9: .. . 50.15. I At I

, 1,1,3. .5: z
, . ., "aw? ..,..s...:...n.,.n£_ »£.n.§3;
(find-3.1.“: . 5

a}. abacus—and. has...
wit. :. s}.‘t$§kx-§.tln~
, {1.3.5 .93.... 35:...
\‘n uyb.’

.c f.) 1 I

.. .ifis‘ktxl

2. 3:80.35

.5 . a. Q I

\. I CI..I

I >3...sz

.15 ‘39

1‘ x. 14
4.. . 3.1.,
1.3.2:}!
a! at. V :3
v

.e:.::.: IAIu
., . I!
3!.

 

Mqun’

Mn unavéia.

 

v!“

IHESIS
3
gout

 

.‘—'B An r
M'Ch'gan State
University _

 

 

 

This is to certify that the

dissertation entitled

LIVELIHOODS AND LAND USE CHANGE IN RURAL SOUTH AFRICA:
THE UNFINISHED TRANSFORMATION

presented by

Brent McCusker

has been accepted towards fulfillment
of the requirements for

Ph.D. Geography

 

 

degree in

‘ mfl’f‘fia“

 

Major prdfoeeo/ '

Date Ajbar ‘29 2.937

MSU i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

1M9». Q ZBZIUU5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6I01 c:/CIRC/DaIeDue.p65-p.15

LIVELIHOODS AND LAND USE CHANGE m RURAL SOUTH AFRICA :
THE UNFINISHED TRANSFORMATION

by
Brent McCusker

A DISSERTATION
Submitted to:
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Geography

2001

ABSTRACT

LIVELIHOODS AND LAND USE CHANGE IN RURAL SOUTH AFRICA :
THE UNFINISHED TRANSFORMATION

By

Brent McCusker

The principal issues discussed in this dissertation stem from the
introduction of six land reform projects into six different rural communities in
the Northern Province of South Africa. I uncover the impacts of the projects by
integrating traditional social science methods with geographic information
technologies such as GIS and remote sensing. Data sources include household
survey data collected by the author and six Landsat images. I argue that
livelihood systems have not significantly benefited from the introduction of land
reform projects, nor have the expected changes in land use occurred. I conclude

by presenting the major reasons for the failure of the land reform projects.

Copyright by
Brent McCusker

2001

ACKNOWLEDGEMENTS

Without the support and patience of my wife, Eileen, this dissertation
would not have been possible. I appreciate the help and diligence of Dr. David,
J. Campbell, my advisor, in the long process we started in 1995. I extend my
thanks to all of those people who made my field research possible: Prof. Paul
Fouche, Marubini ”the Octo” Ramundzuli, Parvin Shaker, Flip Schoeman,
Stephanie Dippenaar, and my research assistants. Further, I wish to thank my
committee, Dr. David Lusch, Dr. David Skole, and Dr. Anne Ferguson for
reviewing my drafts and providing pivotal insights. The ordeal of being a
graduate student was made easier by Laura Cahue. Finally, thanks to my family
who have been most supportive through this process.

Nkosi Sikelel iAfrica!

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................. Vii
LIST OF FIGURES ........................................................................... ix
LIST OF TERMS AND GEOGRAPHIC NAMES .......................................... . xi
CHAPTER ONE
INTRODUCTION ........................................................................... 1
CHAPTER TWO
REVIEW OF THE LIIERATURE ............................................................. 8
2.1 Introduction ............................................................ 8
2.2 Contextualizing Rural Livelihoods .............................. 9
2.2.1 Definition, Composition, and Debate .................... 9
2.2.2 Gender ........................................................... 14
2.2.3 Research Heritage ............................................. 20
2.2.4 Discussion ...................................................... 22
2.3 Land Use and Cover Change .................................... 23
2.3.1 Research Heritage ............................................. 23
2.3.2 Theories on Land Use ....................................... 24
2.3.3 Scale .............................................................. 26
2.3.4 Examples and Case Studies ................................ 28
2.3.5 Land Use Change in South Africa ........................ 32
2.4 Land Alienation and Reform in South Africa ............... 34
2.4.1 Landlessness ................................................... 34
2.4.2 Tenure ........................................................... 38
2.4.3 Land Reform Choices and Policy ......................... 43
2.4.4 The Community Association Model .................... 57
2.5 The Impact of Change Through Time : Examples
from the Resettlement Literature ................................. 60
2.5.1 Resettlement and the CPAs .................................. 61
2.5.2 Lessons from Latin America ................................. 63
2.5.3 The Issues of Time and Transition .......................... 65
2.5.4 State Farms and Cooperatives: Lessons for the CPA
Program ......................................................... 68
2.5.5 Example from the Land Reform Literature .............. 73
2.6 Hypotheses and Research Questions ............................ 74

2.6.1 Hypotheses Made in This Study ............................ 74

2.6.2 Research Questions ............................................. 75
2.7 Summary .................................................................. 77
CHAPTER THREE
RESEARCH SETTING AND STUDY SITES ............................................... 78
3.1 Introduction ............................................................. 78
3.2 Demographics and History .......................................... 79
3.2.1 Political Geography ............................................. 79
3.2.2 Structure of the Population .................................... 80
3.2.3 Economy, Infrastructure and Development .............. 81
3.2.4 Physical Geography, Weather and Climate ............... 82
3.2.5 Political History ................................................... 83
3.3 Census Data on the Northern Province ........................... 84
3.3.1 Demographic Data ............................................. 84
3.3.2 Land ............................................................... 85
3.3.3 Livelihoods ...................................................... 86
3.4 Land Reform in the Northern Province ........................ 87
3.5 Summary ............................................................... 91
CHAPTER FOUR
DATA ANALYSIS ........................................................................... 93
4.1 Introduction ............................................................ 93
4.2 Structure of the Research ........................................... 94
4.2.1 Methods ........................................................... 97
4.3 Interview Data ......................................................... 99
4.3.1 Demographic Data ............................................. 99
4.3.2 Farming as Income Generation .............................. 100
4.3.3 Identification of Food Shortages ............................ 101
4.3.4 Coping Mechanisms to Food Shortages .................. 102
4.3.5 Livelihood Systems ............................................. 104
4.3.6 Land Use and Land Authority .............................. 107
4.3.7 The CPA and Perceptions of Livelihoods ................. 110
4.3.8 Analysis by Gender ............................................. 111
4.3.9 Relationships in the Data ..................................... 113
4.4 Assessment of Land Cover Change Derived from
Satellite Imagery ........................................................ 120
4.4.1 Methods and Scene Properties ............................... 120
4.4.2 Findings ............................................................ 127
4.5 Social Driving Forces of Land Use Change ..................... 133
4.5.1 Land Use Patterns on the CPAs .............................. 133
vi

4.5.2 Land Use Patterns on the Control Farm ................... 135

4.5.3 Perceptions of Change ......................................... 135
4.6 Summary ............................................................... 138
CHAPTER FIVE
DISCUSSION AND SYNTHESIS ............................................................. 141
5.1 Introduction ............................................................ 141
5.2 Evaluation of the Impacts of CPAs on Member’s
Livelihoods ............................................................ 147
5.3 Analysis of Satellite Imagery ....................................... 151
5.3.1 Verifying the Interviews ...................................... 151
5.3.2 “People and Pixels” or ” People or Pixels” ............... 152
5.4 Summary .............................................................. 155
CHAPTER SD(
NARRATIVE - THE UNFINISHED TRANSFORMATION - IMPEDIMENTS TO
LIVELIHOOD ENHANCEMENT AND LAND USE CHANCE ........................... 157
6.1 Introduction ............................................................ 157
6.2 Impediments at the Household and Farm Scale ............... 161
6.3 Trends or Transitions? ................................................ 179
6.4 Impediments at the Regional Scale ............................... 185
6.5 Comparing Results ................................................... 191
6.6 The Social Forces Hindering Land Use Change ............... 192
6.7 Accomplishments of Land Reform ............................... 193
6.8 Synthesis: The CPA, the Landscape, and the Geography
of Livelihoods ......................................................... 194
6.9 Summary ................................................................ 197
CHAPTER SEVEN
CONCLUSION ............................................................................... 199
7.1 Contribution to Theory and Method .............................. 201
7.2 Limitations of This Study and Future Research ............... 209
7.3 Summary ................................................................ 211
MAPS ...................................................................................... 214
BIBLIOGRAPHY ............................................................................ 241

vii

LIST OF TABLES

Table 1 : Livelihood Activities in Rural South Africa .......................... 22
Table 2: Summary of the Types of Tenure in Rural South Africa ......... 36
Table 3 : Options in Land Reform .................................................. 48
Table 4 : Specific Study Sites in the Northern Province ....................... 98

Table 5 : The Range of Livelihood Strategies and Relative Importance... 106

Table 6 : Selected Statistics ............................................................ 117
Table 7: Selected Statistics ............................................................ 118
Table 8 : Relationships in the Data ................................................... 119
Table 9: Detailed Scene Properties ................................................... 121
Table 10: Error Matrix .................................................................. 126

Table 11: Percentage Land Area Conversion from 1988/89 to 2000 by

Land Use / Cover Code ................................................... 128

LIST OF FIGURES

Figure 1 : Location of Study Sites in the Northern Province .................. 214
Figure 2 : False Color Image of Mahlambandlovu CPA 2000 ................. 215
Figure 3 : Land Use on Mahlambandlovu CPA 2000 ........................... 216
Figure 4 : False Color Image of Mahlambandlovu CPA 1989 ................. 217
Figure 5 : Land Use on Mahlambandlovu CPA 1989 ........................... 218
Figure 6 : Land Use Change on Mahlambandlovu CPA 1989-2000 ......... 219
Figure 7: False Color Image of Monyamane CPA 2000 ........................ 220
Figure 8 : Land Use on Monyamane CPA 2000 ................................... 221
Figure 9 : False Color Image of Monyamane CPA 1988 ........................ 222
Figure 10 : Land Use on Monyamane CPA 1988 ................................. 223
Figure 11 : Land Use Change on Monyamane CPA 1988-2000 .............. 224
Figure 12 : False Color Image of Rondebosch CPA 2000 ...................... 225
Figure 13 : Land Use on Rondebosch CPA 2000 ................................. 226
Figure 14 : False Color Image of Rondebosch CPA 1988 ...................... 227
Figure 15 : Land Use on Rondebosch CPA 1988 ................................. 228
Figure 16 : Land Use Change on Rondebosch CPA 1988-2000 .............. 229
Figure 17: False Color Image of Muyingiseri CPA 2000 ..................... 230
Figure 18 : Land Use on Muyingiseri CPA 2000 ................................ 231
Figure 19: False Color Image of Muyingiseri CPA 1988 ...................... 232

ix

Figure 20 :
Figure 21 :
Figure 22 :
Figure 23 :
Figure 24 :
Figure 25 :
Figure 26 :

Figure27:

Land Use on Muyingiseri CPA 1988 ................................. 233
Land Use Change on Muyingiseri CPA 1988-2000 ............... 234
False Color Image of Control Farm 2000 .......................... 235
Land Use on Control Farm 2000 ...................................... 236
False Color Image of Control Farm 1988 ........................... 237
Land Use on Control Farm 1988 ..................................... 238
Land Use Change on Control Farm 1988-2000 239

Difference between On-Screen Digitizing and Unsupervised

Classification ............................................................ .. 240

LIST OF TERMS AND GEOGRAPHIC NAMES

Africans- black indigenous people (non-white, non-Indian, non-coloured)

Apartheid- the legal separation and discrimination based on race. Dates
effective: 1949-1994.

Apartheid South Africa - the Republic of South Africa under minority
white control that affected, codified, and enforced racially discriminatory acts
and actions against the majority black population.

Apartheid state - the institutions of governance during the apartheid era.

Bophuthatswana- the homeland for the Tswana people under apartheid
located in northwest South Africa along the border with Botswana. One of the
four so-called 'independent’ republics.

Control Farm- the group of residents in a rural area chosen for interviewing
due to the nature of their farming system. The answers of this group was meant
to provide a control, or check, against the answers from members of Communal
Property Associations.

Communal Property Association (CPA) - the program under which land
was redistributed from 1994-2000. Refers to the entity consisting of transferred
farmland and any capital transferred with the land.

Department of Land Affairs IDLAj- the state agency charged with land

reform. Part of the Ministry of Agriculture and Land Affairs.

Escarpment - the physiographic region in the east center of the Northern
Province characterized as the border between the lower plateau to the east and
the higher plateau to the west.

Gazankulu- the homeland of the Gaza people under apartheid located in
the east of the Northern Province.

Gauteng - the province that encompasses the major urban core of South
Africa including Johannesburg, Soweto, Kempton Park, Germiston, Pretoria, and
Midrand. Adjacent to the Northern Province on the south.

Homeland- an area designated for blacks under apartheid. Homelands
were ethnically constituted. Three major political groupings developed: 1) the so-
called ’independent’ homelands, the self-governing homelands, and the
dependent homelands. The degree of ’independence’ from the Republic of South
Africa differed across the groups, but the effective control of the affairs of all
homelands lay in Pretoria at all times.

Highveld - the area to the west of the escarpment region in the Northern
Province. Characterized as higher in elevation and drier.

L322!!! - the homeland of the Northern Sotho people under apartheid.
Located in two discontinuous large blocks in the center and small units scattered
through the east of the Northern Province.

Lowveld- the area to the east of the escarpment region in the Northern
Province. Characterized as lower in elevation with more humidity than the

highveld.

xii

Northern Transvaal - anecdotal reference to the northern part of the old
Transvaal province. Not a distinct province before 1994.

Transvaal- the northern most of South Africa’s four provinces until 1994.
The province included present-day Mpumalanga, Gauteng, and Northern
Provinces and half of the Northwest Province.

Union of South Africa- South Africa under English settler control loyal to
the crown. Became the Republic of South Africa after 1949 when the Afrikaner
white minority gained control of government.

m- the homeland of the Venda people under apartheid. Located in the
far northeast of the Northern Province. One of the so-called ’independent’

republics.

xiii

Chapter One

Introduction to the Dissertation

Land reform, rural poverty alleviation, and social transformation were
intended to go hand-in-hand in post-liberation South Africa. Land reform since
1994 has included land restitution, land reform, and land tenure reform. Under
land restitution, communities that were illegally removed from their homes,
communities, and land under apartheid era law can petition the government for
restitution of, or compensation for their property. Land tenure reform seeks to
end the widespread chaos in tenure arrangements created under the apartheid
system. Multiple and overlapping tenure systems are being simplified and
codified. The transfer of land formerly owned by whites through a willing-buyer,
willing-seller program to blacks, called land reform, is the focus of this
dissertation. I examine the impact of the Communal Property Association
program - the land reform delivery system - on the livelihoods of the members
and land use change on the transferred land.

The research questions raised in this dissertation stem from the
introduction of six land reform projects into six different rural communities,
namely:

1) What are the specific responses of rural livelihood systems to such a

change - how did land reform effect in situ livelihood systems?

2) Is change on the transferred lands evident in land cover and land use
patterns - can the social driving forces of such change be identified from
the remotely sensed images?

3) Did land reform result in the expected changes in both livelihood
systems and land use change?

These research questions are drawn from three broad and interrelated
theoretical bodies of work, specifically livelihood systems, land cover and use
change, and land reform. Throughout this work, two themes repeat and provide
pivotal insight into the story of land reform, livelihoods, and land use change,
specifically gender relations and issues of scale. Additionally, I draw upon the
literature on resettlement and cooperatives to address issues of time and

transition.

Livelihood Systems

livelihood systems are explored in this study and are contextualized as the
mechanisms that people pursue to ensure social reproduction, garner income,
and meet the various obligations of self and society. Livelihood systems in South
Africa are diverse and dynamic. For clarity and precision I preface the term
’livelihood’ in this dissertation with the adjective ’rural’ whether visibly printed
or not. This dissertation will not explore urban or pen-urban livelihoods.

Livelihoods in rural South Africa can be understood in the broader African

context, yet retain locally specific tendencies. Livelihoods throughout the

continent have increasingly incorporated non-farm and off-farm activities
(Bryceson, 1997). This is no more so the case than in the Northern Province of
South Africa, the study region. In fact, non- and off-farm employment constitute
the core of rural livelihoods in the Northern Province (Baber, 1996). Migration,
petty commodity production, and employment in the civil service are significant
contributors to rural livelihoods.

Agriculture is in transition in many livelihood systems across Africa
(Bryceson, 1997) and the study area is no exception to that trend. Agriculture
remains an important cultural icon and provides subsistence for many
households, but is rapidly becoming a less important source of actual household

income.

Land Use and Cover Change
Land cover and use change is defined in this study as the temporal shifts in

vegetative cover and human uses for a given parcel of land. The hypothesis here
is that land use will have intensified on transferred lands. To reject or support
this claim, land cover patterns are identified via satellite imagery and interpreted
to derive land use for two periods, 1988/ 9 and 2000. The two periods are then
compared to understand patterns of Change on and around the six land reform
projects.

In this study, land use / cover change provides a contextual basis for

analysis of the impact of land reform on the landscape. Land use and cover

change, or the lack thereof, can be related to the social processes driving
livelihood change and visa-versa. The scope and spatiality of change illustrated
in such an exercise broadens the range of theoretical questions. Further,
understanding land use change in this manner helps provide clarity and
verification or rejection of both hypotheses and information garnered through
less explicitly spatial methods such as interview schedules and informal

discussions.

Land Reform

Defined here as the transfer of previously white-owned land to black
communities, land reform constitutes the core of this study in that the land
reform event precipitated the changes under examination in this dissertation.
Land reform is narrowed from its legal definition in South Africa. The
Department of Land Affairs defines land reform as the transfer of land from
whites to any individual or group of black citizens (Dl.A, 1995). I exclude transfers
between individuals in this study as the dynamic of social transformation in such
transfers is far more limited than those that involve individual to community
transactions.

The underlying theme of this dissertation is focused on land reform,
begging the question - has it worked? I contextualize the land reform program in
South Africa against counterparts across the world, measuring progress

temporally to provide points of comparison and departure. The resettlement and

cooperatives literature is examined to address issues of transition and the timing
of this investigation in the project life cycle. I address the question ” how long
does it take for livelihoods to adapt to a land reform project?" in order to gauge
the results of the study, either as trends or transitions. I then compare other case
studies with the South African land reform project - the Communal Property
Association (CPA). The CPA program, operating from 1996-2000, facilitated land
transfers by pooling the resources of black communities in order to purchase
farms that would have been otherwise unobtainable. The program was designed
to provide start-up and program capital as well as technical assistance and a
legal operating framework. Using the data collected during fieldwork and
drawing on the literature, I analyze the progress of the Communal Property
Associations at their fourth anniversary and relate the findings to livelihoods and

land use change, seeking clear and scientifically defendable relationships.

Dissertation Outline

The next chapter sets a theoretical basis for the discussion of livelihoods,
land use change, and land reform in rural South Africa. I review the relevant
literature regarding each as well as the important contributing themes of gender
relations, scale, and resettlement. The contributing themes, in many instances,
shed as much light into the dissertation as the so-called ’main themes’. In

discussions, I have chosen to integrate the literatures and themes to present a

more holistic, rather than rigid systematic, analysis. The major assumptions are
stated and the research questions are then raised.

The context of the study is set in Chapter Three. I provide background
information on the Northern Province and the specific study sites. I employ
government census data to provide a broad understanding of livelihoods in the
province and describe the progress of land reform.

I lead into the core of this work with the presentation of the data in Chapter
Four. I provide information on the methods utilized and then segment and
discuss each. I start with a review of the interview schedule data, focusing on
income generation, food shortages, coping mechanisms, the constituents of
livelihood systems, the systems of land use and authority, and members’
perceptions of the CPA. I then center one of the recurrent themes, gender, and
gather together the key findings. Statistical relationships in the data are next
explored. Finally, I describe the method of analysis for the identification of land
cover and use change. I show the scope of change for each study site and ponder
the social driving forces of change. I specifically attempt to link social process
and satellite-derived interpretations of change.

In Chapter Five, I address the research questions presented in chapter two
providing explanations of the key findings in the context of the larger literature
and the local environment. An evaluation of the impacts of the land reform
projects on livelihoods is undertaken and a discussion of the utility of satellite-

derived change analysis follows.

I construct a narrative regarding the relationship between the land reform
event and livelihoods and land use change in Chapter Six. In this chapter I
attempt to provide explanation for the ”unfinished transformation” by
specifically grounding my arguments within the local context and the relevant
literatures. This chapter addresses the questions

1) Has land reform delivered the anticipated level of rural poverty

alleviation?

2) What were the reasons for success or failure at the farm level?

3) What were the reasons for success or failure at the regional scale?

4) Are the processes identified trends or symptoms of transition?; and

5) What were the expectations and what could be expected based on

examples from the literature?

I' conclude in Chapter Seven by locating the study in the literature,
identifying its relevance to the discipline and science in general, and posing
questions for future research. I identify the assumptions and limitations of the

study and provide recommendations.

Chapter Two

Review of the Literature

2.1 Introduction

This review of the literature brings together three fields of study that have
been only weakly linked together. The relationship between these three topics is
central to this dissertation. I will attempt to show the complementarity of the
livelihoods, land use change, and land reform literatures in order to ground the
findings of this study presented in the following chapters.

I begin each section with an overview of the dominant themes in the
general literature. I then emphasize the topics most pertinent to this study and
re-examine them focusing on the study area - rural South Africa. When
reviewing the literature on livelihood systems I pay particular attention to the
definition of livelihoods, livelihood strategies, the construction of the term
’household’, gendered relations, and the literature covering the above issues that
has particular relevance to South Africa. Regarding land use and land cover
change, I situate the broad literature and focus specifically on scale in land use
change studies, examples of broad relevance, and case studies and issues
pertaining to South Africa. The discussion on land reform encompasses land

alienation issues, land tenure types and land tenure reform, and land reform

choices and policy direction. This chapter concludes with a synthesis and critique

of these three literatures.

2.2 Contextualizing Rural Livelihoods

The term ”livelihoods” has engendered a lively debate between
theoreticians examining livelihoods either as a function of household economics
or those that view household livelihoods as a social construct, derived from
dialogue and dialectic between multiple scales, structures, and actors. The rift
between the two competing ideologies has narrowed recently (Ellis 1993) and the
resulting productive discourse has been reflected in a more integrated, although
still somewhat disconnected literature on livelihood systems. In an attempt to be
both conceptually rigorous and thorough, I will first examine the meaning of the
term ”livelihood" and will follow with a review of the various arguments on the
composition of livelihood systems. Finally, I will present the literature relevant to
livelihood in South Africa, with a few specific examples from the Northern
Province. It must be noted here that the focus of this study is on rural livelihoods,
therefore the bulk of the literature discussed here refers to rural rather than

urban or peri-urban livelihood systems.

2.2.1 Definition, Composition, and Debate
Defining the term ”livelihood”, specifically rural ”livelihood”, is a process

facilitated by a wealth of literature on the topic (see Bernstein, et.al., 1992). I wish

to be clear here and first separate constituent components of the term to avoid
lateral confusion between ”livelihood” and other terms, most notably ”coping
mechanisms”, “food security”, ”vulnerability” and ”agriculture/farming”. This
study specifically addresses issues of livelihood, of which the terms coping
mechanisms, food security, vulnerability, and agriculture/ farming are a part of
or closely related to, but not the same as ”livelihood”. A coping mechanism
occurs when a household or individual’s livelihood system has been disrupted in
some way and an alternative path is sought to overcome hardship. Food security
refers to the ability of a household or individual to obtain enough food to
maintain basic life functions. Vulnerability refers to how likely a household or
individual is to suffer hardship due to a change in livelihood. Agriculture is a
component of livelihood, not to be equated with the term livelihood.

Definitions are usually contested and debated and the definition of
”livelihood” is no exception. However, opinions have gathered around a core
idea that refers to livelihood as the ”the means of securing a living” (Chambers
and Conway 1992). Ellis (1998) provides a broad perspective on livelihood noting
that ”livelihood is more than just income” and that ”a livelihood also includes
access to, and benefits derived from, social and public services provided by the
state" (4). ”Livelihoods” can be thought of either as static, a snapshot in time, or
as a process, where change will strengthen or weaken households over time, but

not always in a linear fashion. The concern of this study is investigating whether

or not the diversification of livelihoods, what Ellis (1998) defines as ”the process

10

by which rural families construct a diverse portfolio of activities and social
support capabilities in their struggle for survival and in order to improve their
standards of living”, has occurred among study participants. In this respect,
livelihoods in this study are defined as the composite of multiple processes
operating through time rather than as snapshots in time. Ellis clarifies by noting
that ”livelihood diversification is not synonymous with income diversification”,
the latter referring to transfers of cash or credit whilst the former may encompass
non-income transfers, for instance, access to a free clinic. May (1999) identifies
eight ”livelihood activities” particularly relevant to this study:

1. Agriculture - in any form
2. Small and Micro Enterprise - for South Africa, the ”spaza” shop and
”hawking” are most important

3. Wage Labor- often cited as the single most important source of income

4. Claiming Against the State - primarily pensions

5. Claiming Against Household and Community Members - remittances
constitute the largest single claim made against relatives, but other forms
can be evidenced

. Unpaid Domestic labor - usually women’s work

. Illegitimate Activities - illegal or socially unacceptable activities

. Non-Income Earning Activities- activities that do not generate direct
income, but afford households greater livelihood flexibility, for example
participation in civic organizations.

NVO\

Chambers and Conway (1992) describe the ”provisional anatomy” of
household livelihood in four parts. First are the people and their livelihood
capabilities, second are the activities, or what they do, third are the assets,
tangible and intangible, and fourth are the gains or outputs, what they obtain

from what they do (paraphrased from Chambers and Conway, 1992, 9).

11

Just as the definition of the term ”livelihood” is contested, so is the
definition of ”household”, the unit of analysis for this study. Guyer and Peters
(1987) describe the household:

The household is of variable structure; is both outcome and channel of
broader social processes, and is the site of separable, often competing,
interests, rights and responsibilities. It is as much a ’segmented’ unit as are
labor markets segmented by gender, age, clan, ethnicity and so on.
Moreover the ideological construction of the household, the range of
cultural meanings attributed to domestic units, conjugal and age relation
and residential patterns are also critical to a fuller understanding of the
dynamics of production and consumption that generate the diverse social
units we call households ( 210-211).

Martin and Beittel (1987) position households in relation to labor markets,
arguing that households are ”constituted by a small group, the household is the
unit that ensures the continued reproduction of labour through organizing the
consumption of a collective fund of material goods - a unit therefore different
from the family, co-resident dwelling groups, and kinship structures” (218).

With relevance to South Africa, May, et.al., (1994) summarize concerns over
the use of the term ’household’:

The boundaries of the household, the division of power within the
household, and the division of labour within the household, and the
dynamic nature of households are all issues about which reductionism [sic]
assumptions have been made, and this confusion of the nature of the
household has led to misleading conclusions regarding vulnerability in
rural areas. As an example, the popular use of the household as the
standard unit of analysis in surveys has lead to an unwarranted attention
on female headed households, being a readily identifiable means of
introducing gender into analysis. However, this distracts attention away
from the circumstances of the majority of women in rural areas who live in
households in which a male is the de jure head. Moreover, while the
orthodox notion of the household assumes a household structure that is
applied consistently in all places and at all times, empirical studies would

12

suggest otherwise and the structure and dynamics of households and intra
and extra household capabilities have been found to be more diverse and
complex (8).

May, et.al. argue that the household is better defined in three parts, ”fire
boundaries of the household, the division of power within the household, and
the division of labor within the household” (1994, 8). This alternative definition
of the household is useful in a South African context, with family units being
torn apart historically due to labour migration and recently due to urbanization.

Several authors have differentiated households, primarily by class, into
broad categories. Carter and May (1999) have divided households by livelihood
type into eight classes, namely:

Marginalized

Welfare Dependent

Remittance Dependent

Secondary Wage Dependent

Primary Wage Dependent

Mixed Income with Secondary Wages

Mixed Income with Primary Wages
Entrepreneurial

FNF‘P‘PSPNE"

They describe marginalized households as those with very little actual
income and no cash transfers. Welfare dependent households are those firat rely
on some form of transfer from the state or welfare organizations. In South Africa
this includes old age pensions. Remittance dependent households, common in
South Africa, are those first rely on the money sent back from a relative with
employment elsewhere. Secondary and primary wage dependent households are

those that rely on a family member's informal or formal wage, respectively.

13

Households firat combine several livelihood strategies but are dependent on
eifirer informal or formal wages are characterized as mixed income and are
differentiated by fire type of employment. Finally, entrepreneurial households
are first excel due to agricultural production and / or business. (Carter and May

1999). Households in fire latter categories tend to be more prosperous.

2.2.2 Gender

What is fire contribution of gender to a project on livelihoods and land use
change in rural Soufir Africa? Ngqaleni and Makhura (1996) illustrate a very
basic and relevant point. They note that women perform fully seventy percent of
fire labor on small-scale farms in the Northern Province. Second, the
phenomenon of male out-migration to fire mines of northwest South Africa is
perhaps most pronounced in fire Northern Province, leaving women to tend
farms and maintain the ’rural’ livelihood. Third, at a more theoretical, and
potentially more useful level, fire conceptualization of gender and fire household
is strongly challenged in rural South Africa. Households, particularly firose
termed ”female-headed" do not always match the theoretical expectations of
even fire most careful scholars on rural livelihoods and gender. Most
importanfiy, however, this study explorer and uncovers key differences among
and between women and men. While caution is exercised in the construction of

the household, the relevance of fire broader gender literature to this study is

14

reflected firrough fire ”gender and development” and post-modem feminist
approaches that focus on difference.

Not only have studies of gender moved from women’s role to a focus on
gender relations (Joekes, 1996; Razavi and Miller, 1995), but feminist writers are
also reconceptualizing fire household as a relevant framework for assessing such

relationships. Households are largely constructs of economic theory, often
overlooking the relations that individuals have with ofirer individuals, actors and
institutions in society, and with nature. All too often households are constructs of
convenience. While rural studies have long acknowledged women and their
contribution to and in the household, fire notion of the household in relation to
gender as fire ideal unit of analysis is problematic on another level. Peters
recognizes several key problems wifir fire use of fire household, particularly fire
”female headed household". Though useful in dis-aggregating fire household,
fire notion of ”female-headed households” fails to ”take into account fire
significant differences between, for example, male-headed households and
female-headed households or between small male farmers and small female
farmers” (Peters, 1995,95).

Furfirer, fire notion of female headed household blurs the similarities
between firese categories. Peters notes firree major problems with fire use of fire
term ’female-headed household’ First, female-headed is often equated with

impoverished. Second, household structure is not fire only consideration when

15

examining gender relations, and firird, gender relations and relations with actors
and institutions outside of fire household are obscured.

Gendered analysis of fire household, thus, represents an attempt both to
challenge conventional notions of fire household as one unit and to reexamine
fire perceived notions of ’female-headed households’ as a category of household.
This understanding is particularly potent in rural Soufir Africa where households
have typically centered on women, but where gender relations firat influence
livelihoods extend well beyond the boundaries of fire household.

Thinking on women and gender has shifted through several
conceptualizations in fire past decades (Joekes, et.al, 1996; Razavi and Miller,
1995). The role of gender in development shifted away from the early
conceptualization of women’s roles in the household and society to the
incorporation of women in fire development process, or WID. Advocates of a
WID approach argued fire bringing women into the development process would
necessarily improve bofir fire process and women’s lives. An alternative
approach, gender and development (GAD), maintains firat simply incorporating
women into development is not sufficient. Rafirer, an examination of gender
relations is critical in understanding the dialectic between gender, not just
women, and development. The GAD approach emphasizes investigation into fire

differences between and among both men and women. Young writes:

16

A focus on women alone was inadequate to understand fire opportunities
for women for agency or change; firat women are not a homogenous
category but are divided by class, colour, and creed; firat any analysis or
social organization and social process has to take into account fire structure
and dynamic of gender relations; firat fire totality of women’s and men’s
lives has to be fire focus of analysis, not merely their productive, or their
reproduction activities that women are not passive, nor marginal, but active
subjects of social processes (Young, 1993, p.134).

