geographical Journal, 165(1), 37–46.
Netting, R. (1993). Smallholders, householders. Farm families and the
ecology of intensive, sustainable development. Stanford: Stanford
University Press.
New, C. (1873). Life, wanderings and labours in eastern Africa. With an
account of the first successful ascent of the equatorial snow mountain,
Kilima-Njaro and remarks upon east African Slavery. London: Hodder
and Stoughton.
Niemeijer, D. (1996). The dynamics of African Agricultural History: Is it
time for a new development paradigm. Development and Change, 27,
87–110.
Nygren, A. (1999). Local knowledge in the environment-development
discourse: From dichotomies to situated knowledges. Critique of
Anthropology, 19(3), 267–288.
Oldfield, F. (2007). Toward developing synergistic linkages between the
biophysical and the cultural: A palaoenvironmental perspective. In A.
Hornborg, & C. Crumley (Eds.), The world system and the earth
system. Global socioenvironmental change and sustainability since the
neolithic (pp. 29–37). Walnut Creek: Left Coast Press.
Redman, C. L. & Kinzig, A. P. (2003). Resilience of past landscapes:
Resilience theory, society, and the Longue Durée. Conservation
Ecology 7(1), p. 14, <http://www.consecol.org/vol7/iss1/art14>.
Reij, C., Scoones, I., & Toulmin, C. (Eds.) (1996). Sustaining the soil.
Indigenous soil and water conservation in Africa. London: Earthscan.
Richards, P. (1985). Indigenous agricultural revolution: Ecology and food
production in West Africa. London: Hutchinson and Co.
Rocheleau, D. E., Steinberg, P. E., & Benjamin, P. A. (1995). Environ-
ment, development, crisis, and crusade: Ukambani, Kenya, 1890–1990.
World Development, 23(6), 1037–1051.
Sheridan, M. J. (2002). An irrigation intake is like a uterus: Culture and
agriculture in precolonial North Pare, Tanzania. American Anthropol-
ogist, (1), 79–92.
1262 WORLD DEVELOPMENT
Shetler, J. B. (2007). Imagining serengeti. A history of landscape memory in
Tanzania from earliest times to the present. Athens, Ohio: Ohio
University Press.
Siedenburg, J. (2006). The Machakos case study: Solid outcomes,
unhelpful hyperbole. Development Policy Review, 24(1), 75–85.
Sillar, B. (2005). Who’s indigenous? Whose archaeology? In B. Sillar & C.
Fforde (Eds.), Conservation, identity and ownership in indigenous
archaeology. Special issue of Public Archaeology, 4, 71–94.
Sillitoe, P. (1998). The development of indigenous knowledge: A new
applied anthropology. Current Anthropology, 38(2), 223–252.
Sillitoe, P. (2006). Ethnobiology and applied anthropology: Rapproche-
ment of the academic with the practical. In E. Roy (Ed.), Ethnobiology
and the science of humankind. Journal of the Royal Anthropological
Institute, Special Issue 2006, S119–142.
Sokoni, C. H., & Shechambo, T. C. (2005). Changes in the upland
irrigation system and implications for rural poverty alleviation: A case of
the Ndiwa irrigation system, West Usambara Mountains, Tanzania. Dar
es Salaam: Mkuki Na Nyota.
Spear, T. (1997). Mountain farmers: Moral economies of land and
agricultural development in Arusha and Meru. Oxford: James Currey.
Ssennyonga, J. W. (1983). The Marakwet irrigation system as a model of a
systems-approach to water management. In B. E. Kipkorir, R. C.
Soper, & J. W. Ssennyonga (Eds.), Kerio valley: Past, present and
future (pp. 96–111). Nairobi: University of Nairobi Press, Institute of
African Studies.
Stahl, A. B. (2001). Making history in Banda. Anthropological visions of
Africa’s past. Cambridge: Cambridge University Press.
Stahl, K. M. (1964). History of the Chagga people of Kilimanjaro. London:
Mouton & Co.
Stocking, M., & Perkin, S. (1992). Conservation-with-development: An
application of the concept in the Usambara Mountains, Tanzania.
Transactions of the Institute of British Geographers, 17(3), 337–349.
Stöhr, W. B., & Taylor, D. R. F. (Eds.) (1981). Development from above or
below? The dialectics of regional planning in developing countries.
Chichester: Wiley.
Stump, D. (2003). The soils of Engaruka: preliminary soil exhaustion tests
on a pre-colonial agricultural landscape in Tanzania. In P. Mitchell, A.