Young furfirer notes firat fire sexual division of labor is important as it
”constructs women in relation to men - as their inferiors, fireir chattels, fireir
helpmates, or fireir equals” (Young, 1993, p.141). The sexual division of labor ”is
a critical element in maintaining and/ or recreating gender inequality” (p.141)
and must be done away wifir.

The concepts of ”gender and development” and ”post-modern feminism”
deconstruct fire notion of ”women” as a homogenous unit. Moser oufiines the
basic conceptual shift away from a ”women in development” approach to a
”gender and development” approach:

Approaches to issues relating to women in developing countries became
concerned with fire manner in which gender and concomitant relationships
were socially constructed. The focus on gender rafirer firan women makes it
critical to look not only at fire category ’women’ - since firat is only half the
story- but at women in relation to men, and the way in which relations
between these categories are socially constructed. Men and women play
different roles in society wifir fireir gender differences shaped by
ideological, historical, religious, efirnic, economic and cultural
determinants. These roles show similarities and differences between other
social categories such as class, race, ethnicity, and so on. Since the way firat
are socially constructed is always temporally and spatially specific, gender
divisions cannot be read off on checklists. Social categories, therefore,
differentiate fire experience of inequality and subordination wifirin societies
(Maser, 1993, p.3).

17

Moser furfirer differentiates between fire WID and GAD approach noting
firat fire WID assumes the development process would be facilitated by simply
incorporating women while fire GAD approach argues for an examination of
gender relations in development.

Jackson illustrates differentiation among women regarding environmental
degradation noting:

Women as a group do not experience environmental degradation in a

uniform manner - firese effects are mediated by fire livelihood system. Some

women may have remittances from migrant males, or more diverse
livelihoods, or assets that may be liquidated, or kin0based entifiements

which ameliorate fire effects of environmental degradation (Jackson, 1993,

p.1949).

In addition to fire community level, differences exist between women at the
household scale. Jackson argues firat fire relations between mothers and
daughters, junior and senior wives exacerbate the differences already present,
such as class (Jackson, 1993,p.1949). Furfirer, she explains how gender relations
and differences play out across environmental knowledges, property relations,
space, and intrahousehold dynamics.

Empirical examples of fire importance of examining differences among and
between men and women abound. Sylvester (1995)identified differences among
women in a rural producer’s group in Zimbabwe, while Thomas-Slayter
identified cleavages among women in a Kenyan community dependent on a

local natural resource.

18

A post-modernist feminist approach to fire study of gender and gender
relations, however, calls into question some of the central tenants of a GAD
approach. Marchand and Parpart (1995) critique GAD arguing firat it is rarely
incorporated into development planning and continues to characterize women as
vulnerable (p.15) . In addition, a GAD approach essentializes women. They note
first ”gender does not constitute a super-ordinate category, but takes precedence
only wifirin certain situations” (Marchand and Parpart, 1995, p.173).

Finally, central to firis study is fire feminist political ecology approach to the
study of gender. A feminist political ecology approach, writes Rocheleau:

Begins with fire concern of fire political ecologists who emphasize

decision-making processes and fire social, political, and economic context

firat shapes environmental policies and practices. . .Femirrist political
ecology treats gender as a critical variable in shaping resource access and
control, interacting wifir class, caste, race, culture, and efirnicity to shape
fire processes of ecological Change, fire struggle of men and women to
sustain ecologically viable livelihoods, and the prospects of any

community for ’sustainable development (Rocheleau, 1996, 4).

Rocheleau highlights three important areas of investigation wifirin fire
framework, particularly gendered knowledge, gendered environmental rights
and responsibilities, and gendered environmental politics and grassroots
activism. Gender is included in fire tapestry of forces that shape decisions across
scale, taking political ecology into feminist political ecology.

The literature on gender difference is particularly important to firis study

where class, ethnicity, and access to resources have all been factors firat have

long differentiated rural Soufir Africans, including women. Access and control

19

over resources, livelihood systems, and food security are not issues that can be
dichotomized between men and women in rural Soufir Africa. The impact of
differential development, apartheid, and fire creation and destruction of
”traditional” cultural practices and habits are forces firat have led to pronounced
cleavages among all groups, not just gender groups, in South Africa. The
examination of fireses differences will provided clearer insight into the impact of

fire land reform projects on gender relations.

2.2.3 Research Heritage

As noted above, agriculture is not equivalent to livelihoods, but it is
important to the livelihood strategies of people in rural Soufir Africa. As in much
of fire rest of rural Africa, livelihood strategies no longer depend solely on
agriculture as fire main source of income. While agriculture remains important, it
can not stand alone in an analysis of rural livelihoods. Baber (1996) in two case
studies of fire Norfirern Province showed the significance of remittances, non-
farm activity, and pensions to fire livelihoods of households. While agriculture
remained an important source of income (in fact, the most important in one
study), ofirer activities were essential to household survival.

The literature on rural livelihoods consistenfiy recognizes fire importance of
extra-agricultural sources of income (Bernstein, et al. 1992; Bernstein 1994;
Reardon 1997; Lipton 1996; Lipton 1993; Chopak 1991; Davies 1993; Cabre-

Madhin and Reardon 1989; Haggblade 1989; VanZyl, et al. 1991). Non-farm

20

income in firis study is defined as " income from local non-farm wage
employment, local non-farm self employment, and migration income" (Reardon
1997, 4). Reardon includes migration income as a component of off-farm income,
however, several ofirer authors exclude migration income, preferring to analyze
it separately (Baber 1996; May 1996). Remittances from labor migration are
essential to livelihoods in the Norfirern Province (Baber 1996; de Villiers 1995).
Lemon (1987; 1995) describes the structural factors that led to the high level of
dependence on migration for income, primarily the forced segregation into
homelands. Baber (1996) shows the importance of remittances to household
livelihoods in fire Norfirern Province. The former lebowa homeland was
established primarily as a labor reserve and ”dumping” ground for Soufir Africa's
industrialized norfirwest (Smith 1990).

Table 1 shows fire distribution of livelihood activities in fire rural areas of
Soufir Africa. May points out firat, historically, poverty in South Africa stemmed
from an ”erosion of fire asset base” due to fire forced Bantustanization of fire
population, fire ”impact of a disabling state”, referring to the aparfireid
government, and fire ”direct impact of policies that incorporated gender and

racial discrimination” (May 1999, 12).

21

Table 1 : Livelihood Activities in Rural Soufir Africa

 

 

 

 

 

Activity “/o Households Engaged
Agricultural Production 36.4
Small and Micro- Enterprises 10.4
Wage Labour in fire Primary 22.1
Age Labour in fire Secondary 37.4
Claims Against Household 39.0
Claims Aginst fire State 32.4

 

Source: May, 1999

2.24 Discussion

The literature on livelihoods has evoked several firemes pertinent to firis
study. First, fire review of fire contested nature of the definition of ’livelihood’
and ’household’ has produced a clearer understanding of these terms and has
addressed fire issue of how they will be used in this study. Lateral confusion
wifir food security, coping mechanisms, and ofirer terms has been avoided, thus
narrowing fire scope and allowing for more focus on the topic of livelihood. The
”household” as a unit of analysis and conceptual construct has been reviewed
with attention paid to gender and households. The introduction of gender into
fire discussion on households showed fire fractured nature of fire construct. In a
society such as Soufir Africa, where tradition has been invented and change is
constant, challenging the assumptions on household typology, especially with
relation to gender, will help to avoid overgeneralizations and misinterpretations.
Finally, deconstructing fire concepts in a South African framework (evidence

May) and understanding fire dynamics of bofir livelihood and household will

22

help provide a strong foundation for analysis of data collected about livelihoods

at fire household unit in rural South Africa.

2.3 Land Use and Cover Change

2.3.1 Research Heritage

The conceptualization of human relationships wifir land is a long held
tradition in geography (Sauer 1925; Barrows 1923). Recently, Blaike and
Brookfield (1987) and Feet and Watts (1996) have widened perceptions of firis
relationship by challenging scientists to understand social meanings behind land
and fire environment. Land relationships are not only limited to "use" and
"management", but also to a series of complex interactions among and between
actors and institutions. Evidence of firese processes is abundant (Liverman 1990;
Olson 1990; Moore 1993; Neocosmos 1993; Vogel 1993; Levin and Weiner 1997).
Recognition of land use change as a fundamental element of global change
studies has prompted calls for a greater understanding of impacts firat humans
induce on their environments (Ojima, et a1. 1994; Houghton 1994; Turner, et al.
1993, Meyer and Turner 1994).

The relationship between socio-economic drivers and resultant land use
change, including positive and negative feedback needs close examination. The
dynamics of scale and fire relationship between socio-economic drivers has been

poorly understood and only vaguely conceptualized, alfirough much progress

has been made in recent years (Ojima, et al. 1994; Turner, et a1. 1993; Skole, et a1.

1994; Olson 1990; Liverman 1990; Showers 1994; Snyder 1996).

2.3.2 Theories on Land Use

While excellent at providing descriptive detail of land use and cover change
and identifying broad driving forces, global change studies have generally failed
to provide a mechanism for fire study of fire interaction of people, institutions,
and other elements central to an analysis of bofir livelihoods and land use
change. Complementing land use and cover change/ global change studies,
political ecology provides a conceptual framework and fire mechanisms of
analysis needed to critically evaluate the intersection of livelihoods and land use
change in a complex setting embedded with issues of gender, class, and power,
such as rural South Africa.

Political ecologists are particularly adept at forming understandings of the
structures that influence fire relationship between people and land, especially in
terms of scale, history, fire state, and local social relations of production (Bassett
1988). Blaike (1994), Blaike and Brookfield (1987), Watts (1983), and Campbell
(19%; 1990) describe political ecology as a framework proficient at dealing with
fire many social, political, economic, and environmental forces that land users
must face. Not only must institutions, such as the state, be forefront in any

explanation, but scale must also be examined. The state is an institution wifir

24

strong influence from fire international to local scales and it impacts local
processes in a variety of ways (Blaike 1994).

However, political ecologists have provided less than adequate
explanations of fire more contextual issues such as gendered space (Rocheleau
1996), perceptions of the environment, and cultural symbolism. Analyses based
on gender (Rocheleau 1996) and fire notion of fire social construction of nature
(Escobar 1996) provide a more complete understanding of fire interaction
between humans and fire environment, and provide a much needed contextual
aspect to more conventional political ecology analyses. Ecology itself is changing,
incorporating chaos and disequilibria, and political ecologists, particularly
geographers, can benefit by incorporating these changes (Zimmerer 1994).

Political ecologists also tend to fail to operate beyond specific case studies.
Few attempts have been made to present a global perspective wifirin fire political
ecology framework. The global scale is represented as an input or output factor;
rarely are findings generalized in such a way that a global political ecology could
be constructed. Alfirough scale is considered core to political ecologists
(Campbell 1998), fire vast majority of case studies operate at the household or
community level. Fewer studies integrate regional analysis beyond a spatially
small area, and even fewer attempt to paint a global picture. The global scale
simply becomes a feedback loop in fire system.

This broad body of theory shapes and informs firis study by providing

conceptual categories for operationalizing land use change and livelihood in

25

rural Soufir Africa. I will draw on firese fireories and broad bodies of work to
elicit informed conclusions on fire processes of land use change and fireir

relationships to livelihood systems.

2.3.3 Scale

Scale in fire study of land use and land cover change is a salient, yet
unresolved issue. A political ecology approach could be useful if strong
relationships between global and local processes were readily manifest. One of
fire central difficulties in addressing scale in land use and cover change analysis
is fire use of data from different scales. Political ecology tends not to project
empirical evidence from case studies to fire global scale due to incompatible
research design across case studies.

To date, a global political ecology, one in which an attempt is made to
aggregate data collected at multiple local sites into an overarching theory or
model of land use change, has not materialized. To be sure, global change studies
have not progressed much furfirer either. Proximate variables, such as
population, have been used to attempt to fireorize and model land use change at
fire global scale, however, fire data are just as inconsistent and refutable as would
be an attempt by political ecologists to aggregate disparate case study data.

Meyer and Turner (1994) argue:

26

Though a global view is required for some purposes, a globally aggregate
one is insufficient for answering many pressing questions. The net
worldwide trajectories of land-cover change area rarely duplicated in any
region or locality. Consequently, explanations, forecasts, and prescriptions
developed only from global aggregate data area likely to be worse firan
useless when applied in sub-global units. Nor can adequate global
projections be developed from global aggregate data alone because global
tools represent aggregations of quite dissimilar world regions. (1994, 7)

While discounting fire global approach due to a lack of reproducibility in
regional and local scales, fire aufirors point out fire problems wifir fire local and
regional scale approach as it relates to land use and cover change studies:

Yet fire opposite extreme from fire global aggregate approach, a plefirora of

micro-studies highly attentive to local context and singularity, is equally

unsatisfactory given fire needs and constraints of fire global change research
program. A large literature of small-area studies does exist, and it offers
many insights into fire complexities of nature-society interactions. Practical
considerations, firough, prohibit the separate study for global modeling
purposes of every piece in fire world’s mosaic of environmental and

socioeconomic conditions. Nor could the results, even if collected in a

systematic and comparable way, necessarily be aggregated

unproblematically for higher scale of analysis (1994, 7).

Lambin (1992), describing desertification in Burkina Faso, remarks that ”an
analysis of fire environmental consequences of decision-making often requires a
broadening of geographical scale” (1992, 5). The argument here is firat upward
linkages in scale cannot be overlooked when conducting specific case studies.
Furfirer, Lambin writes ” an approach that employs a nested set of spatial scales
has proven to be appropriate to understand the behavior of land managers
responsible for desertification" (1992, 4). What all of fire authors advocate is a

goal rafirer firan a reality. In reality, many studies tend to focus on one scale of

27

analysis wifir fire ofirer scales being inputs and outputs. This study will be no
exception, however, as analysis at greater than fire local / regional scale has not
been attempted. I have purposefully investigated regional phenomenon in order
to present a broader interpretation of land use change in the study region, but
have refrained from drawing global implications for just the same reasons

illustrated above.

2.3.4 Examples and Case Studies

Several firemes resound in fire literature on land use and cover change,
whefirer fire example is global or local. Both empirical examples and theoretical
studies tend to focus on driving forces of change and scale issues. Furfirer, most
empirical studies rely on remote sensing to characterize fire scope of change.

The driving forces of land use change at fire global scale have received
much attention in fire literature. Population, affluence, and technology (PAT) as
driving forces have been central in several arguments (Sage 1994; Grubler 1994;
Bilsborrow and Okofir-Ogendo 1992; Heilig 1993; Meyer and Turner 1992) while
others have examined institutions (Sanderson 1994) and culture (Rockwell 1994).

Bilsborrow and Okoth-Ogendo (1992) examine fire link between population
growfir and land degradation in developing countries. They note three phases in
fire ”raponses of land-use practices to population growth" (38). Phase one
includes tenurial changes, such as accommodation, fragmentation, and

reclassification. An example of accommodation is fire use of underutilized land,

28

 

fragmentation refers to dividing existing parcels into smaller parcels, and
reclassification entails making land available ”to permanent or traditional
members of fire community at fire expense of ofirer members” (39).

Adjustment in fire second phase is affected through land extensificatiorr,
using land previously outside fire control of fire community. Finally, Bilsborrow
and Okoth-Ogendo argue firat new technologies will be sought to accommodate
increased population. The intensification of land use in response to increasing
population is most closely associated wifir fire work of Ester Boserup (1965;
1%1).

Meyer and Turner (1992) address several issues relevant to fire driving
forces of land use and cover change at fire global scale. They remark on three
points. First, ”fire driving forces of change may vary wifir fire type of change
involved” (1992, 51). Simply put, this means change from forest to agriculture
will not always be driven by fire same forces driving change from wetlands to
agriculture. Different types of change will have different driving forces. Second,
they argue firat ” fire same kind of land-cover change can have different sources
in different areas even wifirin particular world regions" (1992, 51). This reflects
fire problem of aggregation. Specific case studies wifirin a world region may
point to different forces of change, alfirough the expectation might be firat similar
areas experience similar forces. Finally, they point out that ” in fire dynamics of
underlying ’causes’, no agreement exists on the level at which adequate

explanation is achiev ” (1992, 51). The problem of scale is unresolved.

29

Several studies describe land use and cover changes at a broad regional
scale. Heiling (1997) identifies five factors driving land use change in China’s
East. Population increase, rural-to-urban migration, economic growfir, changes in
lifestyle, and changing ”economic and political arrangements” will drive future
change. Skole, et.al. (1994), examining deforestation in fire Amazon region of
Brazil, focus on population, but warn ”simple relationships to population growfir
may not alone describe factors driving deforestation in fire Amazon” (1994, 318).
Furfirer, firey argue firat ”economic and institutional factors” driving agricultural
expansion help explain deforestation.

While empirical evidence at the global scale is mired in debate over quality
and compatibility (Turner and Meyer 1994), regional and local scale data on fire
driving forces of change become more specific and are easier to quantify.
Because firis study does not attempt to address land use and cover change from a
global perspective, I will focus on fire regional and local examples of driving
forces.

Reid, et.al. (1999) highlight specific driving forces of change in an area of
central Efiriopia plagued by trypanosomiosis. The authors found firat a
combination of local forces (fire introduction of trypanosomiosis) and national
forces (resettlement policy; land redistribution) shaped land use change.
Lindblade, et.al. (1998) writing on southwestern Uganda indicate that the
perception of change among farmers had led to a real change in land use

patterns. Alfirough fire aufirors found firat population increases had not

30

substantively affected increased pressure on land, the perception of population
pressure had led firem to reduce fallowing. This case study is particularly
relevant to fire dissertation as perceptional data lays the foundation for
understanding fire social forces most responsible for change.

In a case study of Iraqw farmers in norfir central Tanzania, Snyder (1996)
identifies out-migration, soil fertility, pests, and fire introduction of agro-forestry
as forces of change, but focuses on fire relationship between labor and forestry.
She argues firat ”farmers in the Iraqw homeland are choosing to invest in less
labor and capital-intensive agro-forestry practices rafirer firan increasing fireir
investment in food crop or cash crop production” (336).

Walker, et.al. (1996) in an Amazonian study at the household level found
that fire strongest driving forces of change ”emerge on fire basis of bofir domestic
cycles and fire exercise of social power by privileged groups” (76). The aufirors
furfirer note, ” it is impossible to separate fire ecological problem of deforestation
from fire social issues of land concentration, rural violence, and frontier
development” (77). In a similar study, Scatena, et.al. (1996) discuss cropping and
fallowing sequences as an important household and community driving force of
change. These studies of specific social processes at the local scale relate human

action to landscape change and are, firerefore, useful to firis study.

31

2.3.5 land Use Orange in South Africa

Land use and cover change in Soufir Africa is neifirer unique nor more
important firan in ofirer world regions. I will not try to set land use in Soufir
Africa as somehow special or different. I will argue, rather, firat Soufir Africa is
representative of a broad range of land use and cover changes. I posit here firat two
types of land use and cover change are evident in fire Soufir African landscape.
First is structured. Structure is firat which constrains, such as firrough a legal
system or societal norms (Giddens 1984). This represents known, readily
identifiable and predictable changes manifest largely in land use plans and
management documents. Timber plantations are the largest contingent of such
change in fire norfireast Soufir Africa. The scope and spatiality of such change is
known in timber company documents, town and regional land use plans, etc.
The second type of change, and fire focus of firis study, can be described as
agentist change. Agency is fire freedom individuals have to act upon fireir own
will (Giddens 1984). This category represents less structured change- spurred by
individuals and groups in society where outcomes are neither easily measured
nor planned and often come into direct conflict with structured forms of land
use.

Change in Soufir Africa is neither exclusively structuralist or agentist. That
fire Soufir African state (a structure) seeks to alter fire landscape through land

reform makes fire country a good case study. The state as a force of change is

32

clear and can be evaluated and analyzed as a factor of total change. Institutions
in Soufir Africa, such as fire Department of Land Affairs, seek to influence change
(Department of Land Affairs 1997).

Rural households represent one of fire largest contingents of agentist
change. Decisions made at fire household level do not always conform wifir
structural directives, legal codes, or proclamations by local, regional, or national
aufirorities. For instance, land invasions are a clear expression of an agent’s
desire to occupy land against fire structural constraints of law and law
enforcement. Household decisions, it will be shown in firis study, do not always
take into consideration guidance or instructions from structures in society.

Land use change in fire Northern Province includes fire transfer of ”white"
farms to Africans, occupah'on of land by Africans in formerly restricted areas,
abandonment of land, state-sponsored land reform projects, change in fire mix of
crops, for instance from subsistence crops such as maize to commercial crops
(Upton, etal. 1995). Land use patterns can also change as a result of decisions
made by the chief. This situation occurs as a result of more "traditional" forms of
tenure and political relationships (Cross and Haines 1988). In this study, two
types of land use change will be examined- fire transfer of former "white” farms

to Africans and fire changes in areas near fire transferred lands.

33

2.4 Land Alienation and Reform in South Africa: Examples from the

Literature

2.4.1 landlessness

While fire focus of firis dissertation is on the dynamics between land use
change and livelihood systems, a review of land issues in Soufir Africa will lead
to a greater understanding of how livelihoods and land use are affected by the
rural-historical context. The current situation in rural areas is a direct and
unambiguous result of fire assault of fire colonial and apartheid states on the land
rights of Africans. Land use change in Soufir Africa is not a spontaneous event,
rafirer a long process of contestation between various individuals, groups,
communities, and authorities. To understand the process of land use change, it is
vital we understand fire how fire conditions of current land distribution were
affected.

While land alienation in South Africa did not begin with the 1913 Native
Lands Act, firat act did codify previously disparate laws and statutes into an
overarching and draconian system firat facilitated African land dispossession. In
reality, alienation had been occurring since fire first colonial wars in fire Cape.
Africans faced dispossession on a large scale wifir fire extension of the Cape
Colony eastward into what is today fire Eastern Cape province. Bundy (1988)
maintains firat a peasantry was created and destroyed in the process of nacient

capital expansion. Once capital had penetrated fire interior reaches of the firen

colonies of fire Cape and Natal, and later the Orange Free State and Transvaal, it
set out to destroy fire much more competitive and economically successful
African peasantry.

In fire late 19th and early 20th century, fire concentration of power into fire
hands of fire white minority and the process of consolidation of state aufirority
into fire rural areas gave rise to fears among white farmers over lack of labor. At
first time, fire range of agricultural activities open to Africans was considerably
wider than after fire enactnrent of fire 1913 and 1936 land acts. Africans could
engage in peasant agricultural production on land titled to firemselves or on
communal land, become a labor tenant, sharecrop, become a farm laborer, or
occupy and farm state land. At firis point, the definition of a ”peasant” becomes
rather important. The argument forwarded by Bundy firat the peasantry had
been destroyed by fire time of fire Second World War is central to land issues in
fire present. Krober and Redfield (in Bundy 1988) define peasants as:

1) involved in market activity

2) involved in politics beyond fire village

3) involved in country-wide peasant network

4) representing a small version of the larger society

Shanin (cited in Bundy, 1988) rejects firis culturally based definition and
opts for a more power-based definition, remarking that in peasant societies:

1) fire family farm is fire basis of social organization
2) land husbandry is fire main focus of fire household and fire main source
of livelihood

3) a culture develops firat appreciates small communities
4) peasants are oppressed by ofirer social groups

35

In Soufir Africa, even at fire turn of fire 20‘“ century, it would be hard to find
many rural inhabitants who would qualify for the label ”peasant” by eifirer of
firese definitions. In order to avoid a long fireoretical debate, I will side wifir
Bundy on fire issue of definition. He advocates using parts of bofir definitions
and borrows from ofirer analyses writing:

one arrives at a definition of peasant to typify Africans in the Cape (and

Soufir Africa generally) during fire nineteenfir century. An African peasant

was a rural cultivator, enjoying access to a portion of land, fire fruits of

which he could dispose of as if he owned fire land; he used his own labour
and firat of members of his family in agricultural or pastoral pursuits and
sought firrough firis to satisfy direcfiy the consumption needs of his family;
in addition he looked to fire sale of a portion of what he raised to meet fire
demands (taxes, rent, and other fees) firat arose from his involvement in an
economic and political system beyond his community. Like peasants
elsewhere he had recourse to a specific traditional culture...he was
dominated economically, politically and culturally by outsiders in a wider
society - involved in relations of coercion and obedience... the extent to
which fire state or its representatives could enforce firese relations differed

sharply from time to time and place to place (1988, 9).

From an economic standpoint alone, fire life of South African peasants in
fire late 19‘“ century was clearly unpleasant, however, it was to be looked upon
as a golden age. For instance, in a district representative of fire ’native reserves’
fire average purchases for an African family of six fell between 1875 and 1925
from £20 to £9 while fire average sales for goods fell from £20 to £4 (Bundy 1988,
p.223). This drop was precipitated by fire near collapse of African agriculture as a
result of anti-competitive regulations favoring white setfier agriculture.

The two legal documents most responsible for land alienation in Soufir

Africa are fire 1913 Natives Land Act and fire 1936 Development Trust and Land

Act. The first act prohibited African occupation or ownership of land not
scheduled as African. The act established reserves (four of which were to become
”independent”) based on efirnicity and constituted the only areas where Africans
could legally obtain tifie to land. The 1936 Act scheduled additional lands to be
added to fire reserves and further extended fire ability of fire state to enforce the
1913 Act. A firird act, fire Native Administration Act of 1927, provided a strong
enforcement mechanism - fire ability to forcibly remove Africans from one place
to anofirer. Section 5 (1)(b) of fire Act states:

The Governor-General may whenever he deems it expedient in fire general

public interest, order the removal of any tribe or portion thereof or any

Native from any place to any ofirer place wifirin fire Union upon such

conditions as he may determine: Provided that in the case of a tribe

objecting to such removal, no such order shall be given unless a resolution
approving of fire removal has been adopted by both Houses of Parliament

(Murray and O' Regan 1989, p.18).

The language granting ”objecting tribes" fire benefit of parliamentary
review was altered in subsequent amendments. Planners needed only to consult
Parliament beforehand. ”Objecting tribes” had no voice in and litfie recourse to
such actions. The Group Areas Act (1950) provided for stronger regulation of
efirnic and racial groups, but its statutes were largely affected in urban areas. The
bases for dispossession were firmly set in fire 1913, 1927, and 1936 land acts.

The concrete expression of the various land laws was forced, and often,

violent removals of Africans from fireir land. In many cases, firis was ancestral

land and in ofirer cases it was land individuals and communities had ’purchased’

37

prior to fire 1913 alienation. The timbre of fire forced removals was sharpest from
fire early 19608 to fire mid-19703. The 1927 Bantu Administration Act, described
above, gave fire state sweeping powers of removal. Generally, the landlessness
currenfiy experienced by so many people is a result of the enforcement of firat
act.

Africans were either physically forced into homelands or homeland
boundaries were redrawn to incorporate Africans into firem, a process of forced
nationalization. All land in Soufir Africa was scheduled for specific racial groups.
Where residents did not coincide with the ascribed racial-geographic
classification, fire state stepped in to ’correct’ fire situation. Of course, fire
’correction’ was always at fire expense of Africans. The process of consolidation
of fire homelands (and later ”independent" states) in fire 1970s further

aggravated fire already acute problem of landlessness.

2.4.2 Tenure

Understanding bofir land use change and livelihood systems is dependent,
in firis context, on thorough review of fire types and spatial manifestations of
land tenure. The role of land tenure in land use change and livelihood systems is
most pronounced in rural Soufir Africa. The long historical process of disruption
of African agriculture, setfiement, and tenure by fire setfier state is well
documented. (Cross 1988; Cross 1995; especially Bundy 1988; for fire Norfirern

Province Letsoalo 1987; Adams, et.al. 1999; Claassens 1999). This process was one

38

in which fire primary objective was not to alienate Africans from their land for

European settlement. The primary objective, Ietsoalo (1987) demonstrates, was

not land alienation rafirer:
Only by understanding fire evolving political economy can one decode
what has happened to black landownership and the double-meaning of
”land” (i.e. labour) reform in South Africa. Like all capitalist economies,
Soufir Africa’s economy is dependent on fire existence of a labour
reservoir. The Black community could not perform firis function as
independent farmers (owning land) (41).

A series of statutes and laws governed - or more accurately created chaos
in - tenure systems in fire native reserves, which would later become fire
homelands. At a broad scale, fire homelands themselves were established as
”black areas" for specific racial groups by fire Bantu Authorities Act (1951) and
the Promotion of Bantu Self Government Act (1959). These two acts were further
strengfirened wifir fire passing of fire Homelands Constitution Act (1971) which
gave sweeping powers to fire firen non- ”independent" homeland aufirorities.
These firree acts gave control of agriculture and tenure to homeland governments
all too inept in such matters. The Bantu Areas Land Act (1969) governed day-to-
day management of land and tenure arrangements at small scales. This act is still
significant due to fire fact firat at fire time of writing post- aparfireid Soufir Africa
has still not codified a new tenure system for fire former homelands and many
administrators still rely on fire 1969 statue.

What, firen, are fire tenure systems firat have played (and continue to play)

such a key role in rural areas? Cross (1991), Groenewald (1998), and Letsoalo

39

(1987) describe fire tenure systems in terms of ownership, rights of use, and
access. Cross’ typology of tenure includes communal, trust, quitrent, freehold,
and leasehold (70). Letsoalo generalizes more and divides tenure into locations /
reserves, tribal bought land, private bought land, and trust land (63-65).
Groenewald presents possible tenure arrangements, largely overlapping Cross
and Letsoalo but puts forth several different options including collective and

large scale private farms (113). Table two summarizes fire main characteristics of

each type.

Table 2: Summary of fire Types of Tenure in Rural Soufir Africa

 

 

 

 

 

 

 

 

 

 

 

 

Type Rent Who Controls How Level of Distributio
Paid 7 Land Use Common Security n in Pre-
Decisions? 1994 South
Africa
Communal No Chief or Tribal Very High KwaZulu,
Council Common Ciskei,
Transkei
Trust Yes, to Based on Common Very low Norfirern
fire state conditions laid Tranvaal,
down by fire Bopufira-
state tswana
Quitrent Yes, to Based on Rare Moderate Transkei
fire conditions, but to low and Ciskei
State less restrictive only
than Trust Land
Freehold No Individual Very Rare High Through-
out SA
Leasehold Yes, to Based on Common Very Low Through-
fire conditions set out SA
lessee down by
individual
offering fire
lease; state

 

after Cross 1991 and Ietsoalo 1987.

40

 

The primary reason for fire existence of confusing and often contradictory
land tenure regimes was fire extraction of surplus. Land in trust systems
represented a large portion of total land allotted as ”African”; firis was no
coincidence:

fire question of payment of Trust rent is a cause for great dissatisfaction in

fire homelands. It is considered unfair for two reasons. Firsfiy, what is

termed Trust land is land that formerly belonged to Blacks before it was
forcibly appropriated by fire Whites. Secondly, it is estimated firat if the
rental money was payment towards the purchase of land, fire Blacks
would have already bought more firan fire area firey now occupy... the
rent money goes firrough fire homeland institutions (magistrates) to fire

Trust (Soufir African Government), and firen back to fire homeland

revenues. This is fire circular route money takes ”to assist fire development

of firat particular area” (Letsoalo 1987, p.66).

The spatial distribution of tenure regimes was highly disparate and
continually changing under apartheid administrators. Keeping African land use
systems in flux and uncertainty ensured landlessness and a large supply of labor.

Considering firat all of fire new nine provinces, save the Western Cape,
include parts of fire old homelands, current land administration has become
highly fluid due to uncertainty over precise land rights and conflicting land
tenure systems wifirin the same province. This leads to a great deal of conflict
and uncertainty in land use (Cross 1995). Individuals and communities are
simply unwilling given fire abysmal track record of fire state to provide tenure
security, to undertake substantive improvements on land that has any question

regarding ownership or use rights. Land rights currenfiy are held in a state of

limbo wifir fire passage in 1996 of fire Interim Protection of Informal Land Rights

41

Act. This act simply freezes in place existing tenure arrangements in fire Republic
as of 1996 and has caused considerable confusion among land administrators
(Claassens 1999; Sibanda, 1999). The tenure systems frozen into place are
overlapping, spatially chaotic, and often contradictory. Critics of the system set
forfir in fire Interim Protection of Informal Land Rights Act, known as
”confirmed rights” pointed out firat the language was ”vague and imprecise"
and firat not all rights associated wifir land were conveyed under firis act
(Claassens 1999). The system simply stated firat any current rights to land could
not be alienated from anyone without fireir consent. Because of the contradictory
nature of tenure systems in aparfireid South Africa, firis left vast numbers of rural
people in a much weaker position relative to fireir counterparts. Land rights are
generally understood to include

- fire right to occupy

- fire right to use

- fire right to bequeath to one’s heirs

- fire right to transact (lease or sell fire right)

- fire right to mortgage fire land

- fire right to exclude ofirers (evict)

- fire rights to benefits accruing from fire land

(after Claassens 1999)

The revision of fire system included a set of rights firat were ”conferred on
people by a local process of confirmation or allocation” (Claassens, 1999)
inclusive of fire rights listed above. This new policy seeks to integrate fire existing

rights and fire conferred rights in such a way as to maximize security without

compromising fairness and democratic process (Sinbanda 1999; Makopi 1999).

42

Currenfiy, land tenure in rural Soufir Africa is only slighfiy less chaotic firan
under apartheid (McIntosh and Vaughan 1999). Until a coherent, comprehensive

policy is codified rural development and transformation will be stymied.

2.4.3 Land Rejbrm Choices and Policy

I will abstain from a comprehensive review of the voluminous literature
firat developed from fire mid-19803 and peaked in 1993-95 regarding ”land
reform options”. Now firat policy has taken shape and clear directions have been
mapped out (and have since shifted again) much of the ”options” literature has
fallen by fire wayside. I will begin by sketching the broad oufiines of fire ”land
reform options” literature, but will avoid issues firat have subsequenfiy been
rendered moot by either policy actions or current reality. Much of fire literature
was based on suppositions of how a democratic transition might take place, or
how social policy might be formed, or how homelands may or may not be included
in fire new Republic. I can easily rule out literature based on suppositions firat
have not materialized or where reality has contradicted fire aufiror’ s main
arguments.

International experience in land reform was employed in an effort to
preclude policy makers in South Africa from fire same mistakes repeated in
Africa, Latin America, and East Asia. However, Soufir Africa has a history unlike
any of fire areas firat have undertaken land reform. Kenya and Zimbabwe have

had similar experiences, yet did not suffer from a legal system of separation and

43

racial urrderdevelopment to the extent of South Africa. The crescendo of
academic reference in the immediate pre-1994 era led to multifarious
recommendations on ”what to do”. The plethora of literature focusing on land
reform options in Soufir Africa is overwhelming. Bernstein (1999) notes firat so-
called ”reformed” aparfireid-era academics felt certain that the market—driven
policies firat worked so well for white Soufir Africa could be transformed into
”dispossessed-friendly” policy. South Africa emerged from aparfireid at fire
heyday of World Bank and IMF restructuring programs firat emphasized
”getting fire markets right”. International agricultural economists stressed fire
need for market driven reforms, culminating in fire work Agricultural land
Refimu in South Africa. This volume became the reformers’ central policy
platform. It is fire brainchild of several agricultural economists and planners at
fire University of Pretoria and includes aufirors from fire Development Bank of
Soufir Africa (fire aparfireid institution charged with ”development" of fire
homelands). International experts poured into Soufir Africa in 1993 and 1994
ready to tackle fire challenge of land reform in South Africa. It is in firis context
firat we must temper our analysis of land reform policy options. These are
generally not homegrown options, nor do firey represent options as perceived by
fire large majority of fire African population. It would be difficult to say if firese
options even represent fire African intellegencia in Soufir Africa, as few of firem

have been consulted. Land reform options are being driven by white acadenrics

in fire same institutions firat were largely responsible for agricultural and land
policy development under apartheid.

Binswanger and Deininger (1996) map out the ”different paths” to land
reform in ofirer parts of fire world where land distribution was skewed toward
large farms. They note firat reform in landlord estates firroughout Asia has led to
more stable production systems but can lead to evictions of farm-workers or
tenants. Collective farms fail to provide workers wifir incentives, degenerate into
”wage-labor operated state farms”, and tend to suffer from under-investment.
The aufirors, however, point out fire problems of delaying land reform, arguing
firat smallholder profitability is reduced (83) and that delayed reform has led to
violence and militancy among peasants (84).

Drawing on examples from Kenya and Zimbabwe, Binswanger and
Deininger note firat reform in Zimbabwe was less successful. This was due to
poor financing and linrits on land subdivision, prohibition of non-farm activity,
and fire large size of farms, among ofirer factors (89). Their suggestion for South
Africa is an approach firat emphasizes private ownership yet allows for flexible
forms of communal tenure, redress of grievances, and fire willing buyer - willing
seller method of land redistribution (95—96).

Van den Brink, et.al. (1996) relate land reform more closely to rural
livelihoods. They present models based on fire different levels of beneficiaries,
namely communities, farmworkers, and individuals. At the community level,

common property models reflect a community’ 5 will and are broadly more

45

acceptable in areas where ”tradition” holds fast. Irrigated garden plots can also
be developed at fire community level where small-scale use is more appropriate.
Options for farm-workers range from improving existing work conditions and
term of employment to a farm-worker common property model to production
co-operatives and equity sharing schemes. Individuals could benefit from
irrigated market gardens programs, fire out-grower (tenant) model, or
commercial family farms based on freehold tenure.

To reduce fire financial burden on fire state, fire aufirors recommend fire

following:

1. fire sale of state land

2. a system where beneficiaries make co—payments

3. a reduction of fire start up grant

4. firat beneficiaries must contribute fireir labour to infrastructure
improvement (sweat-equity)

. pension buyouts

. obtaining grants from donors

. elimination of agricultural subsidies

. issuing land reform bonds

“VOW

van den Brink, et.al. 1996, p.447-450.

Describing guidelines for selecting beneficiaries, van Rooyen and Njobe-
Mbuli (1996) suggest firat access to farmland is ”important alfirough not a prime
factor" in reducing poverty, farming does not always result in increased food
production, land reform must address fire needs of the poorest, and a broad
approach must be undertaken (466). The selection of beneficiaries must take into

account inequalities in ownership and access, how apartheid laws created

46

poverty, ”the need to establish land use models which will ensure improved
efficient and sustainable use” and ”democratic (transparent and broad-based)
participatory processes” (473). The disenfranchisement of people under
aparfireid, poverty and fire degree of need, land productivity and sustainability,
a participatory selection process, and reference to case studies should guide fire
criteria for selection of beneficiaries (474—476).

Land reform options firat address farm-workers’ needs include joint
venture / participation schemes and resetfiement programs (Ngqangweni and
van Rooyen 1998). Joint venture / participation schemes involve ”rearrangement
of ownership to include farm-workers and thus include firem in the main stream
economy’ ' (72). Table 3 summarizes fire various approaches to joint venture /

participation schemes.

47

Table 3 : Options in Land Reform

 

 

 

 

 

 

 

 

 

Type Description Advantages Disadvantages
Equity Workers buy into an Workers Resistance from
Schemes existing operation or manage; acquire previous owner;
create new one equity need for high
level of support
Build, Private sector builds, Workers get Untested;
Operate operates and eventually extensive unwillingness of
and transfers a scheme to fire training; low investors to
Transfer public sector start-up costs relinquish assets
for fire state
lease and Workers lease land and Financing Workers may not
Buy equipment wifir option provided by fire be able to raise
Schemes to buy state capital
Agri- Encourage enterprises to Minimizes Enterprises tend
Village acquire and develop delivery costs; toward
land for its employees facilitates land hegemony over
ownership in workers
high land cost
areas

 

after Ngqangwerri and van Rooyen 1998, p72-74.

A successful program of land reform, Christiansen (1996) remarks, will
emerge if fire speed of implementation is fast, farm models are scrutinized
closely for economic viability, if fire program is politically acceptable and

legitimate, fire role of fire public sector is clear and reduced, and if land reform is

”part of a comprehensive program of economic reconstruction” (368).

The market-driven / market-assisted approach to land reform is not fire

only option, however. Berstein (1999) characterizes fire main problem wifir fire

dominant land reform paradigm:

 

as for fire agrarian question, a few individuals charged by fire ANC wifir
formulating (or conjuring) land and agricultural policies were
overwhelmed by representations, models, and prescriptions from a variety
of quarters, South African and external. The former include verligte (or
hastily ’born again’) elements of organized agriculture, agribusiness and
fire Development Bank of Southern Africa, DBSA (established to direct
’homeland development’ , as part of the domestic ’reform’ firrust of PW
Bofira’s Total Strategy). External ideological skills, policy analysts and
advice were supplied by the World Bank and other aid agencies, fire
planeloads of consultants attached to firem, and the rest of fire
internationally mobile usual suspects. (5)

Bernstein demonstrates fire need to alleviate rural poverty and ”reform fire
modes of domination firat reproduce fire political exclusion of rural ’subjects’
”(23). The market driven approach treats rural peOple as subjects to be studied
and transformed by fire dominant (Western, neo-liberal) agricultural philosophy
into ”productive and efficient” small farmers. He warms firat ”black bourgeoisie
land acquisition seeks to exploit ostensibly ’customary’ channels of land
allocation and / or opportunities afforded by government land reform policies. . .”
(24). The dominant agricultural reform literature mentions nofiring of such class
struggle for land.

Land reform policy in Soufir Africa has been ”captured”. levin and Weiner
(1997) expand:

fire ANC-led Government of National Unity put its faifir in comprehensive

land reform and a market-led land reform strategy. ..firis signifies a

triumph of neo-liberalism in fire land debate, and fire privileging of fire

market as fire central mechanism for land redistribution. The political
space for fire consolidafion of neo-liberal firinking around land reform is

due in part to an urban bias wifirin fire democratic left which contributed
to a conception of linear proletarianization of rural Blacks (7).

49

Examining current land policy firat privileges rural elite and newly rural
”peasants” validates firis statement. The ”Commercial Farmer Programme” is
just fire latest in a series of revisions to the original goals of land reform. The first
indication of fire composition of an ANC led land reform strategy dates as far
back as 1955 wifir the Freedom Charter. That document expressed the need for
land reform at fire most basic and fundamental level, in fire redistribution of
land:

Restrictions of land ownership on a racial basis shall be ended, and all fire

land re-divided amongst those who work to banish famine and land

hunger; fire state shall help fire peasants wifir implements, seed, tractors
and dams to save the soil and assist the tillers. . .People shall not be robbed
of fireir cattle, and forced labour and farm prisons shall be abolished (ANC

1955).

Wifir fire end of aparfireid in 1994, concrete steps were taken to reverse fire
lop-sided land distribution of fire past and to accelerate real rural development.
The first policy statement on land (ofirer firan to repeal aparfireid era codes and
statutes) was fire Restitution of Land Rights Act (22 of 1994). This act specifically
set out to ”provide for fire restitution of rights in land in respect of which persons
or communities were dispossessed under or for fire purpose of furfirering fire
objects of any racially based discriminatory law” (Republic of Soufir Africa 1994,
p.3). The Development Facilitation Act (67 of 1995) provided fire terms for land
development, especially housing.

The first comprehensive land statement emerged in 1996 wifir the

publication of fire Green Paper on South Afiican Land Policy, and was revised in

50

1997 to become fire White Paper on South African land Policy. Enshrined in firese
two documents were fire principles of land reform. The White Paper identifies
areas firat needed address, including:

1. The injustices of racially-based land dispossession;

2. The inequitable distribution of land ownership;

3. The need for security of tenure for all;

4. The need for sustainable use of land;

5. The need for rapid release of land for development;

6. The need to record and register all rights in property; and

7. The need to administer public lands in an effective manner.

(Department of Land Affairs 1997, v)

Land reform issues were divided into firree broad categories: land
restitution issues, land redistribution issues, and land tenure reform issues.
Provisions were made for a once-off R15000 grant wifir which individuals could
purchase land ”direcfiy from willing sellers” (ix).

The next important piece of legislation for firis study, was fire Communal
Property Associations Act (849 of 1996). This act allowed for fire establishment of
quasi-communal groups in order to purchase farm land under fire willing buyer-
willing seller program. It was recognized from fire initial stages of planning firat
very few individuals would be able to purchase land with the once-off R15000
grant. The Communal Property Associations Act allowed individuals wishing to
pool fireir resources to purchase white farmland a clear legal framework and

protection of fireir rights. The act called for each Communal Property Association

(CPA) to have a constitution first would, among ofirer firings:

51

- Be fair and inclusive

- Prohibit unfair termination of a member

- Prohibit discrimination

- Provide for regular meetings

- Ensure financial transparency and accountability

In early 2000, land policy shifted significanfiy under fire Minister of
Agriculture and Land Affairs Minister Toko Didiza. Individuals wishing to
obtain land now must match fire government grants:

Following fire ministerial review process, the current grant system will
continue but is to be revised to enhance access to land. Three
redistribution windows are to be created, ranging from small, to medium
to large projects. The windows will refer to the total project cost and not to
the size of the farming enterprise or fire type of farmer. Allocation of fire
government grant will be based on fire grant amount per total project cost
per window and not on grant amount per beneficiary per window.
Government's contribution to each window is set at about 70%, 40% and
20% in the small, medium and large sized projects respectively. Eligible
projects for grant financing are commonage, communal and commercial
farming.

The current setfiement/ land acquisition grant (SLAG) is to be replaced by
a land reform grant, wifir specific categories (relating to fire purpose of fire
grant). For example, grants for residential setfiement and fire creation of a
food safety net will be treated differenfiy from grants aimed at
establishing market-based agriculture. The livelihood and food safety net
grant is aimed at fire poor wifirin communities who do not have land and
cannot sustain firemselves. The grant is intended to give firem bofir land
on which to live and food security.

The new supply-led system will be piloted wifir a more pro-active
approach to managing the allocation of land, coupled with fire more
strategic use of grants to support the Government's integrated rural
development strategy. The new programme is aimed at distributing at
least 15% of farmland in the next five years. It envisages land
redistribution as a mechanism to facilitate long-term structural change in
agriculture. (ULA 1999, 2-3).

52

The focus has shifted to creating a black commercial farming class rafirer
firan righting historical wrongs and alleviating rural poverty. The shift will
inevitably alienate fire poorest of the poor from fire land reform process.

Land restitution since 1994 has been extraordinarily slow and heavily
criticized (Brown, et.al. 1998, p.1). The Department of Land Affairs 1998 Annual
Quality of Life Report was undertaken to assess redistribution projects, however,
suffered from flawed mefirods. For example, out of the total of 62 projects
surveyed, only one was in the Northern Province and firis one project
represented only four households. Thus, one of Soufir Africa’s poorest provinces
and fire one with fire direst need of redistribution was represented by four
households. On fire ofirer extreme, KwaZulu-Natal and fire Free State were
represented by 56 and 55 households on 14 and 16 farms respectively (DLA 1999,
p.5). This imbalance in sample size underlines fire generally haphazard approach
of fire DLA to redistribution. One of fire most stunning admissions in fire report
is fire lack of action on actual redistribution versus needs up to 1998. The
estimated percentage of households who need access to land is placed at 67.7%
for fire entire country and 72.4% for fire Northern Province, yet only 0.2% of
households have received land nationally. The situation in fire Northern
Province is worse wifir only 0.1% of households in need having received land. In
no province is fire ratio of need to delivery rate higher than 0.7% (DLA 1999, p.9).
These numbers are even more shocking when examined in the light of fire

original goals of land reform in Soufir Africa. According to the DLA:

53

The initial land reform target for fire redistribution programme was to
transfer 30% of Soufir Africa’s 99.07 million hectare farmland, between
1994 and 1999, which amounts to 29.72 million hectare. After firree year of
operation, firis is what has been achieved: about 200,000 hectares of land
have been transferred to about 20,000 households, which is 0.6% of fire
total farmland demanded by households and 0.6% of the target set. This
means firat only 0.2% of fire total number of households who demand land
have benefited under firis programme. This indicates firat fire programme
is not performing as well as it was expected, and it is doubtful that fire
target will be met...lt is difficult to comment on fire progress of land
restitution. . . however, at face value firis process seems to be too slow
(DLA 1998, p.10).

In firis same document, DLA defensiveness is clear:

Many do not seem to appreciate the progress of the land reform

programme. As far as fire numbers of households resetfied or gaining

secure tenure are concerned, significant progress has been made in terms

of setting institutional structures, establishing planning procedures, and

learning to identify botfienecks at different levels (DLA 1998, p.9).

Thus, tasks firat should have been completed in fire first six months of fire
land reform process have taken six years. The desperately slow process in Soufir
Africa mirrors that in Zimbabwe, and as events in early 2000 have shown, failure

to deliver land can have dire social consequences.

The 1999 Quality of Life Report reflects the same slow process:

Summary of settled restitution claims
Land restoration

Households receiving land

Land cost

Restoration order by LCC (hectares)
Restoration approval - s42D(hectares)

Final compensation

Households receiving compensation
Financial compensation order (LCC)
Financial Compensation approval (s42D)

Restitution totals

Claims setfied as at 18 Feb 2000
Total claimant households
Total restitution beneficiaries
Total restitution award cost

55

11 194
R87,756,559
173 805

90 063

973
R0,0
R32,639,639.64

1 651

12 167

72 406
R120,396,198,64

The process in fire N orfirern Province is even slower:

Northern Province

land restoration

Households receiving land

Land cost

Restoration order by LCC (hectares)
Restoration approval s42D (hectares)

Financial compensation

Households receiving compensation
Financial compensation order (LCC)
Financial compensation approval (s42D)

Restitution total

Claims setfied as at 18 Feb 2000
Total claimant households
Total restitution beneficiaries
Total restitution award cost

2 570
R0,00
25 000

1
R0,00
R174,157.00

2

2 571

15 940
R174,157.w

From a beneficiary perspective, Zimmerman (2000) explored barriers to
participation of fire poor in fire land reform program. He puts forward five
reasons explaining why fire poor are less likely to participate in land reform firan
could be expected, namely: high up-front costs, fire tendency of fire poor to be
risk averse, a lack of human capital, a lack of time, and the large distances over

which people would have to move to participate in land reform.

2.4.4 The Communal Property Association Model

The Communal Property Association was a statutory model of land reform
delivery in fire Departrrrent of Land Affairs from 1996 until 2000. CPAs were
established in areas targeted by central authorities as having a land imbalance.
Land for transfer is selected from willing sellers (usually white) firat is physically
near historically dispossessed black communities. According to fire members of
CPAs, local government officials informed firem of projects in fireir area. These
officials would post advertisements in rural areas and townships offering to
facilitate fire transfer of farmland. To become a member of the CPA, interested
parties were required to contribute their R15,000 state social grant to a pool of
money firat would be used to purchase fire (white) farm. The number of
members was determined by how much money it took to purchase fire target
farm. Several CPAs in fire N orfirern Province have over 300 members while some

have fewer firan 40. This reflects different costs involved in acquiring fire land.

57

 

CPAs are run democratically with each member having a vote in fire
organization. An executive committee consisting of a chairperson, vice-
chairperson, secretary, and treasurer conducts daily tasks of administration. This
body is responsible for making decisions regarding fire operation of the farm,
finances, and communicating to members.

The constitutions of fire CPAs forbid members from residing on fire
transferred land. The intention of transferring fire land is for agricultural
production rafirer firan residential dwelling. The government stipulates fire
amount of start-up funding each CPA is allocated. In practice, some CPAs are
better funded firan ofirers. There is no overall produce or catfie marketing
association for fire CPAs in a given region or province and firey are largely at
fireir own devices to raise, harvest, and sell any produce or livestock. In practice,
many of fire CPAs are reliant on fireir relationship wifir local government for
resources. Direct transfer from fire government is a common method of capital
injection. No CPA in fire study area had relationships wifir fire private sector
eifirer to obtain capital or supply produce.

Membership participation is based on cooperative production. Each
member, in fireory, is expected to contribute to fire association and collect fire
rewards in fire end. Production was intended to be self-supporting after fire
government covered fire initial capital costs. The investment by the government
was intended to jump-start fire first year’ s production. The profits would be

reinvested to support future production. Government’s role was intended to be

58

limited to ”getting fire CPAs off fire ground”. They were envisioned to be self
supported by fire second or firird year (M. Labuschagne, 1999).

Very few of fire original intentions for fire CPAs have materialized.
Members do not always participate in communal production, the associations
have not been able to sustain production at all, therefore, cannot reinvest. They
are reliant on government capital injections and have formed few partnerships
wifir fire private sector, eifirer for provision of supplies or as an oufiet for goods.
In fact, of fire CPAs in firis study, only one was able to sell agricultural produce
and only one was able to sell livestock. Regarding fire Communal Property
Associations, fire Departnrent of Land Affairs has acknowledged fire poor state of
affairs:

Communal property associations are now just more firan three years old.
As an institutional arrangement, firis form of association has seen bofir
success and failure. On the one hand, the communal property associations
have established a legal mechanism which could be used by groups to
jointly acquire and manage land. The restitution, redistribution and
tenure programmes have relied heavily upon firese institutions to facilitate
land transfer. In fire past three years, more firan 150 associations have
been registered wifir the bulk registered in 1999. Most of these
associations were registered as a result of the redistribution programme
wifir restitution trailing closely.

However, a close analysis of communal property associations has revealed

firat many of firese associations have become dormant and ineffective.

Some reasons cited for firis are fire lack of post-transfer support and a lack

of training of land reform beneficiaries. This might necessitate an urgent

review of fire administrative and institutional processes following fire

establishment of firese institutions. (DLA 1999, p.34-35).

The Communal Property Association Program has been widely criticised

for not turning dispossessed rural Africans into instant peasants. Litfie mention

59

has been made firat fire CPA programme has failed to meet ofirer social needs of
land reform. The new policy directions of fire DLA indicate firat fire former
mantra of distribution to alleviate historical injustices and reduce rural poverty
have succumbed to fire same neo-liberal pressures firat dominated fire pre-1994
debate on land reform. The ’profit or perish’ mentality of market advocates has
been applied to fire Communal Property Associations, fire last institution able to
deliver meaningful reform to fire rural poor. The CPA program has also
succumbed to neo-liberal reform policy. Meaningful rural transformation is

beginning to fade into memory, just six years after the end of aparfireid.

2.5 The Impact of Change through Time: Examples from the Literatures

on Resettlement, State Farms, and Agricultural Cooperatives

One fundamental question firat arises from an examination of the design of
firis study is: Has enough time passed to assess the Communal Property
Associations? An assessment early on in fire project life would yield results
indicative of a transition. People have not yet setfied into fireir new livelihoods.
Therefore, it is crucial to locate the CPAs bofir in the literature and wifir fire field

data to identify any indication firat projects are still in transition.

2.5.1 Resettlement and the CPAs

Classifying CPAs as resetfiement schemes is a misnomer. The members of
CPAs in firis study have not moved eifirer voluntarily or involuntarily to
participate in fire projects. Resetfiement has occurred in Soufir Africa as a result
of land reform largely in restitution cases, alfirough resetfiement did occur on a
few CPAs. In firis study only a handful of members (fewer firan 30) of one CPA
resetfied firemselves voluntarily to participate.

The resetfiement literature is being discussed here as it comes closest to
answering questions of time and change. The land reform literature largely
ignores fire impacts of a land reform event on eifirer livelihoods or land use
change. In order to analyze whefirer or not enough time has passed to
substantively evaluate fire CPA program, fire resetfiement literature must be
drawn upon for empirical examples of the impact of change on livelihoods and
land use. Clearly, more disruption will have occurred in fire case of resetfiement
firan CPA establishment, however, a transition baseline can be established to
support or reject fire analysis of fire CPAs at fire four to six year mark.
Unfortunately the literature does not provide clear and concise answers, as
Kinsey and Binswanger point out ”firere is no generally accepted mefirodological
approach or theoretical basis for fire analysis of resetfiement scheme” (106).

Settlement schemes have a long and not always successful history.

Resetfiement is a widespread phenomenon with various causes and

61

consequences. While fire size, scope, and type of resetfiement varies from
country to country, broad similarities can be found in the literature (Manshard
and Morgan, 1988; Mbifiri and Barnes, 1975; Chambers, 1969; Apfirorpe, 1968;
Hansen and Oliver-Smifir, 1981). A typology of setfiement schemes includes
”administered official setfiements, unaufirorized or spontaneous setfiement, and
assisted spontaneous setfiement (Kinsey and Binswanger, 1996, p.115-119). While
fire empirical evidence on resetfiement is large, the scope is rafirer narrow.
Studies of resetfiement too often focus on the economic rather firan social
outcomes of resetfiement:

because fire bulk of fire literature comes from [empirical approaches],

assessments of performance are weighted heavily by fire disciplines in

which researchers have been trained...evaluations in firis category tend to
present static descriptions and to gloss over processes and relationships.

Unfortunately, fire deficiency cannot be remedied by balancing

conventional examinations with those drawn from fire social consequences

literature, since there is too litfie of the latter to serve as an effective

counterbalance (Kinsey and Binswanger, 1996, p.107).

Colson (1971) was one of fire first aufirors to investigate the social
consequences of resetfiement in her analysis of the Kariba Resetfiement Scheme
in Zimbabwe. She maintains that resetfiement has direct, and often negative,
social impacts. Among firese she focuses particularly on socio-spatial
organization, fire process of recovery, and long-term effects. Salient here is first
Colson provides insight into how rural livelihoods were altered, specifically due

to fire re-organization of labor, political instability, intra-household stresses and

fire consequences to women, and fire uncertainty and volatility of income.

62

Oberai (1988) expands on the problems of resetfiement. Consequences
include "setfier abandonment of land, lack of non-farm employment, fire
concentration of land into fire hands of a few, social tensions, and ecological
problems” (22-28). Further, fire success or failure of resetfiement is dependent on
fire extent to which firese problems are addressed and mechanisms firat are

devised to cope in bofir fire short and long- term.

2.5.2 lessons from Latin America

Due to the fact firat many shared conditions exist, such as a large setfier
community and highly unequal land distribution, case studies from Latin
America provide useful insight into resetfiement wifir implications for Soufir
Africa. Both areas have experienced European setfiement early in fireir histories;
bofir experienced large-scale land alienation of fire indigenous population; bofir
attempted land reform after political reform.

Findley (1988) after Nelson (1973) argues that resetfiement occurs in firree
phases. The first phase is fire selection of the settlers, fire second is fire pioneer
stage, and fire firird is consolidation. She explains:

In fire first or selection phase, the potential colonist ”selects” himself or is

selected as a colonist, and firen proceeds to select a plot. During fire second

or pioneer phase, fire colonist clears his land and begins cultivating crops,

usually subsistence food crops. After a few years, fire colonist begins fire
firird phase, which consists of colonization or commercialization

(Findley,1988,p.279).

63

The aufiror recommends firree important elements for the successful
consolidation of projects, namely sufficient land, capital, and labor. She argues
firat during fire transition phase ” [fire pioneers] can not be expected to arrive and
immediately commercialize” (290). ‘

Findley (1988) sets out fire ”structural conditions” firat led to success in
Latin American reform and resetfiement programs:

1. Appropriate crops were selected, for which demand and marketing

facilities existed.

2. The cooperative provided an effective organization to facilitate
commercialization. In particular, it provided inputs in a timely fashion.

3. Credit to setfiers was based on potential yields, which enabled setfiers to
shift from low-return food crops to high-return cash crops.

4. Setfiers were a fairly homogenous group, drawn predominanfiy from
nearby areas. This resulted in less dislocation and more continuity and
cooperation between setfiers.

5. The project stressed colonist leadership and did not demonstrate
paternalism. (301).

She furfirer notes constraints including ”attempts to supplement incomes
firrough off-farm work had led to significant decline in production and land of
community organization had crippled the potential of fire setfiement to respond
collectively to its problems” (302). Two complaints among fire setfiers in
Findley/s Latin American case study include fire need for ” improved access to
transportation and credit” (306). In order to combat fire structural constraints

facing fire setfiers, Findley recommends:

inrprovements in fire internal terms of trade

improvement in transport services

produced-oriented credit

adoption of mixed cropping systems

generation of off-farm employment [but not in fire pioneer phase]; and
expanded healfir and education services (307).

eweri

Henriques (1988) reviewed Brazilian setfiement schemes and comes to
many of fire same conclusions as Findley. Most notably she shows that ”fire
implementation of a colonization policy has not significanfiy enhanced people’s
well being” (338) due to the fact firat ”credit and technical assistance did not
reach more firan one—third of the setfiers”. In the Andean highlands of Peru,
Carpio (1988) found firat several structural failures inhibited the progress of
setfiement schemes, specifically the lack of overall organization, credit, and

marketing facilities.

2.5.3 The Issue of Time and Transition: How long Does it Take to Get an Accurate
Assessment of Orange?

Given firat fire Communal Property Associations investigated in firis study
are less firan seven years old, fire issue of fire time of analysis arises. Had fire
CPAs been in operation for twenty years I could easily argue firat my findings
represent trends, however, it is possible firat I have identified processes
symptomafic of transitions. If fire findings represent transitions firen they must

be addressed as such.

65

Diaw (1990), in an assessment of resetfiement in Ghana, notes firat trends
are set in fire first years of resetfiement and impact all future events:

The period immediately following evacuation could be said to be fire

watershed in fire lives of fire resetfiers.. .firis formative period was going to

mould fire people’s attitude to conditions in fire new town. To be able to
understand firose attitudes would require some indicators as beacons for
future assessment of trends, events, activities and behavioural patterns of

fire population (76).

Voh (1983) however, takes a more conservative approach to the issue of
transition, noting that ”it could be said firat two years after fire relocation, fire
people still have not re-established firemselves” regarding resetfiement in Kano
State, Nigeria. Colson in her study of the resetfiement of the Gwembe people in
Zimbabwe and Zambia reports firat the respondents firemselves felt that the
transition ended in 1962, five years after firey were resetfied and ”when most
Gwembe people agreed that fire crisis had ended” (Colson, 1971, p.141). Koenig,
writing on resetfiement in Mali, argues firat ”people begin to setfie in quickly”
(8), suggesting people recover in fire first two years, and often in fire first year.
One unanswered question in the resetfiement literature is the length of time
recovery takes and when transitions give way to trends. Most studies present a
snapshot of a resetfiement project. The project is judged at time A (usually one to
five years after fire establishment of fire resetfied town or village) with very litfie

reflection on fire phases of transition from fire moment of transfer. The effects of

fire resetfiement are mapped out, but not placed in time. Consequences are

66

viewed as short-term and long-term, but no timeline is established firat marks
progress and failure on a year-by-year basis over any length of time.

It must be noted here firat resetfiement occurs under dramatically different
conditions from place to place. The degree of disruption must be factored into
any analysis of recovery and transition. Involuntary, inefficient resetfiements will
yield longer, and possibly incomplete, transitions than will voluntary, efficient
ones. In fire context of firis study, fire CPA transition has been much less
traumatic, as very few members have resetfied themselves to participate in the
program. The Department of Land Affairs never envisioned large-scale
resetfiement occurring as a result of fire establishment of a CPA. In firis respect,
and given fire evidence in fire literature, an assessment of fire CPA program at
four to six years on can be justified. The literature suggests resetfied people
recover and establish more stable livelihood systems anywhere from one year
after resetfiement (Koenig) to five (Colson). Nelson (1973) argues firat
resettlement occurs in firree stages, fire pioneer stage (5-10 years) where resetfied
persons establish the ”basic life support systems”; fire consolidation stage (5—10
years) where community organizations are forms and more sophisticated social
development takes place; and fire growfir stage (no specific time range) where
socio-economic maturity is reached. These represent longer stages firan found

elsewhere in fire literature. Kinsey and Binswanger (1996) remark firat:

67

resetfiement projects judged to have been successful in one or more

dimensions appear to have passed through a series of three to five year

stages of evolution, during which setfiers adjusted to fireir changed
circumstances in ways firat affected fire performance and impact of the

setfiement project (108).

Bahrin illustrates an important point regarding level of income and time of
transition from a case study of a Malaysian resetfiement scheme:

At fire outset it must be pointed out firat during the first few years after
entry into the schemes average setfiers’ income is usually lower than firey were
receiving previously. This is because fire setfiers receive only a subsistence
allowance of about Malaysian $100 per month during fire pre-production stage of
firree to five years. . .after firis stage, however, the average income increases wifir
fire period of residence in fire scheme (Bahrin, 1988, p.111)

As fire CPAs were much less dramatic transitions, and were facilitated
initially by a capable program of support, I will show via fire literature and data
collected in firis study firat findings from fire CPAs represented in firis study are
trends, alfirough some symptoms of transition are still present and will be noted.
The above literature investigates resetfiement where former livelihoods are
abandoned and new ones are adopted. As noted previously, firis is not fire case
for fire CPAs in firis study. People maintained fireir livelihood systems; firey were

not forced to give up fireir former livelihoods to adopt new ones.