Haour, & J. Hobart (Eds.). Researching Africa’s Past. New contribu-
tions from British archaeologists (Vol. 57, pp. 101–109). Oxford:
Oxford University School of Archaeology, Monograph.
Stump, D. (2006). The development and expansion of the field and
irrigation system at Engaruka Tanzania. Azania, 41, 69–94.
Stump, D., & Tagseth, M. (2009). The history of precolonial and early
colonial agriculture on Kilimanjaro: A review. In T. Clack (Ed.),
Culture history and identity: Landscapes of inhabitation in the Mount
Kilimanjaro Area Tanzania (pp. 107–124). BAR International Series
1966. Oxford: Archaeopress.
Sutton, J. E. G. (1984). Irrigation and soil conservation in African
agricultural history with a reconsideration of the Inyanga terracing
(Zimbabwe) and Engaruka irrigation works (Tanzania). Journal of
African History, 25, 25–41.
Available online at www.sciencedirect.com
Sutton, J. E. G. (1989). Towards a history of cultivating the fields. In J. E.
G. Sutton (Ed.), History of African agricultural technology and field
systems (Vol. 24, pp. 99–111). Azania.
Sutton, J. E. G. (1998). Engaruka: An irrigation agricultural community
in Northern Tanzania before the Maasai. Azania, 33, 1–33.
Swynnerton, R. J. M. (1949). Some problems of the Chagga on
Kilimanjaro. East African Agricultural Journal (Amani), January,
117–132.
Tagseth, M. (2006). The ‘mfongo’ irrigation systems on the slopes of Mt
Kilimanjaro, Tanzania. In T. Tvedt, R. Coopey, E. Jakobsson, & T.
Østigaard (Eds.). A history of water: Water control and river
biographies (Vol. 1, pp. 488–506). London: I.B.: Tauris.
Tiffen, M. (1996). Land and Capital: Blind spots in the study of the
“resource-poor” farmer. In M. Leach, & R. Mearns (Eds.), The Lie of
the Land: Challenging received wisdom on the African environment
(pp. 168–185). London: James Currey.
Tiffen, M., Mortimore, M., & Gichuki, F. (1994). More people, less
erosion: Environmental recovery in Kenya. Chichester: Wiley.
UN [United Nations]. (1992). Agenda 21: The United Nations plan for
action from Rio. United Nations: New York.
Walker, B., Holling, C. S., Carpenter, S. R. & Kinzig, A. (2004).
Resilience, adaptability and transformability in social–ecological
systems. Ecology and society 9(2), 5. URL: <http://www.ecology-
andsociety.org/vol9/iss2/art5> (accessed October 2008).
Watson, E. E. (2009). Living terraces in Ethiopia. Konso landscape, culture
and development. Woodbridge: James Currey.
Westerberg, L.-O., Holmgren, K., Börjeson, L., Håkansson, N. T.,
Laulumaa, V., Ryner, M., et al. (2007). The development of the ancient
irrigation system at Engaruka, Northern Tanzania: Physical and
societal factors. In M. Ryner (Ed.), 2007: Past environmental and
climate changes in northern Tanzania: Vegetation and lake level
variability in Empakaai Crater. Doctoral Thesis No. 5, Dept. of
Physical Geography and Quaternary Geology, Stockholm University.
Widgren, M. (1999). Islands of intensive agriculture in eastern Africa: The
social, ecological and historical contexts. In M. Widgren & J. E. G.
Sutton (Eds.), ‘Islands’ of intensive agriculture in the east African Rift
and Highlands: A 500 year perspective (pp. 3–14). EDSU Working
papers No. 43. Stockholm: Stockholm University Press.
Widgren, M. (2004). Towards a historical geography of intensive farming
in eastern Africa. In M. Widgren, & J. E. G. Sutton (Eds.), Islands of
intensive agriculture in eastern Africa (pp. 1–18). Oxford: James Currey.
Widgren, M. (2007). Pre-colonial landesque capital: A global perspective.
In A. Hornborg, J. Martinez-Alier, & J. R. Mcneil (Eds.), Rethinking
environmental history: World-system history and global environmental
change (pp. 61–77). Walnut Creek: AltaMira Press.
Widgren, M., & Sutton, J. E. G. (Eds.) (2004). Islands of intensive
agriculture in eastern Africa. Oxford: James Currey.
Yoshida, M. (1985). Traditional furrow irrigation systems in the South
Pare Mountains area of Tanzania. In A. Mascarenhas, J. Ngana, & M.
Yoshida (Eds.), Opportunities for irrigation development in Tanzania
(pp. 33–69). Tokyo: Institute of Developing Economies.