2.5.4 State Farms and Cooperatives: Lessons jbr the CPA Program
While fire resetfiement literature provided a sense of fire time frame
necessary to conduct an analysis of fire Communal Property Associations, it was

noted firat fire structure of fire program was different from fire assumptions

68

made in fire literature regarding resetfiement. In reality, CPAs more closely
resemble state farms and agricultural cooperatives.

Gyllstrom (1991) citing fire ILO describes cooperatives as:

An association of persons who voluntarily joined together to achieve a

common end firrough the formation of a democratically controlled

organization, making equitable contributions to fire capital required as
accepting a fair share of fire risks and benefits of fire undertaking in which

fire members actively participate (Gyllstrom, 1991, p.3).

In theory, fire Communal Property Associations closely resemble
agricultural cooperatives. The history of cooperatives as models of agricultural
production long precedes the CPA program. Under communist control, Eastern
Europe and fire Soviet Union relied heavily on fire cooperative model in fireory
but rarely in actual pracfice. The modern idea of fire cooperative in Africa dates
from fire colonial period, but became more pronounced at independence. Given
firat cooperatives have had a long and diverse history, some lessons can be
elucidated firat can inform fire CPA program. Holman (1990) oufiines some of fire
key aspecb of cooperatives in fire Third World, including fire expectation firat
cooperatives would spur change and fire major problems or criticisms of
cooperatives. He pub cooperatives into perspective noting:

Despite fire somewhat diffuse conceptualization of cooperation, but

strengfirened by bofir the vision of peasant fatalism and by the planning

euphoria firat prevailed during fire 19505 and 1960s, expectations about fire
roles and potentials of cooperatives as development instruments have been

bofir varied and far-reaching. (Holman, 1990, 30).

These expectations have included beliefs firat cooperatives would ”increase

agricultural production, extend fire supply of credit and agricultural inpub, cater

69

for fire poor and minimize social stratification, and establish ’cooperative
awareness’ among members” (Holman, 1990, p.31). These expectations were not

fully met and criticisms arose, including fire argumenb firat cooperatives:

0 bring no structural change

0 do not benefit the poor

0 suffer from bad management

0 are exhausted by government interference
(Holman, 1990, p.33)

Okuneye (1985) seb out several indicators on which fire effectiveness of
cooperatives may be judged, including ”viability over time, facilitation of the
provision of services, promotion of egalitarianism, enhancement of agricultural
earnings and lessening of dependence on foreign imports, and a cost-benefit
analysis” (Okuneye, 1985, p.6—8). An additional indicator of effectiveness, as
Gyllstrom nobs, is firat cooperatives in Kenya were considered dormant if they
ceased to operab for two years (1991, p.98).

The case study librature portrays cooperatives as largely unsuccessful as
well. Gyllstrom(1991) maintains firat cotton cooperatives in Kenya are ”too
rudimentary to allow fire expecbd levels of agricultural output” (p.271). As wifir
Soufir African CPAs, Kenyan cooperatives were ”used by fire stab as an
instrument for promoting agricultural and rural development (Gyllstrom, 1991,

p.267). He nobs firat factors influencing projecb include:

70

Local environment

Local population

Quality of Infrastructure
Level of member education

Economic differentiation
(Gyllstrom, 1991 p.100-104)

Furfirer, fire aufiror argues firat fire performance of cooperatives resb on
fire availability of credit and supplies and the role of fire stab (Gyllstrom, 1991,
p.175-212). Survival of Kenyan cooperatives was due in large part to ”regional
variations in the operational environment” (Gyllstrom, 1991, p.269). The author
summarizes fire situation among Kenyan cooperatives:

In conclusion, the organizational structure devised for agricultural service
cooperatives in Kenya displays fundamental weaknesses when exposed to
fire vagaries of fire environment on which it essentially depends, i.e. fire
smallholder economy. It then becomes evident firat the design resb on a
number of implicit environmenb, such as the ease of interaction in
geographical space, a certain size of operation, management skills, and
stability. These in-built firreshold requirements, imparbd on societies and
unions, can in most cases be met only partially, if at all. Hence in
fundamental respects, firey will condition fire expecbd rationality and
benefib of cooperatives. If accepting firat fire symbiosis wifir government
institutions has favorably influenced fireir survival capacity, it is at fire
same time negatively affecting fireir performance and contributions to
agricultural and rural development (Gyllstrom, 1991, 157).

Kimario (1992) nobd similar circumstances effecting Tanzanian
cooperatives. Marketing cooperatives were largely unsuccessful due to bofir
internal and exbrnal problems. The inbrnal problems included ”lack of support
from members, inefficient [executive] commitbe members, inefficient

management, and the inability to control production losses” (Kimario, 1992, 43-

47). Exbrnal factors effecting fire performance of Tanzanian cooperatives
involved fire government and fire party (T ANU) and included:

Overly ambitious plans

Inadequate personnel

Inadequab provision of transport
Inefficient administration

Low moral among fire administrative staff

Miscommunication
(Kimario, 1992, 48-52)

These findings from Tanzania were replicated in Adeyeye’s study of group
farming in Nigeria. He nobd firat cooperative farming in Oyo stab lacked
effective inbrnal organization and membership participation was poor
(Adeyeye, 1991, 21). Okuneye (1985) supporb Adeyeye’s finding firat Nigerian
cooperatives were largely unsuccessful and illustrates similar examples of failure
from ofirer nafions including Zambia, Uganda, Benin, and Tanzania. In Tanzania
”poor work programmes, mismanagement of funds, small sized communal plob
and poor distribution of communal earnings” led to disillusionment wifir fire
cooperative program (Okuneye, 1985, p.19).

The experience of cooperatives informs firis study on Communal Property
Associations in several ways. First, fire cooperatives librature is more relevant
firan fire resetfiement librature as the structure of cooperatives more closely
resembles fire structure of fire CPAs. This allows for a more direct comparison.
Second, fire performance of fire cooperatives is similar to that of fire CPAs, thus
fire reasons for success or failure can be firought of as broadly similar. Finally,

recommendations and conclusions can be made even before more studies of fire

72

CPAs have been undertaken due to fire fact firat cooperatives so closely resemble
fire CPAs. Many lessons have been learned from fire cooperative movement over

its long history firat can be relabd to the Soufir African CPA program.

2.5.5 Example from the Land Reform Literature

Alfirough it is argued above firat the land reform librature has overlooked
evaluation of land reform projecb from a livelihoods perspective, two empirical
works do review fire impacts of land reform in Zimbabwe. Kinsey (1999)
explains firat fire Zimbabwean land reform program of the early 1980’s did lead
to improvemenb at fire farm level. He noted firat crop output ”is worfir over
four and a half times firat of fire average communal area household” (Kinsey,
1999, p.183). Furfirer, incomes were found to be more variable and average
livestock holdings lower in Zimbabwe’s communal areas firan in fire resetfied
(land reform) areas (p.184-185). Additionally, communal area households are
more dependent on remittances, consumption for non- food ibms is less equally
distributed, and spent more on food (rather firan relying on grain stocks). The
findings from firis study were reinforced using data from the same study areas

collecbd in 1999 (Hoogeveen and Kinsey, 2001).

73

2.6 Hypotheses and Research Questions

The problem firis study atbmpb to address cenbrs on fire introduction of
six land reform projecb into the Northern Province, Soufir Africa. A firorough
assessment of fire impact of firese projecb on fire livelihoods of fire members and
land use of fire transferred land has yet to be undertaken. No clear and decisive
analysis has yet been undertaken to gauge the efficacy of firese projecb - or firis
type of project - in achieving fire stated goal of fire government of Soufir Africa of
rural poverty alleviation and social transformation firrough agricultural
inbnsification. This study will assess fire efficacy of fire CPAs in affecting firese

goals by investigating livelihood and land use change.

2.6.1 Hypotheses made in this Study

Based on fire librature, I have made fire following hypotheses in firis study:

1) The introduction of a land reform project into a rural, poor area will
have the effect of raising incomes and strengthening livelihood sysbms.
The introduction of an additional mechanism to draw on for livelihood
generation may not ibelf strengfiren livelihoods but given firat a high
degree of government support is included in these projecb, and
cooperative work bnds to spread the risks associated among a larger

number of individuals. I assume firat any production firat occurs on

74

project farms will increase income and strengfiren livelihoods particular
in light of fire fact firat members of projecb are not forced, as in
resetfiement schemes, to abandon fireir prior livelihood sysbms. These
land reform projecb represent an additional livelihood option for
members.

2) Projects will not have been abandoned and given fire degree of
government support and general enfirusiasm among fire population for
land reform firat the level of activity and participation on fire farms is
high.

3) A change in fire use of fire land will occur. The change will take fire form
of land inbnsification. The librature suggests firat whitebowned farms
in Soufir Africa are not always fully utilized (Van Zyl, 1996)and given
firat fire transfer of land is from one white farm to a large group of blacks,
I expect to see intensified land use based an increased pressure on fire

land.

2.6.2 Research Questions

I posit here bofir general and specific research questions derived from firese
firree libratures firat I will attempt to address in firis dissertation. While firere are
many more questions firan I will present here, firese are fire key research

questions drawn from fire librature review.

75

First, how do livelihood sysbms respond to a land reform change, such as
firose effecbd in fire Norfirern Province by fire central government? In short, did
land reform in fire Norfirern Province result in more or less secure livelihoods?

Second, did land uses and or cover change after fire introduction of fire
CPAs. Sub-questions include: A) Are the social driving forces of land use change
clearly evident in fire sabllib imagery? B) How useful is the imagery in fire
assessment of land use change in dryland South Africa? And; C) Is it possible to
evidence change reliably and accurately from fire images?

Third, did land reform in Soufir Africa affect any land use and/ or
livelihood change? Is the cause of any change firat may be found attribubd to fire
land reform project alone? Land reform may not equal land use change firat in
turn may not equal livelihood change. Were fire observed outcomes of fire land

reform process intended in fire legislation? Were there spurious outcomes?

Embedded in all of these questions are recurrent questions, specifically:

1) What are the impacb of gender relations on livelihoods firat have been
albred by fire land reform project?

2) How are issues of scale manifest, and what issues can be dealt with at

each scale (household, farm, regional, national)?

76

2.7 Summary

I have presenbd here a firemafic view of three libratures relevant to firis
dissertation. I discussed fire construction of livelihoods and definition of
households firrough the librature, emphasized the importance of how
livelihoods are constitubd, and reviewed fire nofion of gender relations. I
presenbd fire land use and land cover change librature, considered issues of
scale, and relabd major themes to case studies, specifically focusing on Soufir
Africa. I then presenbd the librature on land reform in Soufir Africa relevant to
land alienation, bnure and policy. Finally, I set out specific challenges to fire
librature firat I will reexamine in fire discussion section. These themes were
considered carefully during data collection. Questions on inbrview schedules
and discussions wifir informal groups were purposely designed to examine firese
firemes. The next chapbr details fire provincial conbxt followed by data analysis

and discussion of fire major findings.

Chapter Three

Research Setting and Study Sites

3.1 Introduction

How do the issues discussed in fire previous chapbr manifest firemselves in
fire Norfirern Province? I will attempt here to describe fire provincial conbxt and
to situab my discussion wifirin the firemes elucidabd in the review of fire
librature.

The Northern Province is similar to many ofirer areas of rural Soufir Africa
in ib history and agrarian structure. The residenb of fire rural Norfirern Province
face many of fire same constrainb and opportunities as fireir counterparb across
Soufir Africa. However, more than in any other province, people rely on fireir
own production for susbnance (Lipton, 1996). The Norfirern Province is also one
of fire drier provinces of fire country and is prone to drought, especially during
El Nino Soufirem Oscillation (ENSO) evenb. Wifir an average annual
precipitafion of 381mm -635mm, only fire Norfirern Cape and N orfirwest
Provinces are dryer (Tyson, 1987).

The Norfirern Province is also fire poorest in South Africa (Ievin,1996).
Demographic pressures include the highest population growth rate in South

Africa, high rates of infant mortality, and strong out-migration to fire mining

78

sector (levin, 1996). Furfirer, nearly seventy percent of fire population lives
below fire poverty level (Kirsbn, 1996). Income is derived firrough many sources,
including fire sale of crops, remittances, non-farm activities, and pensions (Baber,
1996). Labor migration to mines and fire industrialized areas to fire soufirwest
ofbn foments hardship in fire rural areas. The head of household is ofbn gone
for long periods of time, leaving women, children, and the elderly to work on fire
farms. Often, renrittances fail to meet the family needs, leaving women to seek
income firrough trading in fire market towns or more importanfiy to maintain a
small plot of land on which fire family can subsist. N gqaleni and Makhura (1996)
illustrab firis, noting firat women perform fully seventy percent of the labor on

small-scale farms in fire Norfirern Province.

3.2 Demographics and History

3.2.1 Political Geography

As with all of Soufir Africa’s nine provinces, fire N orfirern Province is
administered from a provincial capital. Magisterial districb divide fire province
to facilitate local government. Towns fall under magisbrial districb and fire
lowest political aufirority is fire Transitional Local Council. The lowest politico-
geographic construct is the farm. Farms are codified unib with precise
boundaries, regisbred in Pretoria, and often sub—divided into ’portions’. Thus a

’farm’ is bofir a political and production entity.

of.

Pr,

The Northern Province’s largest city is Pietersburg, which is also the
provincial capital. Important cities include Tzaneen in fire east, Potgiebrsrus in
fire cenbr, Tlrabazimbi in the extreme southwest, Messina in fire norfir,
Thohoyandou in fire norfireast, and Phalaborwa in fire east. The norfirern half of
fire Kruger National Park is admirrisbred by fire province, but provides less
revenue to fire N orfirem Province firan the soufirern half of fire park provides to

fire Mpumalanga province.

3.2.2 Structure of the Population

The population of fire Northern Province is 4,929,000. The major languages
spoken are Sepedi (northern Sofiro), Xibonga, Tshivenda, and Afrikaans.
Northern Sofiro people inhabit the central and soufir/soufir-western areas of the
province. The Venda people are concentrabd in fire northeast wifir fire Tsonga
people in fire east abutting Kruger National Park. Afrikaners (descendenb of fire
first whib setfiers) live throughout fire province and small minorities of Tswana,
Mswati, Ndebele, and Mashona peoples are inbrspersed among the main efirnic
groups.

Alfirough spatially concentrabd under apartheid into homelands, recenfiy
a great deal of efirnic mixing has occurred. The ethno-racial boundaries imposed
on fire norfirern Transvaal under aparfireid have broken down in fire transition to
fire new Northern Province. Several key areas of efirnic heterogeneity include

Piebrsburg, fire University of fire North, Tzaneen, and fire soufirern boundaries

80

 

as

of fire province, especially those wifir fire Norfirwest Province (formerly
Bophufirabwana).

One of fire major demographic and economic phenomena in fire Norfirern
Province is fire high degree of dependence on labor migration. Several studies
estimab the number of households wifir members involved in labor migration as
high as 70% (Baber 1995; Statistics South Africa 1997). Overwhelmingly, fire bulk
of migration is to the major industrial areas of Gaubng Province. Migranb will
leave fireir homes in fire north for monfirs at a time, ofbn failing to send
remittances, leaving rural households vulnerable, wifirout cash, and suffering

from a lack of labor.

3.2.3 Economy, Infiastructure and Development

Only fire Norfirern Cape province contributes less to the national GDP firan
does fire Norfirern Province. A lack of manufacturing, few natural resources, and
a large, poor, rural population explains fire very low 3.7% contribution the
province makes to the national budget. The only significant concentration of
industry is at Piebrsburg and firis represents only a handful of opertafions.
Milling and brewing are fire main industries in Piebrsburg. The province has
exbnsive timber, ba, and fruit plantations in fire escarpment, almost exclusively
owned or run by fire whib population.

A well-distribubd road network serves fire province, however, roads are in

a slab of decline. After the 2000 floods, many of the rural areas of the norfireast

81

 

and east became inaccessible by road. Many roads were in fire same state a full
year labr. Rail transporb goods, but very few passengers. The N1 national
highway links the province to Zimbabwe at Beitbridge and to fire
Pretoria / Johannesburg area.

According to fire Premier of fire Norfirern Province, Ngoako Ramafihodi,
41% of fire population of fire province was unemployed in 1998, wifir
government being the largest single employer at 25% of active workers (Soufir
Africa Yearbook, 1999, p.17). The Maputo Corridor promises to link the province
to fire port of Maputo as a stimulus for development. Ofirer firan firis proposed

road, no significant development projecb have been undertaken in fire province.

3.2.4 Physical Geography, Weather and Climate

The Norfirern Province is split almost in half by fire Drakensberg
escarpment. To fire east is the lowveld- a hot and humid plain. The western
regions encompass fire highveld - a hot and dry plain. The bulk of the rain in fire
province falls in fire escarpment region. During the rainy season in firis region,
weeks can pass where the dominant weather phenomenon is eifirer rain and / or
fog. Unpredictable weafirer patterns increase the likelihood of crop failures. Only
fire Norfirern Cape and Northwest provinces are drier on average. Soils are firin
and erode quickly. Trees are scarce oubide of fire escarpment. Vegetation cover

is primarily a savanna-scrub.

82

3.2.5 Political History

The Norfirern Province has undergone a rapid political restructuring in fire
past six years. For most of its history, it was called the norfirern Transvaal, part of
fire Transvaal province, and was administered from Pretoria. The north was
looked upon as rural and agricultural and firerefore secondary to fire needs of
urban, manufacturing areas of Pretoria and Johannesburg. A disproportionab
number of the homelands were located in the norfirern Transvaal. Venda,
Gazankulu, lebowa, KaNgwane, KwaNdebele, and part of Bophuthabwana
consumed a large portion of the land surface. Of South Africa's current nine
provinces, only fire Eastern Cape and KwaZulu-Natal had more of fireir land
area set aside for homelands.

The lebowa and Gazankulu self-governing stabs and fire so-called
independent Republic of Venda were politically unstable and socially volatile
firroughout fireir history. The 1959 Sekhukhuneland revolt was one of many
rural protests against fire ofbn brutal regimes in Iebowakgomo, Giyani and

Thohoyandou (capitals of fire homelands respectively).

83

3.3 Census Data on the Northern Province

I will present here findings of a rural household survey conducbd in fire
former homelands of the N orfirern Province in order to present a picture of fire
regional conbxt. This survey focused of fire living conditions and livelihoods
sysbms and was conducbd by Statistics Soufir Africa, fire stab census agency. I
focus in fire next chapbr on fire specific livelihood sysbms and land use patbrrrs

among farms selecbd for firis study.

3.3.1 Demographic Data

In fire Stafistics Soufir Africa survey, a total of 1090 people were questioned
in fire former homelands of lebowa, Gazankulu, and Venda. The majority (87%)
resided in a rural area. Only 11.6% resided in a semi-rural, denser setfiement. In
fire rural areas, the majority of households have one to two dwelling units. The
majority of housing type is a brick or concreb house (57%) wifir a large portion
(39%) of residential dwellings considered ”traditional” (a structure made of mud
or firatch mabrials).

Provision of services is inadequab but not complebly absent. Thirty-two
percent of households have electricity and 28% have wabr. For firose households
wifirout wabr, no clear clusbring of distance to the nearest wabr source is

evident. For instance, 17% had wabr less firan 100 meters from fireir house, 13%

had wabr between 200m and 500m, while 18% had to fetch wabr from more
than 1km away. The majority of households reporbd having a pit latrine (75%).
The majority of respondenb reporbd between three and eight inhabitants
in fireir household. By gender, the composition of households is similar to ofirer
studies. The proportion of households wifir no male adults is 24% while those
wifir no female adulb is only 5%. Many households did report have at least one

male and female adult present (43%, 42%).

3.3.2 Land

Of fire 1090 respondents, only six percent reporbd having been moved
from land since 1913. This is not unfounded given firat fire respondenb to firis
survey were drawn from fire former homelands. Forced removals occurred in
whib lands. This survey found that fire homelands were not used as the vast
labor reserves to fire exbnt previously believed. This is most likely an artifact of
survey design and sampling framework. Had the survey been conducbd near
fire borders of fire former homeland, I expect firis statistic would be very
different.

The largest mefirod of land allocation in the rural areas is through a local or
tribal aufirority, and 65% of respondenb had acquired land in firis way. Smaller
proportions of respondenb access land firrough inheritance of a tribal land
(15%). Surprisingly, 31 % of respondenb said firat they did not have permission

to occupy fire land firey currenfiy inhabibd. Only 29% had a tifie deed to the

85

fr

land firey occupied. The market value of the majority of respondents land was
less firan R10,000 (82%), with 57% owning/ occupying land worth less firan

R2000.

3.3.3 Livelihoods

The livelihood questions undertaken in firis study were of very limibd
scope and generally do not present a very robust picture of rural livelihoods,
however, very litfie data of firis type exists on such a scale in Soufir Africa.

One-firird of respondenb earned between R401 and 800 ($US 50-100), while
22% earned between R201-400 (SUSZS-SO) in fire month prior to fire inbrview.
The majority of respondenb nobd firat their primary source of income was from
wages of household members firat actually lived with them (43%), while 28%
relied on old-age or disability pensions (or someone wifir such a pension).
Twenty percent of respondenb relied on remittances and only 2% relied on
farming activities as fireir primary source of income. As for second most
important sources of income, 33% relied on pensions, 29% on pensions, and only
8% on agriculture. Agriculture, in fact, only becomes a significant source of
income as a firird source. In fact when queried regarding fireir reason for
farming, nearly all respondenb (94%) said it was to produce enough food for
household consumption rafirer firan as a source of income.

Relatively few participants in fire survey have had formal training in eifirer

farming (5%) or animal husbandry (3% ). Size of farming areas used for field

86

crops ranged from 0.5 to 2 ha. Unpaid family labor is the largest source of farm
labor (76%). The year prior to fire survey, 1996, was a drought year wifir 87% of
respondenb reporting that firey had some sort of crop failure in firat year, due
eifirer to drought (63%) or poor soils (13%). Grazing lands were available to only
37% in fire previous year with 96% of firose wifir access to grazing lands
reporfing fire land was held communally. Small livestock consists mainly of
chickens (75% reporbd 16 or fewer). None of these statistics differ drastically

from studies on the rural Northern Province (specifically see Baber, 1996).

3.4 Land Reform in the Northern Province

The N orfirern Province has experienced the slow and unpredictable pace of
land reform firat has effecbd fire rest of South Africa. Reform in fire Norfirern
Province to dab consisb of eighben CPA projecb involving 3,518 households. In
firis study, CPAs were randomly selecbd and stratified by size. The 2000 floods
rendered two of fire farms scheduled to be visibd inaccessible firus albrnabs
were chosen. Of fire CPA projecb selecbd, fire beneficiary list reflecbd fire
involvement of a total of 763 households. Afbr conversations wifir the
chairpersons and secretaries of firese CPAs, first-hand observations, and perusal
of participation records it was found firat only 326 households could be
considered acfive participants. Thus, only 43% of households involved in land

reform projects were active. At the provincial level, given fire number of projecb

87

and beneficiaries, we can assume firat only 1512 households are acfive in fire
CPA program in fire Norfirern Province. Knowing firat fire population of fire
province in 1999 was 4,929,000 and given firat fire average household size is five
individuals, a mere 0.002% of fire provincial population direcfiy benefibd from
land reform. Nob firat firese numbers differ sharply from fire official figures. The
official figures assume all households are active, which firis study will show to be
inaccurab.

The study sibs in firis research included five established CPAs and one
control farm. All of fire CPAs were in fire Northern Province and most were
locabd in the central region. Figure 1 shows the location of fire CPAs and fire
control farm. The CPAs were chose at random afbr being stratified by size. Two
of fire original list had to be excluded as firey were rendered inaccessible by the

2000 floods. Two albrnabs were selecbd.

Mahlambandlovu CPA

Mahlambandlovu CPA is locabd approximably 20km from Piebrsburg
just across fire border of fire former lebowa homeland. The farms firat constitub
the CPA were purchased by fire communal association from a whib farmer in
1997 and formally established as a full-fledged CPA in 1998. The
Mahlambandlovu CPA lies in a drier area of the province. The membership
Stands at 396, of which fire secretary estimabs fewer firan 160 active members.

Afbr firree monfirs of constant contact I was only able to inme 120 members

88

(which still provided enough respondenb to satisfy fire sample design). The CPA
has a small truck, two tractors, and an array of farming implements. They raise

chickens and catfie on fire bulk of their land.

Monyamane CPA

The Monyamane CPA is locabd 15km north of the University of fire Norfir
on fire border between fire Piebrsburg and Thabamoopo magisbrial districb,
which formerly served as the border between fire Transvaal and lebowa
homeland. The CPA was established in 1997 and officially serves 201
households. The farms firat constitub the CPA lie in exceedingly sbep brrain.
Accessibility is severely limibd and much of fire land can be reached only by
foot. The members of fire CPA reside just to the norfirwest of fire CPA itself. The
area represents a transition zone between the wetbr escarpment to fire east and

fire drier highveld to fire west.

Baroka CPA

The Baroka CPA lies between Potgiebrsrus and Roedtan on fire N11
highway. It is fire second smallest of fire CPAs visibd in bofir size (Sha) and
membership (41 people). The farm was established in 1997 and finalized in 1998.
The farmland was part of a larger whib farm firat was sub-divided and sold. The
members have construcbd a small irrigation scheme firat provides wabr to a

Small field producing vegetables. The other portion of fireir land has been set

89

aside for bean, maize, and tomato production. The farm is in fire drier part of fire

province.

Rondebosch CPA

The Rondebosch CPA is situabd in fire escarpment along fire Duiwelskloof
Road. The thirty members of fire CPA reside in fire Sekgopo community, which
lies just to fire norfireast of the actual CPA land. The farm was established in 1997
and finalized in 1998. A whib farmer had previously owned the farm purchased
for fire CPA. The Great Ietaba River separabs the community from the farm. In
lab 1999 and early 2000 devastating floods struck norfirern Soufir Africa,
including the Sekgopo community and Rondebosch CPA. Widespread damage
was observed at fire farm and community including the destruction of most of

fire road network, fire CPA’s wabr pump, crops, and livestock.

Muyingiseri (Trust) CPA

The Muyingiseri Trust is locabd northeast of Tzaneen on the Phalaborwa
(R71) road. The CPA was established in 1997 and finalized in 1998. The trust is
the smallest in land area of all of fire farms visibd and membership officially
stands at firirty-nine. At fire time of interviews the road into fire farm was

complebly inaccessible, also due to the 2000 floods. Production on fire farm was

limibd to a small stand of tree crops and a few vegetable gardens.

Platklip Control Farm

The Plafidip farm in firis study is also referred to as fire ”control farm”. This
area repr a ”typical” farming sysbm, rather firan a farm per se. The brm
”farm” comes from fire fact that all of South Africa is divided into ”farms”
whefirer functionally a farm unit or not. This particular community is cenbred
on fire farm named Platklip, but in reality fireir farming sysbm exbnds beyond
fire Platka farm ibelf.

Estimabs derived from hand counts undertaken by researchers at the
University of the North put fire number of inhabitanb at approximably 1000-
1500 persons. The area was part of fire former lebowa homeland in fire Nebo
District. The farming sysbm is representative of ofirer areas of rural Soufir
Africa. The land is used primarily for farming and livestock ranching. The
residenb are incorporated into the larger South African labor market, wifir many

respondenb reporting a member of their household off to work in Guabng.

3.5 Summary

The statistics presenbd here reinforce fire librature on fire livelihood
sYSbms of fire Norfirern Province. The broad patbrns evoked in firis data will be
illI-lstr'abd in fire next chapbr on data analysis of household surveys. The
Statistics Soufir Africa survey failed to collect important information on coping
IINE‘rchanisms to stress and furfirer failed to probe for fire exact nature of

livelihoods. For instance, cabgories of employment are lumped togefirer in

91

”wage income”. I will atbmpt to dis-aggregab firis cabgory to understand fire
dynamics of livelihood change. The broad patbrns presenbd above, however,

were largely corroborabd.

Chapter Four

Data Analysis

4.1 Introduction

The broad, provincial level picture painbd by fire census bureau statistics
in fire previous chapbr will be compared in firis chapbr wifir data collecbd on
five Communal Property Associations and one rural farming area in fire
Norfirern Province. The analysis of fire census data showed firat fire provision of
services in fire rural Northern Province is inadequab, local tribal authorities
allocab land in most areas, fire majority of respondenb earned less firan
R500/monfir, and firat livelihood strabgies are similar to firose presenbd in fire
review of librature as appropriab for fire Norfirern Province.

This chapbr will be divided into three thematic areas covering fire data
COllecbd as part of field research - livelihood sysbms, land use change, and land
mform. Data collecbd includes 1) descriptive statistics on livelihood sysbm
cOtrrposition and land use change; 2) data on perceptions of land use change; 3)

qualitative inbrviews with individuals; 4) a suib of remobly sensed images and
GIS data; and 5) archival information on land reform including government

documents and academic reporb.

93

Afbr a general introduction to the structure of fire research, data on fire
firree firematic areas will be presenbd. I will first present fire descriptive and
cross-tabular statistics on livelihood sysbms and land use. Iwill firen present
data on fire perceptions of land use change, followed by an analysis of land use
and cover change derived from remobly sensed data. Finally, I will report on fire
land reform program as it relabs to the field sibs. Inbrspersed in firis discussion
will be information derived from qualitative inbrviews. Iexpect firis
presentation of data to support my general notion firat it is precisely fire nexus
between livelihood sysbms and land use change firat will give us fire most

insight into processes of change in rural South Africa.

4.2 Structure of the Research and Methods

In order to investigab fire relationship between livelihoods and land, it was
necessary to find farms where land use change had taken place. In fire original

proposal for firis research, it was hypofiresized firat fire bulk of change had

oCcurred outside of the realm of fire stab sponsored land reform program. While
Privab land transactions between Africans and whites have occurred in the
NOrfirern Province, firese transactions involved mainly an elib, emergent proto-
caIJitalist, black farnring class and as such fire implications for the livelihood
sD’Stems of fire masses of poor were extremely limibd. The goal of firis research

Wis not to understand fire dynamics of farm change from rich whib to rich black

94

farmers, rafirer to understand the processes firat might most impact
transformation of fire bulk of fire historically poor, disenfranchised African
population. Thus, fire original hypofiresis was mistaken. Land redistribution that
effecbd poor Africans was complebly wifirin the confines of the stab program
on land reform. Anecdotal evidence poinbd to four areas where land invasions
had taken place, but under further investigation communities were found
unwilling to discuss the extra-legal sbps firey had taken to acquire land. It
became evident firat fire focus of fire research would have to be on the stab-
sponsored projecb.

The stab-sponsored projects represented land redistribubd from whib
capitalist farmers to nacient African farmers by fire central government under a
comprehensive land reform program. Under the Reconstruction and
Development Programme a once-off R15, 000 (~ $2140) social grant is provided
to each historically disadvantaged citizen. Finding whib farms far too expensive
for individuals or even communities to purchase out-right, fire Department of
Land Affairs facilitabd fire land redistribution through fire Communal Property
Association (CPA) program by encouraging Africans inbresbd in farming to
pool fireir resources in order to purchase productive farms“.

Testing fire relationships between land use change and livelihood sysbms
on stab-sponsored projecb, however, was not enough. How are we to know firat
processes uncovered firere were different from firose in ofirer rural areas? How

could I conclusively evaluab the stab-sponsored land reform projecb and assess

95

any difference from ofirer rural areas? Could these farms alone clearly indicab
positive or negative feedback to livelihoods from land reform and use change?
These lingering questions led to an albration of fire initial research design. A
”typical” rural area was selecbd as a ”control farm”. The control farm selecbd
was fire Plafidip farm in the Nebo magisbrial district in the former lebowa
homeland. The people in firat area have experienced no land redistribution,
rafirer have had continual access to land under communal arrangemenb. If firere
is such a firing as a ’typical’ rural farming area in the Norfirern Province, the
Platklip Farm might be considered such. The area is under bofir communal and
freehold land bnure sysbms, firus represenb many ofirer rural areas in first it
has overlapping land righb. Most importanfiy, asking fire same questions on the
control farm served as a mefirod of differentiating fire redistribubd farms from
ofirer rural areas to isolab fire effer of fire land reform program. It also
provided an addifional mechanism for supporting or rejecting fire first
hypofiresis of fire study, namely that land reform projecb will raise incomes and

strengfiren livelihoods.

4.2.1 Methods
Specific mefirods included (lisbd sequentially):

1. PILOT STUDY - A pilot study was conducbd to refine questions
and debrrrrine fire validity and conbnt of questions.

2. FARM SELECTION - Selection of CPA farms from fire land
reform program list for the Norfirern Province was firen conducbd. Farms
were purposely selecbd by size of membership. Two large farms, one
medium farm, and two small farms were selecbd. Participanb were
selecbd randomly from farm lisb.

3. QUANTITATIVE INTERVIEWS - An inbrview schedule was
conducbd on the selecbd farms. The CPAs were stratified by size for
inclusion into fire study and selecbd randomly. Respondenb were drawn
randomly from membership lists for each CPA.

4. QUALITATIVE INTERVIEWS - Informal discussions and follow-
up meetings were conducbd to get participant reaction. Qualitative
inbrviews were also conducbd with key informanb. By key informants, I
mean firose people who were in pivotal positions who could help me
understand fire actual transfer process. Other key informant included firm
participanb who were active in the entire process, and not in a position of

aufirority in the CPA.

5. CONTROL GROUP INTERVIEWS - Quantitative and qualitative
inbrviews were conducbd on the ”control” farm in a manner similar to
fire CPAs.

Furfirer, each CPA and fire control farm was identified and geo-referenced.
Maps were provided to fire CPAs for farm planning. Soil samples were taken on
the farms firat undertook agriculture. A compleb soil survey of fire control farm
was obtained from fire Department of Soil Science, University of fire N orfir. The
CPAs and control farm included in firis study are lisbd in Table 4.

Table 4 : Specific Study Sibs in the Northern Province.

 

 

 

 

 

 

 

 

 

 

 

PROIECT LOCATION HECTARES
Baroka Roed tan 214
Mahlambandlovu Seshego 2159
Rondebosch Sekgopo 341
Monyamane / Bjafiadi Dikgale 5587
Muyingiseri Trust Tzaneen 147
Plafidip (Control ) Phokwane Not a contiguous

 

All statistics presenbd below are derived solely from fire inbrview
schedule. All variables were summarized and relationships in fire data have been
identified using Chi-Squared bsb. The bulk of fire data is nominal, yes/ no data
and as such is not suitable for regression analysis. In all cabgories, fire data will
be distinguished between the CPAs and fire ”control” farm (CF). Data from all
CPAs has been aggregabd togefirer. Discussion of major findings will follow in

the next chapbr.

The two business plans detailed here were obtained direcfiy from fire CPA
commitbe. Of fire five CPAs visibd, only two had access to their business plans
or constitutions. The ofirer three had no readily available copies of fireir plans,
and in fact, several members requesbd firat the aufiror assist firem in obtaining

copies of firose documenb from the DLA.

4.3 Interview Data

4.3.1 Demographic Data

Demographic questions were asked in order to cabgorize and differentiab
respondenb. Data on participant gender, marital status, age, current income,
prior income, number of household occupanb, lengfir of residence, and prior
residence was collected.

On CPAs 54 % of respondenb were female, while 70% were female on fire
control farm. The majority of respondents were married (75% CPA; 64% CF),
and fire control farm showed a statistically normal age distribution. CPA age
structure bnded toward older age cabgories.

hrcome clusbred in bofir cases toward the lowest cabgories, under 300
Rands ($40) per monfir (45% CPA; 57% CF). The CPA had more people wifir low
incomes firan did fire control farm. In both cases, over a firird of respondents
reporbd family members working in Gaubng Province (35% CPA; 34% CF).

One quarbr of CPA respondenb had a family member working in Piebrsburg,

99

fire provincial capital. The same was not true for the control farm. It is relatively

isolabd from secondary cities such as Piebrsburg.

4.3.2 Farming as Income Generation

When asked what firey spent most of fireir time doing to generab income,
fire majority of respondenb replied firat firey farmed (62% CPA; 85% CF). Nob
firat firis does not equab to the individual’s largest source of earned income.
Rafirer firis question asks how much time is spent pursuing a particular income
source relative to ofirer income options. Although, the majority said they spent
most of fireir time farming, fire majority also responded to a follow-up question
firat farming contribubd ”very litfie” to their actual cash flow (46% CPA; 56%
CF). Only 17% (CPA) and 15% (CF) respondent firat farming provided firem wifir
”most” of fireir family income. Farming as a method of income generation is
clearly not fire most important source of income on either CPAs or fire control
farm. Given firat CPAs were established to promob development and transform
fire rural proletariat through productive agriculture, firis finding stands out as
significant. As a consequence of firis line of questioning, several respondenb
volunbered to participab in qualitative discussions firat undertook to investigab
why people continue to farm although it provides very little of their income.
Responses included: ”I have just always farmed”; ”We Africans can not fail to
farm”; ”Blacks who do not farm are lazy”; ”Ofirer people will take my land”; ”I

must keep busy so I farm”; and ” I can’t get employment elsewhere”. Clearly

100

firere is a strong social imperative toward farming, regardless of ib
unprofitablilty as a time versus income generating activity.

Crops farmed included maize, beans, melons, and vegetables. This
represents no significant variation from findings produced in ofirer studies or
from fire authors own research in KwaZulu-Natal (McCusker, 1997). The
majority of respondenb in bofir cases said firat their family farmed as a unit and

firat they farmed all of fire time.

4.3.3 Identification of Food Shortages

The first of several distinctions between fire inbrview schedules as
admirrisbred on fire CPA and control farm occurred when atbmpting to identify
periods of food shortage. The distinctions occurred to accommodab fire differing
conbxt between fire CPAs and the control farm. Originally, fire inbrview
schedule read ” when have you and your family experienced a food shortage
since you joined the CPA” and ”has firis occurred more ofbn since you joined
fire CPA”. This line of questioning was clearly inbnded to assess fire impact of
CPA membership on livelihoods. However, firis question would not make any
sense to participanb on fire control farm. Most of firem had lived in firat area fire
majority of fireir lives. The question was modified to ”how many times have you
experienced a food shortage since 1980”. In response, 45% of CPA respondenb
reporbd firat firey have experienced a food shortage since joining, but first fireir

food situation is better now firan before joining fire CPA. Their food availability

101

was assessed firrough a suib of questions regarding fire price of food, the
number of times firey had to rely on fire shops for food, and fireir general
perception of whefirer or not firey are ”better off" since joining fire CPA. On fire
control farm, 19% of respondents have experienced a food shortage since 1980,
wifir most respondenb experienced shortages in the period between 1993 and fire
present. Generally, food shortages were found to have occurred in fire periods
1980-83, 1988-89, 1992-4, and 1998-99 and affecbd widespread parts of fire

Norfirern Province, including fire study sibs (Baber, 1996).

4.3.4 Coping Mechanisms to Food Shortages

Coping mechanisms to food shortages in fire Norfirern Province show
remarkable lack of resilience. As in KwaZulu-Natal, mechanisms are largely
confined to reliance on shops, shop credit, and a very linribd suib of
accompanying mefirods (McCusker 1997). On CPAs fire frequency of visits to
shops markedly increases in times of stress (from one a month in normal times to
every ofirer day during shortages). The control farm showed more stability in fire
reliance on shop-bought food. Both cases press fire argument for structural
dependence of rural people on shop-bought food. This will be discussed furfirer
in fire next chapbr.

Purchasing food at fire nearest shop is fire first coping mechanism to food
insecurity on bofir fire CPAs and fire control farm (54% CPA; 68% CF). Fewer

people relied on credit purchases as their first response to food shortages (14%

102

CPA; 24% CF). The second method of the majority of respondents was borrowing
from neighbors (CPA 17%; CF 56%). The firird sbp was asking relatives (CPA
22%; CF 5%). As in KwaZulu—Natal, few coping mechanisms could be identified
past fire firird mefirod. In discussions wifir participanb, the aufiror’ s concerns
regarding these findings were presenbd to discussion groups who responded
firat fire majority of people were almost complebly dependent on fire shops,
credit, and their neighbors/ relatives in times of need. No wild food gathering or
asset sale could be identified. When queried about asset sale, many participanb
responded ”we have nofiring to sell”. Observation of households by fire aufiror
confirms the relatively low number of productive assets. Of fire randomly
sampled members of fire largest CPA, only two were observed to have access to
tractors. Asset sale in brms of livestock will be discussed in more detail in fire
next section, but generally have very litfie impact. Respondenb did recall
slaughbring goab and chickens, but reporbd firat firey ”would only do firat if
firey had bad credit and could not purchase from fire shops”. When queried
about fire frequency of animal slaughbring as a coping mechanism, several
respondenb reporbd firat firey did slaughbr small livestock in severe cases of
food shortage. The frequency of such occurrences can not be debrmined wifir

certainty as firis did not show up in fire quantitative sample.

103

4.3.5 Livelihood Systems

The firrust of firis research was debrmirring fire livelihood sysbms for rural
residenb in fire Norfirern Province and relating fire findings to land use change.
A suib of questions was asked to attempt to uncover the exact nature of
livelihood sysbms. As shown above, fire amount of time spent in agriculture is
clearly disproportional to its return as measured by income.

Labor migration is an important facet of rural economies in South Africa,
and was represenbd as such in the study sibs. Industries in Gaubng drew
significant numbers of household members for labor (35% CPA; 34% CF).
Secondary cities provided fewer employment opportunities for fire control farm,
but as nobd above, 24% of CPA respondents worked or had a family member
working in Piebrsburg. Working as farm laborers on whib farms was
significant to only a few people (10% CPA; 5% CF).

A series of questions were asked regarding income generafing activities and
fire importance of the activity to the family income. Table 5 summarizes fire
major findings. The list of activities was debrmined firrough exbnsive pilot
study quesfioning to debrmine the range of activities in which households might
be involved. The list was compiled from all answers garnered during fire pilot
phases and then incorporabd into the questionnaire. Redundancies were

inbntional.

104

As shown in Table 5, members of the CPA have a broader income base firan
do members of fire control farm. The percentage of respondenb reporting
activity per cabgory is much broader among CPA members. That one-firird of
control farm households rely on pensions is remarkable as is fire number of CPA
households firat rely on selling petty commodities.

These figures indicab a diverse livelihood strabgy in among bofir groups,
however, fire members of the control farm are markedly more dependent on
enfifiemenb, eifirer from ofirer family members or the stab.

Livelihood options include raising and selling livestock, particularly catfie.
In bofir cases, fire majority of respondenb on the CPA held between 1-2 catfie (54
%) wifir large proportions of fire control group holding no catfie (63%). Among
CPA participanb, 43% owned only two catfie. Fewer than 2% held more firan 10
catfie in bofir cases. Change in fire number of catfie owned bnded toward losses,
but fire majority reporbd no change (37% CPA; 44% CF). No significant increase

was shown in eifirer case.

105

Table 5 : The Range of Livelihood Strabgies and Relative Importance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Activity “/0 of CPA % of Control
Hand-craft making 1 % (14) 1 % (10)
Brewing Beer 14% (5) 9% (6)
Traditional Doctor 6% (11) 1% (10)
Selling Cooked Foods 4% (12) 2% (9)
Dressmaking 8% (10) 1% (10)
Mine Worker 3% (13) 1 % (10)
Government Employee 24% (2) 13% (5)
Maid / Domestic 12% (7) 9% (6)
Farm Laborer 21 % (3) 22% (4)
Works for Whib Farmer 10% (9) 5% (8)
Sells Fruit 1 1 % (8) 11 % (6)
Pensions 13% (6) 30% (1)
Part-time or hrformal 16% (4) 7% (7)
SellingPetty Commodities 37% (1) 27% (2)
Remittances 16% (4) 26% (3)

 

 

 

 

Regarding small livestock such as goab and chickens, fire majority of
respondenb hold between 0—2 goab (90% CPA; 62% CF) and change has bnded
toward losses but fire majority have experience no change in fire number of goats
firey own (68% CPA; 44% CF). Nobworfiry is that 32% of participanb on the
control farm have reporbd losing between 2 and 11 goab over fire past bn years.

The majority of CPA respondenb (79%) reporbd holding between 0-2
chickens, while fire range of fire number of chickens per household was much
more diverse among fire members of fire control farm. Only 31 % held 0-2
chickens wifir a clusbr of respondents (37%) reporting firat fire held between 5

and 10. The BaPedi people of the study areas belong predominanfiy to fire

106

Zionist Christian Church firat forbids fire consumption of pork, firerefore, pigs

were not included in small stock.

4.3.6 Land Use and Land Authority

This section will deal wifir quantitative inbrview data collecbd exclusively
on fire control farm. Because all land use decisions are made by the governing
commitbe of fire CPA and are set forfir in the business plans and constitution,
firere was no compelling reason to ask individual households about land use
decisions on fire CPA. They had no individual control over firose decisions ofirer
firan voting for or against them. Data presenbd here regarding land use
decisions on CPAs is stricfiy qualitative, obtained from discussions with fire
members of fire executive commitbe and from the business plans firat l was able
to obtain.

Fifty percent of respondenb on the control farm reporbd firat fireir chief
direcbd firem as to which pieces of land to farm and raise catfie. Regarding the
ofirer fifty percent, nearly all (94%) said they made land use decisions on fireir
own, many adding fire knowledge was drawn from their ”past experiences”.
Shifting from one farming plot to anofirer is not common. Once a household is
allotbd a field, firey bnd to keep that piece of land as fireir own (94%). This is not
surprising given firat the area is still largely under communal bnure

arrangements. Fallow is still practiced by 40% of households.

107

Land use change was direcfiy queried. The amount of land a household
controls has been stable in fire period since 1990 as evidenced by fire fact firat
86% of participanb have neither gained or lost land to others, however, a
significant degree of farm-land to range-land conversion seemed to occur. One-
fiffir of participanb have stopped farming a piece of land previously under
cultivation. Farming is viewed as important, but less reliable firan raising catfie,
as 87 % of respondenb have grazed cattle on land firat firey previously cultivabd.

Only 9% of respondenb have, since 1990, increased fire number of catfie
they own. The majority (52%) have neifirer increased nor decreased fireir herd
size, while some reduction in herd size has taken place (20% of respondenb).
These findings would seem to indicab firat farmland on fire control farm is being
abandoned for betbr land or used for more exbnsive purposes such as raising
catfie. At first firis would suggest a large amount of land use change. Under
furfirer investigation, it was discovered that firis was intra-annual change rafirer
firan inter-annual change and the practice is common in Soufir Africa, in fact all of
Africa. When the crops are harvested, catfie are allowed to graze on fire stubble
to provide fertilizer to enhance the next season’s crops. Land designabd for
raising catfie is considered not suitable for farming, therefore, no cultivation
would occur.

On fire CPAs, land use decisions are made cooperatively, between fire
membership and fire farm commitbe and largely adhere to fire business plan. On

fire two CPAs where fire business plans were obtained, fire communities did not

108

deviab from fire plans, alfirough one had not fully implemenbd fire business
plan. The Baroka CPA had taken fire advice laid down in fire business plan and
had undertaken a manageable degree of cultivation given fireir relatively small
number. They also raised catfie on some of the open fields. The
Mahlambandlovu CPA suffered from an inability to obtain an additional piece of
land on which firey had planned to cultivab crops. The secretary of fire CPA
indicabd firat firey had been tied up legally from obtaining fire land. The CPA
possessed a wealth of farm equipment, well maintained and managed, but
unused. The CPA had full time and part-time paid laborers firat it drew from fire
membership. Following the business plan, fire Mahlambandlovu CPA raised
chickens for sale as well as catfie. Community members were required to pay a
monfirly fee to have fireir catfie bnded on the CPA.

On fire ofirer CPAs, general disorganization plagues fire farm operation.
The structure of aufirority between fire chief and the commitbe of the CPA has
been complebly blurred on fire Monyamane CPA. For instance, when fire author
met wifir fire CPA commitbe to arrange fieldwork, it was the chief who headed
fire meeting, rafirer firan fire chair of fire CPA. Members indicabd firat it was fire
chief firat made land use decisions rather firan the governing body of fire CPA.
Upon inbnsive furfirer investigation, no clarity was gained on firis issue.
Conversations wifir another academic investigator confirmed fire general

confusion over ”who was running firings at Monyamane“.

109

The Rondebosch CPA had a unified commitbe and active use of fire
farmland. They indicabd firat firey grew maize (corn) and tomatoes among other
crops such as beans and vegetables. Unfortunably, fire floods firat hit
norfireasbrn Soufir Africa and Mozambique in February 2000 devastabd fire
Rondebosch CPA and fire neighboring Sekgopo community. The Great Ietaba
River rendered fire CPA’s farmland inaccessible early in February. Many roads to
fire Sekgopo village were accessible only by 4x4 and ofirers were complebly
inaccessible.

Members of fire Llwanlemeetse CPA, fire pilot study group, reporbd
inability to access funds necessary to begin farming. Furfirer, fire head of fire CPA
commitbe reporbd fireft of fire community’ s taming implements wifir no
resolution provided by local law enforcement. As such, fire membership felt very
pessimistic about fire prospecb for fireir CPA. The land acquired by fire CPA was
used for raising a nominal amount of catfie. Refer to fire discussion chapbr for

more on fire general situation of firese CPAs.

4.3.7 The CPA and Perception of Livelihoods

People’s perception of how they have fared since joining fire CPA is largely
neutral to slighfiy positive. The majority (39%) said fireir income was about fire
same, while a firird indicabd firey are a litfie betbr off. However, more people
said firey had become ”much richer”(14%) than firose who responded firat firey

had become ”much poorer” (4%). Nearly firree-quarbrs of fire respondenb

110

believe firat, in general, members become wealfirier fire longer firey are a member
of fire CPA.

A large number of respondenb (38%) reporbd firat firey now sell crops, but
had not done so before joining fire CPA, but an equal number reporbd firat firey
have never sold any of fireir agricultural production. In real brms, however,
most respondenb felt firat since joining the CPA food was too expensive for firem
(76%) but firat it is not more expensive firan before joining either (44%). This
indicabs firat while firey feel they are better off, food remains a critical drain on
fireir budget. As an indicator of the impact of the CPA on their own subsisbnce
production, respondenb were queried on their dependence on store-bought
food. One-firird reporbd depending on shops much more, while half said firey

relied on shops somewhat more or about fire same.

4.3.8 Analysis by Gender

Analysis of fire household inbrview data by gender revealed important
differences and corroborabd several assumptions on gender relations. One of fire
most prevalent firemes firat permeab the librature on gender is fire inequality of
income. Men bnd to have much more control over family finances, ofbn
disproportionab to fire amount they contribub to fire total household budget.
Women in firis study were more likely than men to earn under R300 per monfir
(50% of women compared to 38% of men). Men concentrabd fireir activity only

slighfiy more on farming firan did women. In brms of the mix of crops, however,

111

women showed more willingness to plant beans, melons, and other smaller field
crops firan men. This is largely a result of fire gendered division of space firat
allocates ”garden” or small area agriculture to women and ”field” or large area
crops to men. As a result of firis division of space, differences in income arose.
Men were more likely to respond firat taming gave firem ”most” or ”some” of
fireir income while 52% of women said farming provided ”very litfie” of fireir
income.

Involvement in the CPA has increased men’s ability to sell crops, but fire
reverse is true for women. Respondenb were asked if they sell more or less of
their crops since joining fire CPA. While 42% of men responded firat firey did sell
more, 44% of women said firat they had sold crops before but have not done so
afbr joining fire CPA.

When asked about non-agricultural sources of income, women were nearly
twice as reliant on entifiemenb, such as pensions or remittances, while men were
twice as likely to have informal or part-time employment. Dressmaking and beer
brewing were notable sources of income firat were dominabd by women.

Women reporbd firat they were less reliant on shop-bought food firan did
fire men (53% to 31%). Men concentrabd on maize production, which left firem
more reliant on food from the shops than were women (39% to 17% ). In times of
stress, men are more likely to rely on credit from formal food retailers firan
women, firus reflecting Soufir Africa’s long-standing bias against women in fire

credit market. The percepfional data reflecbd firat women were more likely to

112

believe it important to farm to produce food. Men were more likely to see

farming as a mefirod of obtaining cash.

4.3.9 Relationships in the Data

The relationships presenbd here provide information on fire nature of
livelihood sysbms in firis study. The findings here allow for probing and bsting
of specific topics deemed relevant in the librature regarding livelihood sysbms.
Gender, age, and income have been idenfified a key variables for understanding
fire nature of livelihood change, particularly in norfirern South Africa (Baber
1996). These findings allow for further refinement of the impacb of fire CPA on
livelihoods by differentiating not only between age group, gender, and income
group, but by location. Relationships that might occur on fire CPA but not fire
control farm (or visa versa) must be explained. Testing for firese relationships can
uncover key differences between the CPA and the control farm, thus providing
insight on whefirer or not fire members of CPAs are significanfiy and statistically
different from ofirer rural people. This will allow for identification of fire impact
of fire CPA on its membership.

Given fire nature of the dataset, the statistical mefirod employed to probe
for relationships in fire data was the chi-squared bst. This bst only supports or
rejects fire presence of a relationship and does not indicab directionality, as do
correlafion and regression bsb. The nominal and ordinal data collecbd in firis

study do not meet fire requiremenb of either correlation or regression bsb and

113

transformation of fire data to meet fire requirements was beyond fire scope of firis
dissertation. Relationships commenbd on below will be followed by fire
corroborating statistic. As firis dataset is in reality two data sets, one for CPAs
and one for fire control farm, bsts for relationships were run separably. The
resulb will firen be discussed.

The CPA sample size is 286 while 99 persons were sampled at fire control
farm. Table Six summarizes fire variables bsbd in bofir data seb. Rafirer firan
”rummaging” firrough fire data to try to ”find” relationships, I referred to fire
librature to identify variables that would have the strongest controlling
influence on how livelihoods are constitubd namely age, gender, and income. It
is important here to nob firat some variables firat may seem important were
omitted as firey had too litfie variation to produce a meaningful statistic. For
instance, very few respondenb reporbd firat firey did not farm, so an important
livelihood variable, farming, was not included in fire bsb. Of the firree
”independent” variables, ”age” and ”income” were re-coded to produce fewer
cabgories in order to garner a more meaningful statistic. ”hrcome" was reduced
from eight to firree cabgories: poor (R0-100/ monfir), average (R300—600/monfir),
and wealfiry (>R600/monfir). Age was compressed from seven to firree
cabgories: young (18-29) middle aged (30-59), and elderly (above 60).

Overall, more relationships were found in fire CPA data firan fire Control
Farm data. Chi squared bsb on the CPA data showed relationships between

gender and 1) age [AGE]; 2) change in food security [CHGCPA]; 3) fire amount of

114

crops sold [CRPSELL]; 4) food shortages [EXPSOME]; 5) level of income
[INCOME]; and 6) fire hiring of laborers for farming purposes [LABOR2].
Furfirer bsting of fire CPA data found relationships between income [INCOME]
and hiring laborers for farming [LABOR2]. Variables that have a relationship to
age include: 1) fire contribution of farm income to total income [FARMR]; 2) fire
respondenb gender [GNDR]; 3) fire respondenb income [INCOME]; 3) fire hiring
of laborers for farming [LABOR2]; and 4) the respondenb marital status
[STATUS]. A discussion of firese findings will be included in fire next chapbr.

Very few relationships were uncovered from fire bsfing of fire control farm
data. Statistically significant relationships included: 1) age [AGE] and income
[INCOME]; 2) gender [GNDR] and the allocation of work time to specific tasks
[MOST]; 3) age [AGE] and fire allocation of work time to specific tasks [MOST];
3) income [INCOME] and fire allocation of work time to specific tasks [MOST];
and 4) income [INCOME] and marital status [STATUS]. Again, fire findings
presenbd here will be discussed in fire next chapbr.

Of fire central poinb of firis dissertation is to evaluab the expecbd benefit
of transferring land to rural South Africans via the Communal Property
Association land reform scheme. As such, one mefirod of evaluation I have
chosen to use is fire direct statistical comparison of data obtained on fire CPAs
versus firat obtained on the ”typical” rural farm (Control Farm). Table Seven
summarizes fire findings of chi-squared bsb employed to debrmine whefirer or

not membership in a CPA versus living in a typical rural area was relabd to a

115

suib of variables. I am bsting here to debrmine the impacb of CPA
membership. Relationships were found between fire respondenb place of
residence (CPA versus Control Farm) [PLACE] and age [AGE], coping strabgies
[COPING], whether or not the participant farms [FARM], gender [GNDR],

income [INCOME], and labor [LABOR}.

116

 

«058:.

 

 

 

 

 

 

 

 

 

 

«Show So» 2 235:8
262 6an mo> 282 252 232 252 9.253 moon 5:... >6: 12min.
$2. $5. coon 8:38an
282 232 282 282 262 .Nuiuo mo> 2&3 32 So» «a: wSOmaxw
«<8 2: See
me... 3 8:7. 820 50> .0
<2 252 $2 232 <2 .muuo mo> 80. co EOE =8 so» on jmwamo
$65
952 2.02 262 282 282 0.32 no» on 820 fies waomo
team»... 32
232 232 232 262 982 952 5.3 88 so» on to: 02:50
9&0 352
3H: 3....» 33202. 3
$2 252 $2 282 $2 .muuo 06> 333.... 2:33 econ. (nooro
two» 96 a 5 ounces...
252 232 0:02 252 252 232 so» on use. 52: :6... >383
in?» :2...
232 252 2.52 232 252 232 c. :03 £22. 38 «a: 593.?
motor find
(2 $2 .vruo 8> 25: oeoz .muuo 8> Eoeeoaaom 3 oa< mo<
E...“— 3350 (no .Eau 3.3.30 (.3 .Eau .2230 (ac
39¢ 5.3 80¢ 5.3 853:. 5.3 858:. 5.3 20300 5.3 cause 5.3 15:
35:253. 3:28:38. 3:30:88... 3:28:13. 9:25:30: 2:85:38. eon-claim 0385!,

 

 

 

 

 

 

 

 

 

sunfism wait " x. «3:.

I!

117

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$59 39% 366
.vnuo mo> .onmo mm> 2oz ccoz 232 0:02 5:52 38288: 03.52%
0:02 0:02 232 0:02 032 0:02 2:85 3 82:8 350 02.5.0
cmm. 3 non: 3:6 358 o5. .59. Co
_ "no mu> ocoz .Nuuo mo> o:oz .Fuuo mo> acoz $2: 8on :9. on 55> 50$.
«.55
owed. find so» 52. 2 £885
0:02 032 202 .vuuo mo> ocoz .Nuuo mo> 85: 3>o =9. 95... $593
«Ea.
mama: :5. 5o: 2 £285
202 .euuo mo> 8:02 0:02 0:02 0:02 .388 82.. 93 :09)
2:85
5 3:20 52.. 3
032 302 0:02 ocoz 0:02 082 3.383.. mEoecoamv: a0:maoz_
woo.» oE85
$2 $2 $2 $2 202 .N nun. 8.» >_:.:oE £5288: m2002.
<8 2.. 9559.
$2 802 $2 0:02 $2 032 85m 0885 5 3:20 01002.
(a0 2..
$2 962 <2 0.52 $2 232 955°. 223 9:85 8302.
on.»
8:02 .«uuo 9.; 032 0:02 0:02 082 38.5 :020
:5". .0530 <a0 :5: .9530 (:0 .Ean. .9330 (no
"239:. 5.3 358:. 5.? 3a< 5.3 3a< 5.3 "3.3.0 5.3 2330 5.3 2.32
5:23.33: 5:30.33: 5:23.33: 5:35.50: 5.33.33: 5:23.53: 3.3:!an 035:;

 

 

 

 

 

 

 

 

mueflsm 38$ 5 2.3

118

 

 

2:85

 

 

 

 

 

 

 

 

 

 

80.3 5qu 00> 55:2: 3:33:00: 92002.
53. .Fumo 00> 80:00 :020
«2:85
So» 2 235:8
0:02 053:3 008 :05: 3: :2:<....
Ewe Fun 00> 2:0. .5> 00 2:3.
«203:0 s
0.33 50:83: :0» «>35
0:02 Ea: 0:888 88 moves 0:02 :9. on 88 30:2. 0:05
8.80 5:88... 82
0:02 .050: 3:03:88: 03.5.0 with .0 two 00> 5.3 0:8 3.. on 3: 02300
2:85 :0» 5.? 3:0.
252 s 82:8 .23 022.5 232 5:3 2:5 se. 88 5.5.2
358
25 59. co .85 «3%... :2...
0:02 0:08 so» on 30:2. .50.: 0:02 5 :00: «>020 08. 00: $03.2 ,
2:03
8» 50: 8 £880.
condo 3qu 00> 58:0 20:. 2< :03 «no.0 .Nnuo 00> 3:09.88: 3 02 m0(
«.50.... .0330 0:82. «5.0... 33:00 0:03..
(:0 0.3 :. 5:80:50! 03.02 (:0 0:3 :. 5:035:02 0E0...
5.3 5:35:00: 330:0.axm 0.:0...0> 5.3 5:33.00: 330:0...“ 030:0>

 

 

 

 

 

 

 

are 2: a azaouss. " .3 ~35

119

4.4 Assessment of Land Cover Change Derived from Satellite Imagery

4.4.1 Methods and Error Assessment
"Images in this thesis/dissertation are presented in color."

To assess the scope of land cover change on both the redistributed and
control farms, three Thematic Mapper satellite images from 1988/ 89 and three
Enhanced Thematic Mapper satellite images from 2000, specifically path/ row
numbers 169/077, 169/076, and 170/076 were obtained. Table 6 includes specific
details of each scene. These images encompassed all but one of the study areas.
Images were collected that would represent the study areas before and after the
transfer of land from white to black owners. Variation due to seasonality was
controlled by selecting scenes with similar acquisition dates across the two time
periods. One intervening factor that could not be controlled for was the amount
of moisture present in the landscape. As noted elsewhere, the year 2000 was a
flood year, with the heaviest rains falling in February. Lower than normal
rainfall occurred in 1988 and 1989. While not considered drought years by the
national weather bureau, the earlier imagery represented a landscape with much
less moisture present than did the 2000 imagery. Although differences in
vegetation could be expected, three factors mitigate the influence of seasonality
in this analysis. First is the date of acquisition of the images. All of the scenes

were acquired well after the end of the rainy season, with the exception of

120

path/ row 170/076 for the year 1989. Second, the 2000 rains were intense in
February, but ended in late March as is normal. Finally, the images were
classified manually. The interpretation of land cover was undertaken with the
knowledge that 2000 was a rainier season. Furthermore, the primary
differentiating factor for land cover types was spatial patterns in addition to
spectral reflectances. The influence of seasonality, while not completely
overcome, was considered and steps were taken to ameliorate its influence.

Table 9: Detailed Scene Properties

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Scene ID Path Row Date of Rectification Sensor
(Product #) Acquisition RMS
Error
(pixels)
0119907130135CXD1 169 077 06/ 02/ 1988 1.1220 Landsat 4 (TM)
0750009180043_(X)05 169 077 06/03/ 2000 1.2845 Landsat 7 (ETM+)
01100092801180003 169 076 06/02j1988 1.2208 Landsat 4 (TM)
075000918043_0003 169 076 06103[ 2000 1.3996 Landsat 7 (ETM+)
011(XD9280118W1 170 076 03108/1989 1.5888 Landsath)
075(XXJ918(X)43_(XX)1 170 076 04/ 23/ 2G!) 1.4408 Landsat 7 (ETM+)

 

Each scene was geo-rectified (UTM WGS 84 [Zone 36 South] ) and the geo-
rectification was verified across the two time periods to ensure accurate
representation of change. The RMS error of each scene is detailed in Table 6. All
scenes were classified using the unsupervised isodata clustering method.
Supervised classification was also undertaken. Neither method satisfactorily
delineated classes needed for analysis of land cover change in the study areas.
Change in the study areas was anticipated to be largely a fluctuation between

agriculture, grazing, and fallowed or abandoned land. The ”study area" for land

121

cover and use mapping was defined as the Communal Property Association
itself and the areas around the CPA where members resided. Thus, the vicinity,
or environs, included any rural townships where a CPA member resided and its’
associated farming areas and grazing land. Baroka CPA was not included in the
land use and cover analysis due to the expense of acquiring the two additional
scenes. The Baroka CPA farm is exceedingly small and is discontinuous,
therefore, the majority of the farm would fall under the minimum mapping unit.
It would have been the only study site in the scene (path 170 row 077) and thus
the expense was determined to be unjustifiable.

Spectral separation was difficult to achieve and proved to be inadequate.
Figure 27 demonstrates this problem in the area around the Monyamane CPA.
The isodata classification method was unable to differentiate grassland from
agrarian land. Notice on the map that the area delineated as agricultural by the
unsupervised classifier is far different from the polygons designated as such
using the on-screen-digitizing method. The unsupervised classifier was often
unable to differentiate between short grass areas and agricultural areas.

To overcome the inadequacy of the spectral classifiers, the relevant areas of
each scene were digitized manually using on-screen digitizing where the image
is displayed on the computer screen and polygons are created with the mouse
rather than using a paper map and a digitizing tablet. Each area was divided into
one of five land cover classes: agriculture, grassland, forested, residential, and

other. Figures 2, 4, 7, 9, 12, 14, 17, 19, 22, and 24 show the spatial-spectral

122

properties of each study site based on a standard Band 4, 3, 2 false-color image
representation. Note that agricultural lands in the study areas are not rectilinear.
African agriculture is quite different spatially from industrial or white
agriculture. While white/ industrial agriculture is largely a collection of
rectangular and square fields with an occasional center-pivot irrigation circle,
African agriculture is spatially and visually represented as a patchwork of small
fields, often following contours, without definitive boundaries. Grazing land lies
between and beyond farmed areas, however, because of the need for a priori
knowledge of the area to determine which areas are actual grazing lands and
which are simply open grasslands, the classification will remain ”grassland”.
Forests are not widespread in the study areas, save Rondebosch and Monyamane
CPAs, which are in the forested escarpment region. For the other study sites,
forested areas are those in inaccessible areas such as gullies or high peaks, where
farming or grazing is minimal.

The land use change maps were created using ARC/ INFO, ArcView, and
Microsoft Excel. The digitization of land cover polygons was undertaken in a
manner that would reduce error as much as possible. The coverage for each time
period was digitized separately to avoid visually biasing one image based on the
other. The earlier time period was digitized first, followed by the later period.
After the initial digitization a second image analysis iteration was conducted.
The images from each period were compared in the same image viewer (Erdas

Imagine) using a 5,4,3 band combination followed by a 4,3,2 combination. If the

123

boundary of a polygon was found to differ significantly and justification was
weak (meaning spectrally indeterminate) then the boundary was left as created
in the earlier time period. Upon investigation during groundtruthing, the sliver
polygons were found to be especially problematic. Respondents often noted that
when fields were added to an existing group of fields the most common method
was to simply ”add another row of fields to the edge”. Therefore, the error
inherent in the rectification process could not be isolated. When groundtruthing,
care was taken to stay out of the sliver polygons and edge areas. Points were
taken 100 meters from any given boundary. Once digitized, each polygon was
coded in the database to correspond with the visually interpreted land use on the
underlying image. Next, the two coverages were merged using the ”union"
command in Arc / Info with the ”nojoin” option at the minimum fuzzy tolerance.
The union coverage contained the unique identifiers for each polygon from the
two original coverages, but also assigned new unique identifiers in a separate
variable column in the dataset as many of the land use polygons overlapped
creating ”slivers”. Because the unique identifiers for each original coverage
remained in the new union coverage, the coding of the new polygons for land
use could be done manually. In order to generate a ’land use change’ variable,
the land-use codes for the two original coverages had to be added together. To
do this, the land use label for the 1988/ 9 scenes was recoded from single digits to
hns. For instance, the code for grassland was changed from ”2" to ”20” and so

on. This would allow for unique codes to be generated in the final ”land-use

124

change” variable for the union data set. Had one of the land use codes from the
two original variables not been re-coded, the ’land use change’ variable would
not have had distinguishable change classes.

Error assessment of the images was undertaken in two steps. The first step
was to identify highly unlikely types of land conversion, for instance from
residential to wooded. These errors resulted from the on-screen digitizing and
were small discrepancies along land cover boundaries. During follow-up field
work, the accuracy of the image classification and the resulting land use and
cover maps was assessed in the field with the maps created in the first step.
Transects of each area were conducted. At points in each study area, the land
cover (and interpreted use) was recorded using a GPS and visually documented
with a camera. The sample was not random and did suffer from overemphasis
on areas that were diffith to classify during the initial image-processing phase.
The error matrix (Table 10) shows that most errors were derived from improper
classification of wooded land. The error was primarily of commission -
incorrectly assigning pixels to the wooded category. This resulted largely from
the inability to distinguish between wooded and grassland/ residential land at
the border between such cover/ use types. The overall accuracy was 91 .8%.

The error assessment was calculated for the 2000 scene only. Due to the
inaccessibility and expense of the National Land Cover Map of South Africa, no
comparison of error could be undertaken for the 1988/ 89 period. Given that the

2000 and 1988/ 89 data come from the same sensor, at the same time of the year,

125

and with the same haze levels, the level of error for the 2000 period is indicative
of the level of error in the 1988/ 89 period.

Table 10 : Error Matrix

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Agriculture Grassland Wooded Residential Row
Total
tare 9 0 0 0 9

Grassland 0 16 1 0 17

Wooded 0 2 7 1 10

Residential 0 0 0 13 13

Column Total 9 18 8 14 N-49

Column - Clamified

Row I Known

Producer’s Accuracy User’ s Accuracy

(Omission) (Comission)

Agriculture 100% Agriculture 101%

Grassland 88% Grassland 94%

Wooded 87% Wooded 70%

Residential 92% Residential 1(1)%

Overall Accuracy I K-hat -

91.8% 0.893

 

 

 

 

In an attempt, to thoroughly investigate possible classification error in the
1988/ 89 scenes, intensive interviews were conducted. A local resident who lived
in the area and knew the land use history was taken to a given spot where some
classification confusion had entered into the process (usually an area that was a
sliver polygon on the land use change coverage). The person was then queried
about what the land had been used for (if anything) in 1988/ 89. This less precise
method of error assessment was conducted for all of the study sites except

Muyingiseri CPA. While not definitive nor statistically defendable, this method

126

helped to clarify confusion encountered during the analysis of 1988/ 89 land

cover and use.

4.4.2 Findings

In the areas where the most land cover change was anticipated, those farms
transferred under the CPA program, the least change was observed. In the case
of Muyingiseri CPA there is scant evidence of any change whatsoever. Only
marginal shifts in existing patterns were observed. In the Monyamane,
Rondebosch and Mahlambandlovu CPAs extensification and abandonment
occurred. Areas that had been more intensively utilized in 1988/ 89 were less
used in 2000. Areas outside of the actual redistributed farms, however, showed a
variety of land use changes. This ’external’ change is mentioned here, as the
owners of the farms do not actually occupy the transferred lands. The CPA Act
expressly forbids residence of any kind on transferred lands. Thus, the owners of
the CPA live in neighboring areas, with the exception of the Muyingiseri Trust.
The Muyingiseri Trust is located in the middle of an area of largely white
inhabitants and the owners must drive in from approximately 10-20 kilometers
to access their farm. In the following descriptions of land use change on
individual CPAs the term “agriculture” refers to both farming and herding while
”grassland” refers to spatially undifferentiated grassy scrubland. “Grassland"
should not be taken to mean land on which cattle are raised. Table 11 quantifies

land conversion for the study areas.

127

Table 11: Percentage Land Area Conversion from 1988 / 89 to 2000 by Land

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Use / Cover Code
Monyamane CPA
Land Use Category CPA CPA CPA Environs Environs Environs
1989 2000 Change 1989 2000 W
Agriculture 0% 0% 0% 16.3% 14.7% -1.6%
Grassland 11 % 22.3 % +11 .3 % 42.9% 40.7% -2.2%
Wooded 88.9% 75.8% -13.1 % 37.4% 39.7% +23%
Residential 0% 1.9% +1 .9% 3.2% 4.7% +15%
Mahlambandlovu CPA
[and Use Category CPA CPA CPA Environs Environs Environs
1989 2000 Chagngg 1989 2000 ChangL
Agriculture 17.1 % 2.5% -14.6% 18.2% 15.6% -2.6%
Grassland 82.4 % 97.5 % +15.1 % 78.9% 78.5 % -0.4%
Wooded 0.4% 0% -0.4% 0% 0% 0%
Residential 0% 0% 0% 2.5% 5.3% +28%
Water 0% 0% 0% 0.2% 0.4% +02%
Muyingiseri CPA (Muyingiseri has no membership residential environs)
Land Use Category CPA CPA CPA
1989 2000 ChangL
Agriculture 0% 0% 0%
Grassland 71.4% 64.6% -6.8%
Wooded 28.5% 26.6% -1.9%
Residential 0% 0% 0%
Water 0.1 % 8.7% +8.6%
Rondebosch CPA
[and Use Category CPA CPA CPA Environs Environs Environs
1989 2000 Change 1989 2000 Chang;
Agriculture 8.8% 3.4% -5.4% 23.8% 24.0% +0.2%
Grassland 4.7% 0% -4.7% 23.8% 6.3% -17.5%
Wooded 85.5% 94.7% +9.2% 49.2% 46.6% -2.6%
Residential 0% 0% 0% 2.5% 20.6% +18.1%
Water 0.8% 1.8% +1% 0.4% 2.3% +1 .9%
Platklip Control Farm
Land Use Category CF CF 2000 CF Environs Environs Environs
1989 Change 1988 21%!) e
Agriculture 36.5% 37.4% +0.9% 59.1% 56.0% -3.1 %
Grassland 56.8% 48.4% -8.4% 34.0% 35.7% +1 .7%
Wooded 0% 0% 0% 0% 0% 0%
Residential 65% 14.1 % +7 .6% 6.8% 8.2% +1 .4%
* CF - Control Farm

128

The Mahlambandlovu CPA is the second largest CPA in area and the
largest in terms of membership in the study group. As evidenced by Figure 6, the
membership is spread across a large rural area to the west of the CPA. The CPA
is located in the drier area of the province to the west of Pietersburg. Rainfall is
significantly less than on the CPAs nearer the escarpment region. On the CPA
itself, land use change consisted largely of extensification. In the northern area of
the CPA, the pink area represents land that had previously been used as
farmland and has since become grassland. A large area toward the center-east,
represented in light green, indicates that some shift toward agriculture has
occurred. However, examination of the satellite imagery (Figure 2) for 2000 and
my records indicate that the shift was not into farming, but rather herding.

Outside of the CPA, where the membership resides scattered among other
non-members, intensification is evidenced just to the north of the CPA
(represented by light green areas) while extensification can be seen to the south
and southwest of the CPA. Noticeable expansion of settlements (rural townships)
is shown in dark red and dark green. The overall pattern of change on the CPA
itself, then, has been extensification. Although one area shows a change from
grassland to agriculture, the change was not to farming, but rather herding cattle,
which for this study is still considered an extensive activity.

The Monyamane CPA is on the western side of the escarpment region and
although in a relative rain-shadow still receives more rain than several of the

other CPAs. As such, the Monyamane CPA has more wooded area. In several

129

parts of the CPA itself (Figure 11), wooded areas have been cleared to make way
for grassland. The imagery does not provide any indication (i.e. fences, cattle
paths) of the land being used for grazing cattle and at the time of field work in
late 1999 and early 2000 I was told the areas were not specifically used for
grazing cattle. A small residential area developed in the far northwest corner of
one of the farms that is part of the CPA. Overall, very little intensive activity is
occurring on the transferred farmland. This is not surprising given the rugged
terrain of the redistributed farms. I found it exceedingly difficult to even access
the majority of the farm other than on foot.

Outside the CPA, several rural townships sprang up between 1988 and 2000
while others expanded in size. Most notably, and relevant to the CPA, are the
two large settlements just to the northwest of the CPA represented by the letter
"A” on Figure 11. These are the areas where the majority of the membership
lives. It is important to note here that this is the only CPA where any of the
membership moved in order to be nearer the farm itself. This project benefited
the Magoeba people who were displaced from Magoebaskloof near Tzaneen.
Since that time, they were forcibly removed from several locations and scattered
throughout the area between Pietersburg, Duiwelskloof, and Tzaneen. This
project had the effect of re-uniting about thirty of the dispersed community.

Similar to the Mahlambandlovu CPA, extensification is widespread in the
area off the CPA. Large areas of agricultural land have reverted to grassland or

wooded land. Polygon ”B” in Figure 11 represents the farm land utilized by CPA

130

members in the adjacent townships. The land under cultivation has fluctuated,
but not more or less than would be expected in a typical African farming system.
According to interviews with the chief of the Magoeba people and evidenced in
Figure 11, the areas to the southwest of the CPA have seen a reduction in the
amount of land cultivated as people are seeking employment at nearby
Pietersburg and the University of the North.

Tucked deep in the escarpment, the area around the Rondebosch CPA has
experienced a large growth in population since 1988. The dark green area in
Figure 16 shows a dramatic increase in the amount of land area used for
settlement. This representation is somewhat misleading, however, as it would
indicate that no settlement existed in this area in 1988. In fact, the imagery from
1988 does show small areas of settlement. These areas are very small and nearly
indistinguishable from the surrounding grassland in 1988, while in 2000 the area
has clearly become residential. A great deal of change has occurred since 1988 in
areas other than the actual redistributed farm. Surprisingly, given the clear
increase in population, agriculture on the slopes of hills along the southern
border of the study area has receded. In group discussions, this was attributed to
a lack of labor and capital to clear fields.

On the CPA farmland itself, extensification has occurred. Areas that were
clearly used for agricultural in 1988 have reverted to thinly wooded areas. The
membership of the CPA consistently complained about the lack of equipment,

labor, and skills necessary to efficiently utilize the CPA farmland. The mean age

131

of the respondents on the CPA was 50-59 (nearly 50% of the membership was
included in the sample of this CPA). An older population would certainly find it
difficult to farm the land without the assistance of younger people.

Of the CPAs in this study, the Muyingiseri Trust is the smallest in area. The
Trust lies in the escarpment on the wetter east side near Tzaneen. The land
redistributed was a sub-division of a much larger farm. Land use change here
could be described as ”breathing” (Figure 21). Certain areas on the border
between agricultural and wooded areas and between grassland and wooded
areas have shifted, but no significant change in any given parcel of land has
occurred. The center of the CPA was inundated with water from a small stream
during the 2000 floods. The primary land use observed during my visit to the
farm in early 2000 was tree crops and some small areas of maize and bean
production.

The control farm represents one of the best examples of a typical South
African rural farming system. Some of the population still abides closely by the
decisions of the chief while there is a considerable part of the population that
does not, rather abiding by the Transitional Local Council. Tenure is also
fractured between traditional, chief and tribal council - controlled land, land that
is controlled by the Transitional Local Council, and freehold land. The latter
category represents only a very small fraction of the land authority in the area

and is nearer the town of Phokwane than the Platklip farm itself.

132

Land use change in the area, as on the Muyingiseri Trust, can be described
as ”breathing”. Figure 6 shows that land use has been very stable. Few shifts
have occurred and those that have are fluctuations on the border between the
areas of cultivation and those of grassland. The only remarkable change has been

in some expansion of settlement.

4.5 Social Driving Forces of Land Use Change

4.5.1 Land Use Patterns on the CPAs

The original intention of this research was to document and explain
patterns of land use change among farms that had been redistributed from white
owners to black associations under the Department of Land Affairs Communal
Property Association (CPA) program. The primary change was the transfer of the
land from white to African owners and as such, it was expected that production
would intensify due to pressure from the membership for access to the land. In
all but one of the CPAs, agricultural production was reported to be less than
before the transfer. The reasons ranged from lack of membership commitment to
unclear division of authority between the chief and the CPA chairperson. The
smallest CPA, in fact, was the most productive. CPAs range is size of membership
based on the price of the farm the committee wishes to purchase. The larger
and / or more potentially productive the farm, the more expensive it becomes. As

a result, more individuals are needed to pool their grants to purchase the farm.

133

The Mahlambandlovu CPA has a membership of 396. The farm was used to
raise approximately 500 cattle and 200 chickens. No farming had been
undertaken.

At Monyamane CPA, the land was used for farming and herding. This
CPA, however, suffered from conflicts between the committee of the CPA and
the traditional chief. The chief, for instance, led a meeting with the community to
discuss my work on land use and livelihoods on CPAs. Normally, the
chairperson of the committee would fill this role. It is the chief, not the
chairperson, who makes land use decisions. As evidenced from this CPA,
community level politics factor strongly in land use decisions. This represents
one of the most pressing problems in rural South Africa overlapping claims to
community authority. When explaining community level forces of land use
change, it is impossible to make clear statements on the role of community
authorities, as they themselves are overlapping and often in conflict.

The smallest (and most productive) CPA visited had a membership of 40.
This CPA, near Potgietersrus, had a small irrigation system and raised corn,
beans, and melons. A higher proportion of farm members was regularly present
on the farm. The chairperson of the CPA reported that he had fewer problems in
getting fire membership to participate in farm activities than did the chairpersons
of the larger farms.

In fact, I was unable to examine fire social "driving” forces of land use

change on fire CPAs themselves due to the fact that very little or no change had

134

occurred on these lands. No change means no driving forces. However, it also
means firat there are forces hindering land use change. These forces will be

discussed in detail in Chapter Six.

4.5.2 Land Use Patterns on the Control Farm

The Plafidip control farm allowed for testing of hypotheses regarding fire
role of traditional authority on land use and allocation. The Communal Property
Associations have an elected chairperson and committee rather than a chief. As a
result, CPAs tend toward greater democracy than under a traditional authority,
alfirough the chieftancy should not be viewed as exclusively anti-democratic.

When queried, only 49% of respondents said firat their chief instructed
firem on where to farm. The traditional aufirority controls fewer land use and
occupation decisions firan expected. The overwhelming majority of respondents
noted that they have always farmed fire same fields (93%) and 39% have at some
point let fireir land go fallow. The amount of land a household controls has been
stable in fire period since 1990 as evidenced by the fact firat 86% of participants

have neifirer gained nor lost land to others.

4.5.3 Perceptions of Change
Participants’ perceptions of land use are rooted in a more complex system
of societal and tenure arrangements. Lindblade (p.32 of firis document)

highlighted fire importance of perceptions in land use change. Members of fire

135

Community Property Associations differed slighfiy from participants on fire
control farm. When queried if farming (for one’s own consumption and trade)
was still important to firem, 91% of CPA participants and 92% of the control
group responded positively. However, far fewer (49% -CPA; 21 % -Control Farm)
agreed wifir fire statement ”do you firink firat even if people can’t farm or don’t
have fire money to farm firat firey should still have access to Ian ”. The primary
reason for fire strong disagreement with this statement among bofir groups was
firat firose wifirout money would have a difficult time raising fire necessary
capital to undertake farming. Simply possessing of land does not explain fire
widespread push for land reform and can be ruled out as a possible social
driving force. I distinguish here between land hunger as fire desire for land
regardless of what is fire intended use and land hunger firat assumes active use
of fire land. It would be logical to expect that communities feel their land should
be returned to firem regardless of how they would or would not use it since it
was alienated from them (often brutally) under aparfireid. However, firese
communities View land as a productive asset first and a social asset second.
Simply put, land must be used. The social value of land (our ancestors lived here,
we have relatives buried here) has been subsumed by fire production efiric.
Under apartheid, subsistence farmers were often viewed as unproductive and
irrefficient, firerefore, many rural people still fear firat ”unused” land might be

taken from firem. The difference between the two groups stems from fire fact

136

firat many of fire CPA members have experienced acute landlessness and are
more willing to grant access to firose in need.

The importance of land for food production is highlighted by fire fact that
fire majority of respondents in bofir groups noted firat fire primary reason they
farm is to grow food for their own consumption (65%-CPA; 71%- Control Farm).
Farming for income generation was a distant second (26%-CPA; 28%-Control
Farm). Thus, subsistence farming can be considered a social driving force.

Heritage and lineage are strong societal forces in African society. When
queried, fire overwhelming majority of participants responded positively to fire
statements ”do you firink it is important to farm because your ancestors farmed"
(95% -CPA; 98%—Control Farm) and ”do you firink it is important to farm because
not many young people are farming anymore” (93% -CPA; 98%-Control Farm).
These strong cultural perceptions are likely drivers of change.

Researchers often note that apartheid skewed fire system of land
distribution, leaving Africans landless and unable to farm (Smifir, 1990).
Accordingly, fire majority of respondents agreed with fire statement ”do you
firink it is important to farm because the land was taken from people under
aparfireid and it is now their right to farm” (90%-CPA; 97%-Control Farm). The
return of land alienated under apartheid also drives change. People all over fire

province wait for fireir land and plan how they will make use of it.

137

4.6 Summary

The data needed to tell the story of CPAs in fire Northern Province has been
presented in firis chapter. I have shown that demographically fire sample
population included in fire interview schedule was similar to that in the official
census. Marital status, income, age, and gender in fire study population were
found to be typical of fire province. High levels of labor migration were found.

The majority of respondents farm, but it was not found to be as important
to income generation as ofirer activities. Just under half of CPA members have
experienced food shortages, but report firis happening less since firey joined fire
association. However, firey are more dependent on shops, indicating that shop-
bought food has become a more important coping mechanism. In contrast, only
20% of inhabitants of fire control farm reported a food insecure situation since
1980, and in firis respect, are better off than their counterparts on fire CPAs. The
level of dependence on shop-bought food as a coping strategy is exceptionally
high (54%) and households in firis study are considered to be structurally
dependent on shops. CPA members have a broader suite of coping mechanisms
firan do fireir counterparts on the control farm. Other coping mechanisms are
similar to firose found in the literature, but much fewer in number and scope to
fire point that serious concerns are raised over fire ability of households to cope

in extreme situations.

138

CPA members have a broader income strategy than inhabitants of fire
control farm. The majority perceives no change in fireir income as compared to
fire period before joining fire CPA.A remarkable dependence on entifiements was
found on fire control farm while CPA members are actively engaged in petty
commodity trade.

Far fewer people on fire control farm raise catfie than do CPA members,
largely do to differential access to land. Land use aufirority on fire control farm is
typical of rural Soufir Africa wifir half subscribing to fire chief for instructions
and fire ofirer half looking elsewhere for instructions on land use including
firemselves and fire local authorities. Land use change was minimal on the CPA
farms firemselves, but change was occurring in fire areas where association
members reside just outside the officially transferred farmland. Land use change
was negligible on fire control farm, representing a stable farming system
(corroborated by fewer incidents of food shortage among residents).

Regarding fire management of fire CPAs, general disorganization prevails
on fire majority of CPAs. They are generally under or improperly capitalized,
suffer from a lack of skilled farmers, do not obtain adequate extension, and
membership participation is abysmal, at best.

Women are far more likely to be poor firan men and are more willing to
plant crops other than maize than men. Accordingly, men were more likely firan
women to report that farming income gave firem most or some of fireir income.

Given firat maize is also a primary cash-earning crop, firis is not surprising.

139

Joining fire CPA increased men’s ability to sell crops, while the opposite was true
for women. Women were nearly twice as reliant on entifiements firan men. Given
fire labor nrigration pattern firis is expected. Women had a far more difficult time
accessing credit than men, leading to higher incidents of poverty.

These findings will be utilized in fire next chapter to address fire research
questions and in Chapter Six to tell fire story, albeit critically, of fire CPA land
reform program and its impacts on livelihood and land use change. Iwill
address issues raised at the end of the review of fire literature and challenge
policy-makers on received wisdom regarding land reform and rural political

economy in Soufir Africa.

140

ChapterJFive

Discussion and Synthesis

5.1 Introduction

I ended the review of fire literature with firree research questions that I will
now take up having addressed fire data in the last chapter. I asked specifically:
how do livelihood systems respond to a land reform change, such as firose
affected in the Norfirern Province by fire national government? In short, is there
any evidence firat land reform in fire Northern Province has resulted in more
secure livelihoods? Second, are the social driving forces of land use change
clearly evident in the satellite imagery? Third, did land reform in Soufir Africa
result in fire intended land use change? The goal of firis chapter is to specifically
address the research questions posed on page 67. In fire next chapter, I weave
together fire findings presented here and the literature to construct a narrative on
land and livelihoods.

I will begin by examining the statistical evidence in order to judge fire
impact of land reform on livelihoods and land use change. I will firen address
fire issue of efficacy in satellite-assisted interpretation of land cover change and
determine fire degree to which relationships can be gleaned from the imagery in

dryland areas. Finally, I will challenge the general assumptions laid down by

141

Soufir African policy makers regarding fire intended effects of land reform versus
reality.

Data analysis provided findings on firree general fronts: 1) statistical
findings derived from the interview data; 2) spatial findings derived from fire
satellite and GIS data; 3) and findings derived from qualitative interview. The
challenge of firis chapter is to weave the findings of these firree analyses togefirer
in order to provide a meaningful story. Rather than provide a ”cut-and-dry”
prove/ disprove or support/ reject discussion, I prefer here to tell fire story of
land use change, land reform, and livelihood systems, using fire evidence
presented in fire last chapter to support the main themes.

What, then, are fire main findings? The data analysis of livelihood systems
showed:

0 A relatively representative sample population (income, age, gender,

marital status)

0 High levels of labor migration (34%), but not as high as in ofirer areas of

Soufir Africa.

0 Most respondents farm, but derive litfie income from it, even though

male CPA members sell more crops firan before joining fire association.

0 Just under half of CPA members have experienced food shortages, but

report firis happening less since firey joined the association. However, firey

are more dependent on shops, indicating that shop-bought food has

become a more important coping mechanism.

142

OOnly 20% of inhabitants of fire control farm reported a food insecure
situation since 1980, and in firis respect, are better off firan their
counterparts on the CPAs where a similar percentage of respondents in a
much shorter period (since 1996).

o The level of dependence on shop-bought food as a coping strategy is
exceptionally high (54%) and households in firis study are considered to
be structurally dependent on shops. CPA members have a broader suite of
coping mechanisms firan do fireir counterparts on fire control farm.

0 Ofirer coping mechanisms are similar to those found in fire literature, but
much fewer in number and more limited in scope to the point that serious
concerns are raised over the ability of households to cope in extreme
situations.

0 CPA members have a broader income strategy firan inhabitants of the
control farm. The majority perceives no change in their income as
compared to fire period before joining the CPA.

o A remarkable dependence on entifiements was found on fire control farm
while CPA members are actively engaged in petty commodity trade.

0 Far fewer people on fire control farm raise catfie firan do CPA members,
largely due to differential access to suitable grazing land.

0 Land use authority on fire control farm is typical of rural South Africa

wifir half of fire population subscribing chiefly authority while fire ofirer

143

half adhere to ofirer sources of aufirority including firemselves and the
local state agents.

0 Land use change was minimal on fire CPA farms firemselves, but change
was occurring in fire areas where association members reside just outside
fire officially transferred farmland.

0 Land use change was negligible on fire control farm, representing a
stable farming system (corroborated by fewer incidents of food shortage
among residents).

0 General disorganization prevails on fire majority of CPAs.

o CPAs are generally under or improperly capitalized, suffer from a lack of
skilled farmers, do not obtain adequate extension, and membership
participation is poor, at best.

0 Women are far more likely to be poor firan are men.

0 Women are more willing to plant crops ofirer firan maize than are men.
Accordingly, men were more likely than women to report firat farming
gave firem most or some of their income. Given that maize is also a
primary cash-earning crop, firis is not surprising.

o Joining fire CPA increased men’s ability to sell crops, while fire opposite
was true for women.

0 Women were nearly twice as reliant on entifiements firan men. Given fire
labor migration pattern firis is expected.

0 Men have far more access to credit.

144

Significant relationships were presented in fire last chapter. What do firese
relationships mean? Given firat the statistic used does not indicate directionality,
I will interpret the relafionships based on qualitafive data and descriptive
statistics. Table Six on page 90 summarizes fire main relationships.

First, gender influenced several variables. Age and gender were related
largely because fire population is generally older, but older women were even
more pronounced firan one would expect in a random sample of fire total rural
population. The principal of self-selection of CPA members and the clear state
directives at gender equity explain firis relationship. Older women stated firat
firey “had more time" to participate in such activity. Age and income are related
and a review of the descriptive statistics indicated that the direction of fire
relationship is expected. Older participants have fewer income options and tend
to be poorer.

Food security since joining fire CPA [CHGCPA] was shown to be related to
gender. In fire summary of findings I noted that men were more able to sell fireir
crops firan women do since joining fire CPA but had a narrower range of crops
that firey farmed. Male reliance on a single crop, maize, has restricted fire
options firey may pursue to strengthen their food security. Thus it is not
surprising to see fire relationship. In firis instance I suggest firat fire relationship
reflects fire fact firat men are more likely firan women to rely on shops. Women,
by widening their crop options, are less likely to have to sell maize to fire shops

for milling, for instance. In times of maize failure, women have ofirer crops on

145

which firey can rely. This is corroborated by fire statistical relationship between
gender and change in fire amount of crops sold since joining fire CPA
[CRPSELL]. Here fire explanation lies in the fact firat men sell fireir crops more
often firan women do since joining fire CPA. Furfirer corroboration comes from
fire fact firat a statistical relationship exists between instances of food shortage
[EXPSOME] and gender. Women reported experiencing more food shortages
firan men did. However, firis is understandable because men's ability to procure
food at fire shops is related to fireir differential access to credit. In short, due to
fire structural dependence of rural households on shop-bought food, women’s
broader cropping options do not ensure their food security. Food security is
ensured not by fire range of crops farmed rafirer it is access to credit at shops firat
provides security.

Statistical relationships of variables to age showed firat older people tended
to be poorer on fire control farm. Most elderly people rely on pensions as fireir
main source of income on fire control farm and tend to rely on hired labor for
assistance in farming. Neither of firese relationships are surprising or lend much
explanatory power to fire analysis of farming systems but rafirer confirm firat fire
control farm is similar in these respects to other areas in rural Soufir Africa.

Income had fire fewest statistical relationships in fire data set. Relationships
were found only between income and age, income and fire hiring of labor
[LABOR2], and income and time spent on various income generating activities

[MOST]. As noted above older people tend to be poorer, and firose wifir more

146

income at fireir disposal are more likely to hire laborers. Income is affected by
activities an individual pursues in order to garner an income. In firis instance, it
is hypofiresized firat poorer individuals will rely on entifiements while wealfirier

ones may pursue agriculture or be involved in paid labor.
5.2 Evaluation of the Impacts of CPAs on Member's Livelihoods

One of fire key research questions centers on fire effect of a land reform
event on livelihoods and land use change. Did fire introduction of the Communal
Property Association land reform alter, stabilize, strengthen, weaken, or have no
effect on livelihood systems and changes in land use?

Some differences were found between CPAs and the control farm (see Table
8, page 104). To simplify firis discussion I refer to ”place” meaning whefirer a
respondent was a member of fire CPA or fire Control Farm. Age was shown to
have a relationship to place. This is largely due to fire fact firat people were more
clustered into fewer of fire age cohorts on fire control farm. Coping mechanisms
are also related to place. On fire control farm, coping mechanisms are focused on
firose firat in some way relate to shop-bought food, such as saving money to buy
food in shops, or purchasing food on credit from shops. CPA members generally
have a broader suite of responses to food shortages, such as relying on neighbors
and seeking income from petty commodity trade. Members of fire CPAs have

experienced more shortages since joining fire CPA, firus would have developed

147

more coping mechanisms than people on fire control farm who have experienced
fewer shortages.

Farming is more prevalent on fire control farm than at CPAs ([PLACE] vs.
[FARM]). This is not surprising given findings mentioned above regarding fire
broader suite of income strategies pursued by members of CPAs and fire clear
lack of land use change. Interestingly, fire level of income is related to place.
However, no clear indication of directionality is indicated in the data. It would
appear firat CPAs have a more even distribution of incomes while on the control
farm incomes clustered around the levels <R100 a month and R400-600 a monfir.
A significant number of unemployed persons were found on fire CPA. This
’downward’ skew of fire income on CPAs could also be leading to fire
relationship. The data does not clearly indicate firat one group of people is more
or less wealfiry firan the other, but rafirer firat income distribution among
participants is more even in fire CPAs firan on the control farm. There are far
more poor people on CPAs. This last statement should not be firought of as
causal. Membership in a CPA does not cause poverty, rafirer it is more likely firat
it simply reflects fire constituency base. CPAs have self-selecting membership
based on fire once-off R15000 social grant. More poor people are likely to band
togefirer to purchase a farm firan are wealfiry people. Wealfiry people in most
cases have ofirer, non-farm, sources of income firat firey view as more profitable,
firerefore will not join CPAs in large numbers. Lack of return to fire investment of

time and money can also dampen income levels.

148

The data suggests a small degree of improvement in livelihoods after
joining fire CPA, but fire benefit accrued to men disproportionately. This is quite
striking as CPAs were purposely designed wifir gender equity in mind. No
significant positive change was detected in income or fire mix of income
generating activities. Men showed less likelihood to experience food shortages
firan prior to joining fire association. Income strategies are more diverse among
CPA members than on fire control farm, but participation in fire CPA is not
clearly fire cause. It must be remembered here that the CPAs are nearer the larger
cities of fire province that may provide members with ofirer income generating
activities. In fact among CPA participants one of fire key income activities was
petty commodity trade, which is much more possible near large population firan
in "deep” rural areas such as the control farm.

As indicated in other studies, fire range of coping mechanisms to food
shortages is limited. The dependence on shop-bought food is so high as to
strongly suggest that rural Africans are structurally dependent on shops for fireir
food. This structural dependence can be direcfiy attributed to fire aparfireid
state’s social engineering firat sought to create cheap rural labor pools. Furfirer,
rural blacks’ lifeline could be immediately disrupted in fire event of rural mass
action. A rural African farming class could threaten both the state’s interests in
maintaining order and capital’s interests in maintaining a labor pool dependent
on wage labor. The two are not unrelated. Capital and fire aparfireid state

worked hand-in-hand to produce just such conditions. In fact, without a

149

structural dependence on shops, rural Africans would not need to engage in fire
labor market to fire same degree and wages would necessarily rise. Ensuring
African dependence on shop-bought food guaranteed their participation in fire
cash-labor economy.

The evidence regarding the impact of Communal Property Associations on
the livelihoods of its members is firat the programs tend to broaden coping
mechanisms to food shortages and have begun to lead to broader income
strategies. Based on fire literature, it is also highly likely firat the central
government has ended the program (and support to it) too early. I have found no
clear and decisive evidence to conclude firat fire CPA program has yet been a
successful transforming mechanism in rural Soufir Africa, however, long-term
trends have had insufficient time to materialize. The CPAs included in firis study
have not yet contributed to members’ livelihoods in a manner significant to
render firem firus far successful. Members expend time, money, and energy and
receive very litfie in return. Agricultural transformation leading to rural poverty
alleviation envisioned by fire planners of post-apartheid South Africa have
simply not produced results. In frustration the central government has

abandoned rafirer firan reorganized fire CPA program.

150

5.3 Analysis of Satellite Imagery

Research question two will be addressed in firis section: 1) How well did
change show up in fire image versus what was told by the interviewed, and 2)
Are firere patterns firat emerged in the imagery across CPAs that could be related

to social process?

5.3.1 Verifying the Interviezvs

On fire first question, fire answers provided by interviewees and fire
findings derived from satellite imagery analysis were fire same. On all of fire
CPAs, members were clear that litfie change had happened on fire redistributed
lands. An apparent conversion from farmland to grazing land on fire control
farm indicated on the survey instrument was identified as erroneous by fire
imagery and was clarified with follow-up questions. What is important here is
first fire analysis of satellite imagery helped to illustrate certain phenomena and
evoke questions on others. The best example of firis was the farmland to grazing
land discrepancy on fire control farm. The interview data would never have been
clarified had it not been for fire imagery telling a different story. The large degree
of change indicated by the participants was not corroborated by fire imagery. The
question asked was ”have you ever grazed catfie on land firat you used to farm”.

The intention of firis question was to query for land use conversion. The meaning

151

of fire question translated but fire lack of a time qualifier led respondents to
answer ”yes”. When crops were harvested catfie were often allowed to graze fire
stubble. The expected drastic change in land use did not appear on fire imagery
leading to a re-evaluation of fire question and follow-up group discussions.

The imagery was also useful in identifying changes firat respondents were
unable to recall or of which firey had no knowledge. By showing fire imagery to
key informants, persons wifir fire necessary knowledge could be contacted. The
imagery helped people understand fire spatial extent of their land and relate
their farms to surrounding areas. The imagery also helped verify conflicting
claims to land use changes. When respondents disagreed over fire scope or
spatiality of change, images could be shown to firem and an agreement over

change could be established.

5.3.2 ”People and Pixels” or ”People or Pixels”?

Discerning social process from satellite images of vegetative cover is
fraught wifir difficulties. Alienating a specific landscape change and relating it to
social change in reliable and reproducible ways presents a formidable challenge.
Wifir these cautions, however, some general statements can be made regarding
land use change and social process in firis study. Attempting to model change in
firis study, given the context, would be erroneous, as fire events leading to (a lack
of) change on CPAs is known - an unsuccessful land reform event occurred.

Modeling is necessary only in instances where information is incomplete and

152

general observations must be drawn from a small sample area and projected
onto a larger spatial scale. This is not necessary in fire Northern Province, as we
know how and why change has failed to occur on CPAs. Where modeling might
be useful is in fire identification of possible land use changes on non-state
sponsored land transfers, such as land invasions or private market transactions.
Land use change in these instances are not well documented and in fire case of
land invasions, rarely documented at all. Furfirer, it has been shown here firat fire
real impact of land reform on land use change is in areas surrounding CPAs
where fire membership resides.

Identifying fire social driving forces of land use change, particularly
changes firat occur when land is transferred from white to black ownership,
requires a temporal analysis that includes a pre-1994 scene and fire most current
scene available and basic information on fire population structure and political-
geographic setting.

Once fire basic irrformation has been gafirered some hypofireses can be
made as to where change would most likely be occurring. These findings apply
not only to transferred lands, but also to land use change in general.

1. Border areas of fire former homelands were observed to be more likely to
experience change. All of fire CPAs in firis study were formerly white
areas near former homeland borders. Monyamane and Mahlambandlovu
were adjacent to fire actual border between the Transvaal and lebowa

homeland.

153

2. Change from white to African land derived from analysis of fire CPA
lands can be evidenced by extensification. Structural impediments have
prevented intensive use of transferred lands.

3. Areas wifir pronounced expansion of rural townships tend to produce a
land use change nearby. The driving force of change in firis case is fire
expansion of fire township or in fire case of CPAs fire transfer of
ownership precipitated fire setfiement change. Furfirer investigation will

need to be conducted on the ground to determine causality.

Social driving forces derived from fire survey instrument were discussed in
fire previous chapter, but how well can social process be determined from an
image? A one-to-one correspondence was impossible to achieve, but I observed
several fiends regarding social process and images. First, white farmers and
African farmers leave very different spatial footprints on the landscape. White
farmers almost always plant their fields in large rectilinear blocks, eifirer
rectangles or squares. Circles are a relic of center pivot irrigation systems. This
pattern is similar to farming patterns in both Europe and North America as white
Soufir African farmers imported bofir technology and farming practices from fire
norfir. African farmers, however, rarely plant their field in large or rectilinear
blocks and have had very litfie access to center-pivot irrigation systems due to
fireir prohibitive cost. African cropland is fragmented into very small parcels,

usually following contours, and often near rural setfiements.

154

Given firese patterns, one can infer several social driving forces based on
observed change. At firis stage I am unable to provide empirical justifications for
firese inferences beyond what was observed in fire areas immediately outside fire
CPAs. I was unable to pursue specific tests on firese social driving forces and can
only relate fire general trends I observed while in close contact wifir these areas. I
observed fire following: 1) economic pressure is leading many white farmers to
sell or subdivide fireir farms; 2) a change from clear rectilinear and crop circles to
an African-style pattern indicates social change, possibly land transfer; and 3)
areas where African farmland has been abandoned or extensified results from
out-migration leading to a lack of labor to farm, as shown in fire Rondebosch

CPA example.

5.4 Summary

I have presented fire findings of firis study in firis chapter related to the
specific research questions. I have shown that for the first research question -
how have livelihoods responded to land reform - the answer to be ”less firan
expected” (by bofir my own hypofireses and by policy-makers in Pretoria). I
demonstrated firat alfirough less firan expected, livelihood changes were
differentiated by gender and income strategies were broader on the CPA than on
fire control farm. The range of coping mechanisms to food shortages appeared to

be somewhat broadened by participation in fire CPA, however, members of fire

155

sh

CPA remained structurally dependent on store-bought food (and subsequenfiy
shop credit).

The second research question focused on fire social driving forces of land
use change derived from analysis of satellite remote sensing. I specifically sought
to utilize fire imagery to produce direct statements on social driving forces, but
was unable to make a one-to-one correspondence between a land cover change,
fire underlying land use change, and fire social force driving fire change. I was
able to make broader statements on forces driving land use change, but could not
produce an empirically based validation of such statements.

In fire next chapter I address fire firird research question. I construct a
narrative describing the intersection of fire various forces firat have contributed
to fire unfinished nature of transformation on the Communal Property

Associations.

156

Chapter Six

Narrative: The Unfinished Transformation- Impediments to Livelihood

Enhancement and Land Use Change

6.1 Introduction

The story of how livelihoods, land use change, and land reform intersected
on fire Communal Property Associations will be told in firis chapter. The goal
here is to integrate the themes raised in the review of literature and fire findings
presented in the last chapter to understand how such an intersection created fire
current socio-economic circumstances on fire CPAs. This chapter, firen, will both
contextualize fire findings and situate fire research in the broader literature,
drawing togefirer fire multiple explanations for the unfinished transformation
into one narrative.

The CPA program was designed to empower rural historically
disadvantaged Africans to run productive farms that would strengthen
members’ livelihoods by changing land uses. The transfer of land was fire first
step in fire empowerment process. However, it was not meant to be fire last. The
analysis of land cover and use change showed firat fire actual farmland
transferred has undergone few changes to a more productive state. In fact,

simultaneous intensification and extensification is occurring in fire study areas,

157

intensification off fire transferred land and extensification on fire transferred
land. At Mahlambandlovu, Monyamane, Muyingiseri, and Rondebosch CPAs no
actual crop farming has occurred. Baroka CPA, fire smallest, is fire exception.
Land is being used as productively as under fire former white owner. In areas
outside fire CPAs, where fire bulk of fire membership lives, land use and cover
changes are widespread. Land uses are increasingly being converted from
agricultural to residential uses and areas firat were formerly farmland are
becoming grasslands. I will first present fire evidence that leads me to conclude
firat fire CPAs have failed to make significant progress. I will firen present the six
primary reasons for the lack of progress wifir empirical support drawn from fire
livelihoods and resetfiement literature. This section is devoted to analysis at fire
household / farm scale. Next, I will show that the evidence obtained from the
CPAs represents short-term trends rafirer firan transitions or long-term trends.
Explanations for fire lack of progress will be placed in a regional context followed
by a general assessment of fire successes of and challenges to land reform in
Soufir Africa.

In order to justify a judgement on fire progress of CPAs, I address here fire
question - are fire CPAs at the point of development or maturation we would
expect firem to be compared to case studies from the literature? Alfirough in
existence for between four and six years, the CPAs in firis study appear to have
stalled in the initial phase. When comparing fire available evidence from the

literature from fire evidence garnered from the CPAs, one would have expected

158

much more change to take place. This expectation is supported in fire literature
from fire following evidence.

First, fire livelihoods of CPA members have not enjoyed significant
enhancement. The most relevant literature reviews resettlement. In firese instances
a dramatic change occurs, participants are both physically moved to a new and
often unfamiliar area and livelihoods are completely uprooted. In most cases,
livelihoods recover and transition ends from two to seven years after the initial
change. However, firis study examines change where no resettlement has occurred.
Projects were undertaken near the residences of fire beneficiaries. Participants did
not have to abandon their former livelihoods. The cooperatives literature, however,
describe projects firat were much more similar to fire CPAs but had very litfie
actual assessment of fire impact of cooperatives on livelihood systems. While
lessons learned from cooperatives could be applied to fire structure of the CPAs,
very litfie could be said about livelihoods.

Thus fire CPA could be firought of as a livelihood-enhancing strategy, or
alternately as an additional choice from fire already broad array of options which
participants could select to purse a livelihood (e.g. petty commodity trade,
migration, formal or informal employment). Thus, fire options firat provided fire
least enhancement to fireir livelihoods would eventually be excluded, in effect, a
livelihood maximizing exercise. The early exit of many CPA participants showed
that firey felt fire advantage of participating in fire CPA to fireir livelihood system

was significantly less than their other livelihood options. The process of livelihood

159

transformation has stalled. Memba‘s’ livelihoods are not being transformed in
ways firat ensure active membership participation and reproduction of fire CPA.

Second, no empirical evidence of intensifying land use on fire redistributed
land comprising any of the CPAs is evident. If land use change were occurring
firat indicated a shift to more productive use of fire transferred lands that would
equate to income generation for CPA members, I could argue firat success might
be forfircoming. However, in fire absence of such evidence it is difficult, at least,
to argue firat fire CPAs have affected changes in land use in a way that
contributes to current livelihoods or firat in some way may provide future
benefits. The empirical evidence, derived bofir from interview schedules and fire
satellite analysis, simply does not support such a claim.

Third, if fire South African government were undertaking observable and
empirically identifiable rural economic reform firat would create a supportive
environment for such projects, where members would realize fire ability to
secure labor, capital, and a market for fireir goods, I would be justified in arguing
firat fire trend on CPAs is not of stagnation but of transformation. Again the
empirical evidence does not lend credence to such a statement. Transformation
in rural South Africa is has been slow and uneven and the CPAs in the N orfirern

Province are no exception.

160

6.2 Impediments at the Household and Farm Scale

What has caused fire under-utilization of CPA land and fire failure to
deliver livelihood improvement? Across fire CPAs members point to several
common issues firat cause firis lack of activity. The findings of firis study are
remarkably similar to several studies firat focused on fire impacts of resetfiement
and fire results of cooperative agricultural projects. As detailed in fire Review of
fire Literature, fire themes from firose studies are replicated here and include:
difficulty managing land, rapid social change, lack of a reliable relationship wifir
government officials, overlapping claims to local authority, and lack of labor due
to out-migration (Oyedipe, 1983; Diaw, 1990; Pankhurst, 1992; Voh, 1980; Koenig,
1987; Cernea and Guggenheim, 1993; and Colson, 1971). These issues cut across
scales and are discussed below in a household / farm project context. It is
important to note here firat fire agency charged with land reform has, itself,
identified many of the same hindrances I detail below, including ” [problems
wifir] inputs and services, financial issues, and infrastructural issues” (DLA,
1999,21). Zimmerman (2000), although without empirical observations, argued
many of fire same points when detailing the barriers to participation of fire poor.
He noted the barriers as a high level of risk, high up-front costs, lack of skills and
education, fire lack of rural services such as water, and fire difficulties of moving

over long distances (Zimmerman, 2000, p.1455).

161

Reason 1: General Disorganization (Social Disruption)

First and foremost, members feel unprepared to organize, maintain, and
develop fireir CPA as an economic unit. The farm as unit of production has failed
to materialize largely due to a lack of managerial skills. The membership and
leadership of CPAs were not chosen specifically for fireir expertise in farming,
and many simply lack fire necessary skills to run a farm as a business. Training
on all farms was reported to be minimal. In the case of the Baroka CPA fire
membership has benefited from the fact firat fire chairperson was trained at an
agricultural technikon in the Eastern Cape. The ofirer CPAs have not been as
fortunate.

Members of fire all of fire CPAs recognized at least some degree of post-
transfer government support, however, the support programs did not provide
fire skills necessary to maintain long-term production. The members of
individual CPAs must also bear responsibility for fire general disorganization.
Membership participation is weak at best. For instance, two work-days were
scheduled at fire Mahlambandlovu CPA firat I attended for purposes of
observation. Of fire 396 members barely eighty showed up on the first occasion
and only fifty attended on fire second. The secretary allowed me to peruse her
records and she noted firat members were also required to make cash

contributions. Her records indicated large numbers of people who had neifirer

162

attended nor contributed. She explained firat upon inception of the CPA many
people believed that fire government was ”coming to hand out cash”. Expecting
some sort of direct transfer, many people signed up for fire CPA ”and had no
idea what firey were really doing’ "V. After the first two years, firey saw no return
to their participation and abandoned the CPA.

Members of fire Monyamane CPA’s agricultural production committee
reported widespread failure to participate in even basic CPA functions. The low
level of participation was discussed at several informal meetings. When queried
about lack of participation members responded with a range of reasons from ”we
are very old and cannot work we ” to ”no-one shows up and firen we must do
all fire work” to issues regarding lack of transportation, failure for fireir concerns
to be addressed, and general apafiry. Most telling is firat on one CPA a number of
members firat were so completely disgusted firat they requested a secretary
accompanying me on a visit to ”strike their names” from fire membership list.
Many issues regarding the issues of poor attendance and poor organization are
in large part due to fire age structure of members and members perception firat
fire CPA has become ”a waste of time”.

The literature on cooperatives supports these findings regarding fire lack of
membership participation as a key obstacle. Kimario (1992) noted firat a lack of
support from members was a fundamental problem in Tanzanian cooperatives.
Adeyeye (1991) found similar problems in Nigeria as did Bager (1980) on Kenyan

cooperatives.

163

Diaw (1990), Oyedipe (1983), Voh (1980), and Colson (1971) all found a high
degree of social disruption and general disorganization after resetfiement. Long-
standing social roles change dramatically and power is not always concentrated
in fire same hands after fire transition. In fact, Pankhurst found firat ”former
wealfir was not always a means to success” (Pankhurst, 1992, 156). In fire change
to new production arrangements, such as fire introduction of a CPA, new rural
elites emerge and old ones are swept away. The unpredictable nature of such
transitions leave people vulnerable and often unwilling to contribute to a project
firat may or may not benefit firem. As Cemea and Guggenheim (1993) found new
social factions form and ofirers fade away (241). These factions do not always

cooperate wifir each ofirer, leading to schisms and disorganization.

Reason Two: Too Many People (Farm Size)

Farm size is a focal point of much of fire agricultural economics and
alternative development literature. Traditionally, agricultural economists argued
that larger farms create a more favorable economy of scale, are more efficient and
profitable, and stimulate rural development. This accepted knowledge shifted as
noted by Staatz and Eicher:

Empirical evidence from the late 1960s and early 1970s revealed firat fire

economies of size in tropical agriculture were more limited firan had been

previously believed and that small farms often achieved greater output and
employment per hectare firan did large-scale farming (19).

164

Binswanger and Elgin recount fire main reason for firis advantage- fire use
of less hired labor on small farms. Small farms use less hired and more familial
labor, fire latter being more efficient and cheaper. Furfirer, family members share
the risks (318). Many of fire perceived advantages and economies of scale of
large farms were shown eifirer not to be necessarily beneficial or largely
inconsequential on small farms. Regarding mechanization, larger farms maintain
an advantage in first ”average costs decrease as fire size of holding increase”
(319), however ”small owners can rent out fireir land rafirer firan sell it -and still
keep fire advantage of owning land to raise credit” (319). Binswanger and Elgin
maintain firat ownership can be independent from operation - small farmers can
join together to match the economy of scale of large farms, firus, reducing
mechanization inefficiencies.

Large farms have traditionally been considered advantageous in terms of
technology and management as well. The authors argue here firat:

Some management skills can be rented. If technology becomes too complex,

farmers can hire private extension officers by fire hour or the day to advise

firem...another solution to fire management problem is contract farming,
where large firms provide technical advice, finance, and marketing service

to small farmers (321).

Van Zyl contextualizes fire debate for Soufir Africa emphasizing two points:

1) fire productivity differential favoring small farms over large farms

increases wifir fire differences in size, implying firat it is largest where

inequalities are fire greatest, and

165

2) fire highest output per unit areas is often not achieved by the smallest
farm size category, but by fire second smallest...suggesting firat fire
smallest farms may be fire most severely credit constrained (266).

The author reinforces firese statements with empirical evidence showing
small-scale black agriculture out-performing large agriculture in Soufir Africa.
The evidence points to the fact that ”small farmers may use land much more
intensively firan do large farmers” (274).

Regarding firis study, fire question arises - Are fire large CPAs too large?
The findings of firis study show firat very litfie intensive use of fire land could be
found on any but fire smallest CPA. The above literature can inform firis study as
large white-owned farms were transferred to large groups of blacks. Farm size
did not decrease, and in the case of Mahlambandlovu and Monyamane CPAs it
actually increased. Van Zyl (1996) showed fire inefficiency of large white-owned
farms in Soufir Africa. Transferring these farms to large groups of blacks does not
abrogate fire farm size problem.

What, firen, are the specific problems facing the large CPAs? First, any
decision firat must be undertaken on the CPA must consider a large number of
members wifir widely differing opinions. For instance a land use change firat
might result in increased crop production would require fire consultation of a
large group of people, some of whom want to plant maize while ofirer want to
raise more catfie. How can a coordinating committee take into account fire land

needs of 396 different members on a 2159 ha farm? The solution under fire CPA

166

programs is fire central committee would decide on land use approved by fire
members. However, members who, for instance, do not have catfie have no
incentive to participate. If a group of older women wish to plant tree crops but
are outvoted by men who want to plant maize, fire women will (and have)
simply stop taking an active role.

Second, simple tasks, such a communicating a meeting, become
administratively difficult and complex when membership numbers rise into fire
hundreds. The secretary of one of fire CPAs illustrated several communication
problems fire committee faced due to the large membership number.
Announcing and garnering participation at crucial meetings is almost
impossible. The telephone is almost completely absent from the homes of
members. To communicate a forfircoming meeting the secretary must eifirer
travel to all fire population centers or must rely on word-of-moufir.

Third, fire constitution of fire CPA requires distribution of the profits among
all members. However, profits are not always realized. When catfie, for instance,
are sold, most of the income goes to inputs and running costs leaving very litfie
profit. Difficulties arise over how the small profit will be distributed. Many
members argued firat profits should accrue only to firose active participants.
According fire CPA constitutions it cannot. That in itself is a lefiral disincentive to
any activity at all. In practice, however, those that contribute are firose firat reap

the rewards.

167

Reason Three: Lack of Labor and Skills

Out-migration is widespread in all of Soufir Africa’s former homeland
areas. May (1996) refers to statistics which show fire largest source of income in
rural areas of KwaZulu-Natal derived from wage labor or remittances from wage
labor (in Lipton, et al., p.21, 1996). While acknowledging firat other sources of
income are important, such as agriculture and petty commodification, clear
emphasis is placed on fire role of labor migration. May notes firat ”migration for
employment remains an important aspect of many rural people’s lives, as does
fire reliance of rural households on a share of fire migrant’ 5 income in fire form of
a remittance ” (May, 1996,p.14).

In fire Northern Province, Baber (1996) found in a sample of two villages
firat nearly 70% of households ”include at least one migrant” (280). He maintains
firat firese migrants represent a valuable pool of largely inaccessible labor and
skills- inaccessible, that is, in fire rural areas. Kirsten (1996) and Ngqaleni and
Makhura (1996) present similar pictures of labor and skills loss due to migration.
Low (1986) interprets the impacts of migration on the household:

We have seen that fire opportunity for some members to earn relatively

high incomes in off-farm employment can result in labour shortages at fire

farm-household level, which lead to low levels of labour input per hectare

(126).

Mini (1994) corroborates the statements of bofir fire above aufirors and fire

findings of firis study writing ”fire shortage of labour at critical periods of

168

peasant farming activities was attributed to the migrant labour system, which
deprives firese villages of young adult males” (276).

As evidenced in the statistical analyses in firis study, many people rely on
hired labor to perform even fire most basic functions of agrarian production.
Labor exchange is not uncommon in ofirer parts of Africa during certain points in
the crop cycle. For instance, at planting and harvesting time in sub-Saharan
Africa, labor becomes scarce and exchanges of labor become commonplace. The
normal shortage periods in rural Soufir Africa are highlighted by fire fact firat at
bofir the CPA and control farm nearly one-firird of households have one or more
members involved in fire labor market in Gauteng. Many members of fire
Monyamane and Mahlambandlovu CPA households are involved in fire
Pietersburg labor market as well. This drain of labor impacts fire younger and
middle age groups disproportionately. Those people wifir fire most energy to
engage in agrarian production are absent.

A common myth throughout Soufir Africa is fire belief firat ”all Africans can
farm” or ”fire blacks can farm” (as evidenced in fire qualitative and quantitative
interviews undertaken for firis study). This racial stereotype ignores fire long-
term de-skilling of blacks firat occurred under apartheid. Farm laborers never
received explanation behind fire instructions they were given on large white
farms. They gained very few skills on white farms least of which was how to
manage a farm as an economic unit. Extension into fire former homelands was

erratic and poorly coordinated. Wifir large numbers of people, mainly men. off in

169

Johannesburg, extension was targeted at a very small portion of fire remaining
male population. This is not surprising as homeland resources for such activity
was severely limited. It was in fire best interests of the apartheid state to keep
Africans out of agriculture. Africans able to reproduce their livelihoods based on
productive and profitable agrarian systems would necessarily undermine the
labor market, a labor market that constanfiy petitioned fire aparfireid state to

keep wages very low.

Reason Four: Continuing Gender Bias

Gender bias in rural South Africa is manifest as unequal access to land,
capital, and social services. The Reconstruction and Development Programme
(RSA, 1994) specifically addresses fire plight of women:

Women face specific disabilities in obtaining land. The land redistribution

programme must firerefore target women. Institutions, practices, and laws

firat discriminate against women’s access to land must be reviewed and
brought into line wifir national policy.

But is gender equity necessary for a CPA to be successful? Gender equity is
often advocated on liberationist grounds, i.e. liberating women from the yoke of
oppression. Alfirough firis is clearly a valid reason to argue for gender equity,
might firere be ofirer grounds to argue firat gender bias in rural Soufir Africa
should be eliminated? Ngqualeni and Makhura (1996) show fire high degree of

female involvement in agriculture (70% of agricultural labor is female) in the

Norfirern Province. Mini (1994) found a similarly high proportion in fire Eastern

170

Cape (75%) and Walker notes firat ”land represents an opportunity for women to
meet a variety of needs, including those which are socially identified as primarily
fireir responsibility” (71). Women’s limited access to land in fire form of gardens
can be a mixed blessing, however. Middleton argues:
While fire cultivation of food crops, particularly vegetables, may feed
families and ease the demands on family budgets, firis would not be
without its contradictions. Firsfiy, food cultivation food cultivation for
subsistence purposes was fire responsibility of women. Thus fire cultivation
of gardens and field would have fire effect of increasing fire already heavy
workload of women. Secondly, food cultivation is physically exacting, time-

consuming work...Thirdly the absence of water supplies, farming
equipment, and . ..ofirer resources could result in food cultivation becoming

an additional burden (81).

Reform of policies promoting, direcfiy or indirecfiy, gender bias and
promotion of those firat attempt redress would allow women larger participation
in agricultural projects. Many women are now tied to subsistence production
(vis-a-vis fire ”garden”) and are unable to meaningfully contribute to and reap
from fire operation of CPAs. Further, fireir time allocation to subsistence
production effectively traps firem into poverty. They have very litfie time to
contribute to CPAs (an alternative and additional income generating strategy)
and are socially marginalized (through limited access to credit) when firey do
participate. Men benefit from CPAs precisely because firey have been freed from
subsistence production. Women have not been able to consolidate enough power
wifirin fire CPAs (or more accurately fireir own households) to change firis

situation. Thus, a highly skilled and agriculturally knowledgeable portion of fire

171

 

rural population is effectively blocked from or has limited involvement in
projects firat may stimulate feedbacks in fire rural economy. Women’s full
participation in CPAs would unlock a vast reservoir of knowledge and skills,
somefiring sorely lacking in rural areas. Women’s limited access to credit, land,
and labor is a hindrance to rural development, as Mini points out

fire likelihood of fire female-headed households improving farm output and

farm income is limited, first, by fire labour deficiency which tends to limit

fire size of cultivated areas and secondly by their lack of capital in fire form

of stock and cash (277).

Gender equity has been a key goal of the land reform program since its
inception in 1994 and 1995. This study has shown that fewer benefits have
accrued to women. Men still have better access to credit and benefit from fireir
participation in the CPA while women actually suffer. Women were found to be
poorer in several respects. In firis study firree factors were found to contribute to
female poverty:

1. A high degree of dependence on insecure entifiements.

2. Unequal division of labor and land use; a division firat limits women’s

flexibility and income options.

3. Unequal access to credit.

Discussions wifir women were conducted after initial data analysis showed
firat large numbers of women felt fireir incomes did not improve, and statistical

tests indicated firat firese feelings were not unfounded. Women generally felt

alienated from fire core decisions made at the CPA. In two cases, fire dominant

172

activity on fire CPA was catfie ranching. Women felt that once fire decision had
been made to raise catfie, fire men gained an ”upper-hand” in fire administration
of fire CPA. Anyfiring firat was discussed regarding fire CPA was always done in
the context of fire catfie. Many women became angered over fire perception firat
men had so much control over the CPA. They complained firat firey were not
allowed to plant tree crops firat would have provided firem wifir an income
(women usually sell fruit along side fire road in and around Pietersburg and
Tzaneen).

Finding continuing gender bias was not unexpected, but it was surprising
to find just how disgrunfied some women were regarding fireir involvement in
fire CPA. According to fire provincial administration records female-headed
households constituted the following percentages of households on CPAs:

Mahlambandlovu = 41%

Monyamane = 22%

Muyingiseri *3 77%

Baroka = 48 %

Rondebosch = 25%

This would appear to give women considerable more input into fire
running of fire organization. It must be noted here firat not all women were
displeased wifir their CPA. The most widespread unhappiness was evidenced at
Mahlambandlovu CPA. The female members of Baroka CPA were fire most
pleased wifir fire overall situation. It would be erroneous here to suggest firat all

women or all poor women were unhappy across all CPAs. Women do not share a

unified oufiook on their situation or their farms. In firis respect, firis study

173

reinforces findings from many prior studies in that women participating in fire
CPA program are not a homogenous group and as such several separate gender
relationships exist. For example, poor women involved in fire CPA have different
relationships, bofir with each ofirer and wifir men, firan wealfiry women. Poor
women often provide fire labor to liberate wealfirier women to participate in fire
CPA. Wealfirier women were more able to access resources (such as capital and
ofirer women’s labor) that allowed them to pursue involvement in fire CPA. This
led wealfirier women to a very different relationship wifir men. Gyllstrom
highlighted firis problems in his review of Kenyan cooperatives, noting firat firere
was a ”bias to advantage fire privileged strata of smallholders” (Gyllstrom, 1991,
p.273).At Mahlambandlovu CPA, follow-up fieldwork in January 2001 showed
firat a group of relatively wealfirier women had persisted in convincing fire
governing committee to apply for agricultural training support. These women
were able to pursue such an activity due to the fact that firey:

1) had money to pay for transportation to and from fire farm on a weekly
basis;

2) were able to access other’ 5 labor (largely familial, but in several cases
hired) which allowed firem to devote time to fire agricultural training
sessions;

3) were able to convince several of the men in charge and a voting block
of men firat they (fire women) were already skilled at farming and fire

men firemselves would benefit from output from fire CPA farm.

174

This division of labor empowers certain groups of men and women while
simultaneously disenfranchising others. In many cases poor men and poor women
have more in common firan do poor women and wealfirier women. Issues of power
and class temper gender relations. The central finding in firis respect is first it is
fire range of gender relations firat is key to understanding differences among and
between men and women, rafirer firan gender roles and firese relations are crucial
in understanding fire successes and failures of fire CPAs. This finding is
supported in the literature on gender and development and post-modern
assessment of gender. Moser (1993), Young (1993), Joekes, et.al. (1996), and
Marchand and Parpart (1995) all recognize the importance of examining gender

differences.

Reason Five: Improper, Incomplete, or Under-utilized Capital Investment

The problem of under-investment in rural agriculture in South Africa is
neifirer recent nor unnoticed. A large volume of literature attests to the systemic
bias against black agriculture (Lipton, 1996; Cross, et.al. 1988; deVilliers and
Critchley, 1997; van Rooyen, et.al. 1998). Golino, et.al. (1996) summarizes fire
situation:

In fire former homeland areas, input supply and credit were mosfiy non-

existent. Where these were provided, it was mosfiy firrough parastatal

institutions, generally biased against female and small farmers. In addition,

a large proportion of agricultural production in firese areas came from state

farming where input supply was arranged firrough public sector
institutions (501).

175

The authors remark firat successful reform programs will have to integrate
strong capital and skills training regimes or else fire under-skilled and under5
capitalized rural farming class will not be able to effectively utilize transferred
lands. Thobane (1997) concurs wifir firis assessment noting that under aparfireid:

Some of the services that were subsidized included irrigation, credit,

fertilizers, and marketing. The lion’s share of such privileges went to fire

rich or large-scale [white] farmers parfiy because firey were the biggest
producers and parfiy because firey were able to manipulate fire bureaucracy

and to form alliance with the politically powerful in urban areas (102).

Thobane expresses the problem of service delivery, particularly in firat
many of the African farmers are ”poorly organized and largely inarticulate”
(103). He further remarks that lack of capital and skills extension is a primary
obstacle blocking rural transformation. The sheer volume of need is scarcely met
wifirin fire current extension system, he notes, and a realignment of priorities is
needed to address fire legacy of ”separate development”, especially in fire former
homelands.

Members of all of the CPAs complained of lack of capital to undertake
farming. They complained firat start-up costs, operating costs, and equipment,
fertilizer, and marketing tools were not at their disposal. Unifome across fire
CPAs, members recognized fire need for capital in order to farm for profit. Wifir
fire exception of the Mahlambandlovu CPA, which was exceedingly well

capitalized, fire CPAs generally did not have the equipment at hand needed for

farming or raising catfie. This is not to suggest firat fire government had never

176

capitalized these farms. All fire CPAs benefited from some sort of investment,
however, in many cases the capital was eifirer not maintained or had been
destroyed in some way. A group of women at fire Monyamane CPA related fire
following story regarding fireir diesel water pump:

” fire men went away to town with fire money we had collected for petrol
[diesel]. We had taken a long time to get firat money. We went to each
person’s house and explained to them why firey must contribute. Many
people here are very poor and it was difficult to get money from firem, but
firey gave to us. When the men came back from town firey brought wifir
firem the very small containers of petrol [gesturing to fire small two litre
jugs]. They had spent most of the money on beer. They brought back more
beer firan petrol. We were very angry wifir firem but firey told us to go
away. Now no-one will contribute. When we ask firem [other CPA
members] now they tell us to go the shebeen ourselves and leave firem
alone”".

This state of affairs is not exclusive to Monyamane or to the antics of
mischievous men. The Rondebosch CPA suffered from the floods that struck
norfirern Soufir Africa from December 2000 to February 2001. Respondents told
fire story of the loss of one of fireir most precious capital investments, fireir water
pump:

”as you can see over there, our water pump has been destroyed by fire

[Ietaba] river. The government helped us put that in and now we have no

way of getting water to next years crops. The river has washed away fire

road and we can’t even get over there to fix it”

Members at Mahlambandlovu CPA told a different story of problems wifir
capital, specifically fireir farming equipment:

”here we have beautiful implements, in good condition, but we are unable
to use firem. We have a bakkie [truck] and two tractors with all fire

177

implements. This land firat firey have given us is unsuitable for farming.

We wait for the Bultfontein farm, then we will be able to use firese firings”.

Baroka CPA is the exception to firis general story of the improper or
undercapitalized CPA. The members had developed a small irrigation system for
fireir fields firat worked properly and was the pride of the farm. The farm had a
few implements, but most work was conducted manually. Baroka CPA was fire
only CPA wifir actual crop production and catfie production. The cooperatives
literature presents similar findings from case studies across Africa. Several
aufirors noted firat cooperatives lacked capital and access to credit, thus stifling
fireir development (Gyllstrom, 1991; Adeyeye, 1991; Kimario, 1992; Holmen,

1990).

Reason Six: Skewed Age Distribution

One of the most common complaints, particularly on fire Mahlambandlovu
and Rondebouh CPAs and the Platklip Farm was fire youth are scarcely around
to provide labor. Older respondents on CPAs voiced serious concerns regarding
fire future of fire CPA wifir so many youfir seemingly unconcerned wifir fire
projects. The age statistics presented in fire last chapter showed that on fire CPAs
fire median age was in the 40-50 age cohort wifir significant numbers in the 50-60
age cohort. Two official and complete membership lists were obtained firat
included each member’ 3 age. The total number of cases was 512 and fire mean

age was 46 confirnring firat the sample data is accurate.

 

.. ./—...

This represents an older population firan was expected. The effects of such
an age skew on production are far-reaching. Older members were much less able
to contribute labor in meaningful ways. Often older men would literally hire a
young person to go to ”work days” at fire Mahlambandlovu CPA so fire men’s
contribution to required labor would not go unfulfilled.

Diaw (1990) writes of a similar situation in Ghana. All of fire younger
members of fire project farm had left and the remaining participants complained
firat firey were too old. The high pr0portion of older members at CPAs raises
furfirer questions regarding future viability. If farms are struggling now, how
will firey be able to survive in ten years when the bulk of fire membership will
either be deceased or completely incapable of rendering any service to fire
association? Unless steps are taken to encourage fire participation of younger

people fire CPAs will face mounting problems and more uncertainty.

6.3 Trends or Transitions?

Given that most of the CPAs have been in existence for less firan seven
years, are fire six points listed above symptomatic of short-term trends, long-term
trends, or simply fire result of fiansition? As I demonstrated in the Review of fire
literature (see page 62), several aufirors estimated that resetfied communities
recovered wifirin one to five years. Given firat fire development of fire CPAs have

been slow, but not nearly as traumatic as a resetfiement, I am inclined to view fire

179

current situation on fire CPAs as symptomatic of short-term trends. I have no
empirical evidence to suggest that fire findings presented here are long-term
patterns. Nor do I have empirical evidence firat fire transitions are compleb.
What I have shown is firat a transition has occurred and fire initial change has
been incorporated into fire livelihood systems of participants. I have shown first
fire livelihood systems of respondents from fire CPAs are remarkably similar to
those in fire control group, and that very litfie evidence exists firat livelihoods
benefited direcfiy from the introduction of the CPA. However, to project firis into
fire future would be erroneous, locking fire CPAs into firis given time period of
analysis and denying firat dynamic changes could occur in five or ten years.
What can be said is that given the existing conditions, fire processes facilitating
livelihood enhancement in fire CPA program have stalled.

One finding, however, does have implications for long-term trends. The
participants on CPAs were found to be quite old. Very few young people
participated actively in fire programs. This calls into serious question the ability
of fire membership to build and develop fire CPA into fire future. What will
happen in ten years when the bulk of the population is simply too old to
contribute? The long-term oufiook for fire CPAs is unclear due to the aged nature
of fire membership.

The data have demonstrated that after four to six years the potential of fire

CPAs has not been realized, but does fire literature suggest this is unusual? If fire

180

CPA program had been fully supported, what could we have expected to see
(versus what we see now)?

First, what are some of fire conditions firat might ensure success? Findley
(1988) sets out fire ”structural conditions” firat led to success in Latin American
reform and resetfiement programs:

1.Appropriate crops were selected, for which demand and marketing

facilities existed.

2.The cooperative provided an effective organization to facilitate

commercialization. In particular, it provided inputs in a timely fashion.

3. Credit to setfiers was based on potential yields, which enabled setfiers to

shift from low-return food crops to high-return cash crops.

4. Setfiers were a fairly homogenous group, drawn predominanfiy from

nearby areas. This resulted in less dislocation and more continuity and

cooperation between setfiers.

5. The project stressed colonist leadership and did not demonstrate

paternalism. (p. 301).

She furfirer notes constraints including ”attempts to supplement incomes
firrough off-farm work had led to significant decline in production and land of
community organization had crippled fire potential of fire setfiement to respond
collectively to its problems” (p. 302). These same constraints were found to
plague the CPAs investigated in firis study. Two complaints among fire setfiers in
Findley’ s Latin American case study are the exact same fire findings of firis study,
namely ” improved access to transportation and credit have been identified
repeatedly as critical to the colonists” (p. 306). In order to combat the structural

constraints facing the setfiers, Findley recommends:

181

improvements in fire internal terms of trade

improvement in transport services

produced-oriented credit

adoption of mixed cropping systems

generation of off-farm employment [but not in the pioneer phase]; and
expanded healfir and education services (307).

emewwe

Henriques (1988) reviewed Brazilian settlement schemes and comes to
many of fire same conclusions as Findley, with similar implications for Soufir
Africa. Most notably she shows that ”the implementation of a colonization policy
has not significanfiy enhanced people’s well being” (338) due to the fact firat
”credit and technical assistance did not reach more firan one-firird of fire
setfiers”. In the Andean highlands of Peru, Carpio (1988) found that several
structural failures inhibited the progress of setfiement schemes, specifically fire
lack of overall organization, credit, and marketing facilities.

Empirical evidence on fire progress of fire CPAs in South Africa is derived
from the Department of Land Affairs 1999 Annual Quality of Life Report. This
report details the slow pace of progress of fire CPAs and notes:

The data available indicates that beneficiaries still earn their livelihoods

from conventional sources, such as pensions, remittances, wages, and to a

less extent subsistence agriculture. Land reform projects have not began

[sic] to change fire livelihood activities of communities (21).

The aufirors point out firat overall progress has been slow and firat fire

changes envisioned have not materialized. They note firat some activity has

occurred but ”firere is potential for growfir” (21).

182

Evidence from elsewhere also gives an indication of what could be expected
given more facilitating conditions. Initial success was obtained under fire Federal
Land Development Aufirority Project in Malaysia started in 1955. Bahrin (1988)
shows a steady increase in settler incomes after fire fourfir year. Initially incomes
fell, but by fire firird year firey had surpassed the pre-setfiement level. The
FELDA resetfiement scheme was endowed with significant financial,
institutional, and technical resources to carry out its tasks, in fact, far more firan
South Africa’s land reform program. For instance, Bahrin notes fire following
personnel resources were devoted to each setfiement scheme:

1 manager

1 assistant manager

4 field supervisors

8 field assistants

3 setfier development assistants

2 clerks

2 secretaries

1 driver
1 office boy

(Bahrin, 1988, p.107)

In addition, applicants to resetfie were evaluated to determine their
likelihood of success: ” the people to be awarded fire most points were those who
were landless, around 35 years of age, with a large number of dependents, and
physically fit” (97). This is in stark contrast to the self-selection mefirod
undertaken in Soufir Africa.

Limited progress in projects undertaken to transform rural society (e. g.

resetfiement, land reform) is a shared problem, however. Ofirer nations have

183

experienced similar problems regarding the lack of a facilitating environment for
change. Chole and Mulat (1988) describing Efiriopian resetfiement explained ”fire
settlement programs are too cosfiy and face serious problems of mismanagement
and low land and labor productivity. The inability to allow sufficient time to
prepare, plan, and implement setfiement programs is a cause of waste and high
costs” (182).

Somalian resetfiement programs ran into similar difficulties. Ragsdale and
Ali (1988) cite a program report explaining lack of progress in the Setfiement
Development Agency / Coastal Development Project:

Project implementation and evaluation, and particularly annual budgeting,

were adversely affected by the lack of documents...and of disciplined

recording, reporting, and monitoring procedures. . . Sub-sectoral

programmes appear to lack underlying definition of long range objectives

and strategies of the interrelated roles of their projects (SDR 1979c: 106-7

cited in Ragsdale and Ali, 1988, p.219).

This literature illustrates some of fire factors that both encourage and
discourage success in resettlement projects. Primary enabling conditions
included access to credit, strong organization, provision of transportation, and

technical assistance. The absence of these conditions in one form or anofirer has

led to less successful projects.

184

6.4 Impediments at the Regional Scale

Land reformers failed to recognize firat basic structural inequalities in Soufir
African society had not been addressed at fire time land reform was initiated.
Reformers desired to transform rural society using land reform as a device for
change rafirer firan by addressing fire difficult questions of tenure, aufirority,
gender inequality and unequal access to capital in rural society first. Land reform
was a ’magic bullet’ firat would stimulate the economy and alleviate rural
poverty. The notion firat any one stimulus can redress hundreds of years of
unequal development needs to be re-examined. The next six sections address fire
structural inequalities, biases, or irrefficiencies in fire rural South African
economy inherited from fire aparfireid state that block or in some manner skew
fire development of the rural regional economy necessary for such projects
successful. The literature on successful projects showed firat not only were fire
projects’ internal dynamics critical, but a facilitating local and regional climate
was also necessary for success (Bahrin, 1988; Findley, 1988). In South Africa, fire

following issues impinge upon CPA success:

Tenure Relations
Tenure reform has occurred, albeit in small steps. The tenure system is

confusing and does not provide for widespread security. Women still experience

185

a high degree of insecurity, especially in tenure. Aufirority over tenure needs to
be clarified and codified in fire former homelands, including the issuing of tifies
(Cross, 1988). Rafirer firan freezing existing tenure systems into place, as has
occurred, tenure has to be addressed wifirin the context of rural democracy
building.

Insecurity resulting from fire ’placelessness’ brought on by fire lack of legal
tenure leads to a host of problems for fire household and in firis case the CPAs as
well. Households have to spend vast amounts of time trying to maintain their
tenuous hold over the land firat they occupy, often confronting competing claims
on fireir plot. Moving from plot to plot incurs expense of time and money,
precious resources firat could be invested in fire CPA. Differentiation between
tenured and non—tenured members necessarily advantages those with security
property rights. They have more time and / or money to devote to the CPA
causing cleavages to develop. Reforming the tenure system to provide stability in
rural areas would have the effect of freeing up households’ time and money,
which in turn, could be invested in fire CPA. Subsistence agriculture is a time
consuming task. Constantly relocating or fire fear of expulsion distorts fire
functioning of firis system. Households have to concenfiate more resources into

an already burdening system.

186

Overlapping and Mulh'ple Claims to Local Authority

Wresting control of rural areas away from fire chiefs has proven to be a
difficult task for fire new state. The inherited system of ”traditional authority” is
neifirer traditional nor has universal legitimate aufirority in rural areas. The
current realities of traditional aufirority were created by and suited the aparfireid
state, and as such reflected its interests (Cross, 1988). The strangulation of
democracy at the local level in rural areas must be addressed. In reference to fire
CPAs, aufirority lies clearly with the central committee in all but one CPA. In fire
Monyamane CPA, fire confusion over the legitimate aufirority must be clarified.

South Africa's system of competing aufirorities denies real democracy a
chance in rural areas. Where inhabitants demand a ”traditional” chief system,
strict controls have been placed over the administration. Where possible local
councils (a similarly democratically elected body) should seek to broaden fireir
mandates. In any case, overlapping and multiple claims to authority must be
addressed and one clear and legitimate body must rule. Recent steps to
strengthen fire local councils are encouraging and must continue. Resistance
from chiefs has proven difficult to surmount, however, fire transition to
democracy in rural areas will not be complete until all citizens have a voice in
fireir affairs at fire national, provincial, and local levels.

The membership of CPAs have to function in two roles, one on fire CPA

where authority is clear, and fire other in fireir places of residence where

187

6.4 Impediments at the Regional Scale

Land reformers failed to recognize that basic structural inequalities in Soufir
African society had not been addressed at fire time land reform was initiated.
Reformers desired to transform rural society using land reform as a device for
change rafirer firan by addressing fire difficult questions of tenure, aufirority,
gender inequality and unequal access to capital in rural society first. Land reform
was a ’magic bullet’ firat would stimulate the economy and alleviate rural
poverty. The notion firat any one stimulus can redress hundreds of years of
unequal development needs to be re-examined. The next six sections address fire
structural inequalities, biases, or inefficiencies in the rural South African
economy inherited from fire apartheid state firat block or in some manner skew
fire development of fire rural regional economy necessary for such projects
successful. The literature on successful projects showed firat not only were the
projects’ internal dynamics critical, but a facilitating local and regional climate
was also necessary for success (Bahrin, 1988; Findley, 1988). In South Africa, fire

following issues impinge upon CPA success:

Tenure Relations
Tenure reform has occurred, albeit in small steps. The tenure system is

confusing and does not provide for widespread security. Women still experience

185

authority is not always clear. The CPAs represent an alternative source of
aufirority and as such threaten local chiefs. In fact, several chiefs were afraid firat

fire CPA would draw away participants labor and tribute.

Gender Inequities

Redressing gender bias has been one of the central initiatives in the quest to
transform Soufir African society. Employment reform, access to health care, and
equal status under the law have all been affected since 1994, however, rural
women have yet to feel fire full force of these reforms. Women still suffer
unequal access to the credit and labor markets and continue to receive
disproportionately lower wages than fireir male counterparts. Most glaringly
women still must rely on a male relative in cases involving access to land.

Since large numbers of women are involved in fire CPAs, addressing
gender inequities could strengfiren fire associations. Women would be able to
contribute in more meaningful ways if they had security of tenure in fireir places
of residence and a higher degree of food security. As shown in the Review of fire
Literature, women are the primary agricultural producers in the Norfirern
Province. Gender inequity dampens production and weakens the development

of livelihood systems.

188

Limited and Unequal Access to Capital

The availability of capital in fire rural economy is still restricted, with banks
weary of investing in black agriculture. Credit is extended at shops for food and
furniture, but rarely for agricultural needs. In contrast to fire general situation in
fire rural areas, fire Department of Land Affairs has been active in extending low
cost loans for fire CPAs. Until more credit becomes available, people will still
find entry into agriculture prohibitive.

Credit here refers not to that lent to fire CPA itself, rafirer to households and
individuals. It is also important to note that agriculture tends to be fire primary
focus of credit reform in rural areas (Eicher and Staatz,1998). I maintain firat
credit extended for school fees, transportation (in fire form of subsidies to
transport providers), and microenterprises can also have significant feedbacks in
the rural economy. Of course, credit for agriculture is in dire need in fire
N orfirern Province, but given the lower levels of returns to agriculture there firan
in other places, credit extension in firese other areas will also stimulate the rural
economy. The extension of credit may eifirer free up fire time for, or may directly

facilitate participation in the CPA.

A Skewed Labor Market and the Continuing Urban Bias
Soufir Africa’s labor market is still geared toward providing cheap labor to

fire major industrial centers (Lipton, 1996). Efforts to raise labor wages in rural

189

areas are met with vociferous resistance from organized (commercial) agriculture
and World Bank-oriented government economists in Pretoria. Labor unions have
had litfie impact in firese areas. As long as Gauteng looks enticing as a labor
option, the rural areas will continue to suffer from lack of labor.

Urban bias persists in South Africa (Lipton, 1996). Urban areas receive fire
larger portion of investment and consume fire bulk of tax revenues. For example,
roads, schools, and clinics are far more accessible and far better funded in the
urban centers firan in fire rural areas. Service delivery in fire rural areas in fire
first six years of independence has been abysmal. Housing, land reform, and
water provision have all fallen short of fireir service delivery targets, especially in
fire rural areas. Sanitation reform has been most successful.

The African National Congress (ANC) is a party of urban blacks and as
such has neglected its rural brofirers (Weiner and Levin, 1997). The issue of land
reform has been placed on the back burner. The rhetoric of fire pre-independence
ANC on land reform has failed to translate into action. Fault for fire failure of
land reform clearly rests with the ANC, which has since placed emphasis on

urban issues, such as labor and transportation reform.

190

6.5 Comparing Results

When faced wifir, and compared to, fire performance of cooperatives in
Africa, fire question can be raised, did fire CPAs perform any better? Using
Okuneye’s indicators of effectiveness, the answer has to be ”not much".
Okuneye judged cooperatives on four criteria, namely:

1. Viability over time

2. Facilitation of fire provision of services

3. Promotion of egalitarianism

4. An increase in agricultural earnings

(Okuneye, 1985, p.6)

On fire first indicator, judgment is early but all of the available data points
to fire fact that the CPAs have faltered, or at least stalled, bringing into sharp
question fire ability of the organizations to withstand fire test of time. On fire
second indicator, CPAs have faired somewhat better as they have provided
much needed capital and operating costs and in that respect have created a
facilitating environment for fire provision of services, but have not actually done
much to affect a sustained higher level of service provision. The analysis of fire
data in firis study by gender showed that fire CPAs strive to be more egalitarian,
but have serious issues to face. Finally, there is no evidence firat agricultural

earnings have increased at all. On firis indicator, fire CPAs have completely

failed.

191

6.6 The Social Forces Hindering Land Use Change

The six reasons for fire lack of livelihood transformation on Communal
Property Associations listed above constitutes not only livelihood constraints but
also block land use change firat would ofirerwise occur. Take for example the
case of fire Mahlambandlovu CPA. When the white farmer sold the farm to the
association he also sold firem fire complete inventory of his farming capital from
tractors to implements. The farmer had planted maize crops on that land in
previous years and left the farm in a condition where it could be readily utilized.
Given that fire association that bought fire farm consisted of 396 members, an
assumption of land intensification is reasonable. These people were drawn from
the area near fire farm, they knew the environment, fireir livelihoods were not
disrupted, and fire vast majority expressed both fire need and the desire to farm
crops. Furfirer, firey possessed some knowledge of agriculture, as the majority
had long been subsistence farmers. Why then did so litfie change (and so litfie
change toward intensification) take place? .

Not only are fire above reasons impediments to livelihood transformation,
but firey also stymie land use change. The literature on land use change focuses
on forces firat promote or cause change, wifirout investigating cases where social
forces inhibit change. The oversight is due to fire fact that land use change
studies have a tendency to be data led. Investigation of fire processes of land use

change occurs only after sudr changes have been clearly identified usually in some

192

type of remotely sensed imagery. Forces firat block change would rarely be
identified, because no a priori land use change would ever be detected on the
imagery, or at least not enough to prompt investigations over other areas where
more change had occurred. For instance, firis study would have never investigated
the Communal Property Associations had it been a data driven project. There is far
more conclusive evidence of land use change outside fire CPAs. I am arguing
here that fire ”ofirer” side of ”driving force of land use change” has been ignored
- fire social forces firat block potential change. I have followed firis line of
reasoning precisely because firese blocking forces direcfiy relate to the social
system in which firey are embedded - firey feedback to livelihood systems which
can drive land use change (or not).

This is fire primary reason I have attempted to link these literatures
togefirer. I show here firat the same forces that effect livelihoods also effect land
use change and fire two are intricately connected, but not always in the manner

fire literature has demonstrated.

6.7 Accomplishments of Land Reform

Despite fire failure of land reform to affect widespread change in rural
areas, fire land reform program has had several successes. First, gender equity
programs have been implemented for the first time in many rural areas. The fact

firat so many women have been included in what, seven years ago, would have

193

been considered a man’s domain is notable. Second, land has been redistributed
and more will be redistributed in the coming years. Alfirough fire amount of land
redistributed is paltry, lessons learned in firese first steps will help in the crafting
of future programs. People are being returned to fire land that has for so long
been alienated from them. Third, fire publicity around land reform (and lack
firereof) in Soufir Africa has created a vigorous debate. Credit should be given to
fire Department of Land Affairs for generating publicity and keeping fire issue of

land reform alive.

6.8 Synthesis: The CPA, the Landscape, and fire Geography of

Livelihoods

In firis analysis, the CPA has been isolated from the surrounding landscape,
bofir physical and human, in order to discern its efficacy in delivering change.
Boundaries of fire CPA separate it on the imagery from its surrounding area and
CPA members were surveyed separately. To equate fire entire physical and
human landscape in the assessment of fire CPAs would have been bofir a
misnomer and would have undermined fire core of firis study. It was necessary
from an analytical point of view to set up the dichotomy between the CPAs and
their surrounding landscapes. However, at firis point it is important to place the
CPAs back into bofir fireir physical and human landscapes. The CPAs represent a

bridge between different scales of analysis. They link fire livelihood systems of

194

households wifir fire broader regional patterns - here referred to as fire
geography of livelihoods. They also direcfiy impact fire physical landscape -firey
are not set aside and devoid from interaction. Again, firey provide a bridge
across scales - subsistence production via fire household is shown to impact
broader land use patterns, specifically fire land use extensification on the CPAs.

The CPA has affected an array of disarticulated impacts on both livelihoods
and land use change. The CPA is only part of a suite of livelihood strategies, not
fire central livelihood strategy as envisioned by planners and members. Yet, it
has begun to show signs of contributing to fire strengthening of rural livelihoods.
The CPAs have affected extensive land use changes. Yet fire surrounding areas
have experienced widespread intensification. This disarticulation of space is
symptomatic of fire impact of the CPA on both the physical and human
landscape. While no member claimed to make a living solely from their activities
on fire CPA, many members’ livelihoods were strengfirened by participation. On
the land that could have been expected to undergo rapid intensification, fire
transferred CPA lands, extensification occurred. Yet, the neighboring land
experienced intensification. The full impact of fire CPAs on the physical and
human landscapes has yet to be realized. The differential impacts are due largely
to fire fact firat fire CPAs have failed to become central, organized, and
substantive contributors to livelihoods and fire local rural economy.

The land use change maps presented in firis dissertation belie a geography

of livelihoods. The spatial manifestation of the disarticulated impact of fire CPA

195

on members’ livelihoods is represented on the map. Intensification has occurred
in areas where members live, yet none where firey (are supposed to) work - the
CPA. Livelihood strategies focus on utilizing fire available resources wifirin fire
local space economy. Subsistence farming still occurs outside fire CPAs.
Members expected firat production for profit would occur on the CPAs, thus
minimizing fire risk to livelihoods involved wifir allotting substantial time to fire
CPA. This tendency toward risk-aversion is not uncommon (Zimmerman, 2000).
Given the risk averse nature of subsistence farmers, fire geography of livelihoods
evident in the landscape is not exceptional.

The dynamic between fire household and fire CPA as units of production
fuels fire above noted disarticulation. Individuals face pressure between fireir
responsibilities to fire household versus firose to fire CPA. When fire pressure
mounted and a decision had to be made, firis study has shown firat most
individuals chose their household over fire CPA as fire mode of production and
reproduction, as it was perceived as far more effective in sustaining their
livelihood.

These disparate and seemingly disarticulated forces shaping land
managers’ / household decisions across bofir scale and space, impacted by
political, environmental, social, and economic forces speak to fire fireoretical
framework in which fire study is set - namely, political ecology. When examined
in fire context of a political ecology approach the disarticulation manifest in the

landscape is clearly a result of the forces competing for the interests of the

196

household - some households respond to fire CPA while ofirers have abandoned
it. Certain households have ”scaled up” believing firat in collective action lies
greater livelihood security. Political and economic forces (primarily vested in fire
state) have influenced firose households to focus many of their productive
resources on fire CPA. Being risk averse, however, these same households have
weighed fire potential of fire CPA against its inability to provide material benefits
and have decided to retain their local gardens and fields. In firis respect, CPA
members are operating across scale - retaining their own gardens and fields
(household scale) while at the same time participating in the larger farming
activity of fire CPA (local / regional scale). In the context of a political ecology
approach toward understanding the dynamics of firis case study it is important
to note firat households have not chosen to make only one land use decision, but
multiple land use decisions with multiple impacts across space and on several local
level scales. The seeming disarticulation in fire landscape is a result of fire

multiple land use decisions.

6.9 Summary

I have shown here firat land reform which aimed to change land use
patterns and alleviate rural poverty has, of yet, accomplished neifirer. Evidence

was drawn from statistics firat showed litfie change in livelihoods and satellite

197

images firat showed even fewer changes to more productive land uses on
transferred lands.

I have omitted in firis chapter a discussion of the social driving forces of
land use change. I have focused on the Communal Property Associations and as
noted in chapter four and five, not enough land use change occurred on the CPAs
for me to undertake any analysis on which Icould base scientifically defendable
statements. On several of the CPAs, no land use change occurred. Therefore, it is
impossible to explore fire social driving forces of change where no change or
very litfie change is evident. In effect, I presented in firis chapter the social
hindering forces of land use change.

I have shown six reasons for fire lack of change, bofir in livelihood systems
and land use - general disorganization (social disruption), the large farm sizes,
lack of labor and skilled members, gender bias, capitalization problems, and fire
age structure of fire membership. I presented evidence firat fire development of
the CPAs has stalled. The transforming potential of fire CPA program has only
begun to affect change in livelihoods and land use. I firen discussed fire issue of
time, specifically whefirer or not fire CPAs had matured enough to conduct an
analysis. I next raised fire issue of expectations of progress, drawing on case
studies to gauge what could be expected from fire CPAs at firis stage in fireir
development vis-a-vis ofirer projects from around the world. Finally, I
highlighted fire successes and failures of fire land reform program, specifically

addressing reasons the CPAs have faced such difficulty in a regional context.

198

Wm

Conclusion

I have shown in firis study fire failure of land reform to substantially alter
bofir livelihood systems and land use among fire study sites in fire Norfirern
Province, Soufir Africa. Specific questions were raised in firis dissertation relating
to fire impact of land reform- did it deliver, what was the scope of change, did it
impact livelihoods in a substantive way? I will now review the assumptions and
findings of firis study, fire contribution to science and fire literature, and will
conclude wifir a discussion of fire limitations of fire study and implications for
future research.

Land reform has not delivered the results expected by either government
planners or project beneficiaries. However, expectations can far outpace realities.
Immediate post-liberation Soufir Africa was deluged in a flurry of promises,
high-hopes, and ambitious service delivery figures. I assumed in firis study firat:

1) The introduction of a land reform project into a rural, poor area would

have fire effect of raising incomes and strengthening livelihood systems

2) Projects have not been abandoned and, given fire degree of government

support and general enfirusiasm among fire population for land reform,

the level of activity surrounding fire farms is high.

199

3) Significant changes in land use change on transferred lands would have

occurred. Given fire high degree of land pressure and fire need to
strengfiren and broaden livelihoods in rural areas, members of
Communal Property Associations would have actively mobilized and

fully utilized fire new land resource opportunities.

The findings of firis study include:

1)

2)

3)

A lack of clear and compelling evidence firat fire land reform projects
directly and unambiguously increased incomes and strengthened
livelihood systems. I did find, however, an indication firat livelihoods on
land reform project farms are improving slighfiy -it is probable that
some limited contributions to livelihood strengthening has occurred due
to participation in CPAs.

Projects have not been abandoned, however, the level of activity
(production or ofirer use) on fire farms was much lower firan expected.
In many cases firere was no activity on fire transferred farmland.

Land use on fire transferred lands is largely extensive rafirer than
intensive. It was found firat fire bulk of land use intensification has
occurred on land near to CPAs. This was attributed to fire fact firat the
membership of CPAs are not allowed to reside on the transferred farms

firemselves and are drawn to participate from nearby communities. No

200

one-to-one correspondence was found to indicate fire CPA was fire

driving force of fire change in fire bordering areas.

7.1 Contribution to Theory and Method

This dissertation informs fire literature and fireory specifically regarding
issues of livelihoods, land reform, land use change, and integration of social
theory and geographic information technologies.

Livelihoods

An understanding of livelihoods systems is essential to firis study for
several reasons. First, large numbers of people suffered severe economic
hardships as a result of fire aparfireid system. The current government strives for
social transformation, yet for firis social transformation to occur an
understanding of fire dynamics of livelihoods is essential.

Second, fire vast majority of rural South African households are still
dependent on local environmentally based mefirods of securing an income,
namely agriculture and livestock husbandry. To understand how land reform
might effect livelihoods, fire livelihoods firemselves must first be understood in
fireir local environmental context.

Livelihoods were shown in firis dissertation to be similar to firose in ofirer
areas of Africa and Soufir Africa. In fire study areas, people rely on remittances,
pensions, and petty commodity production for subsistence. This dissertation

firus provides furfirer evidence to fire trends identified in fire literature on

201

livelihoods firat show increasing dependence on entifiements (Baber, 1996;
Bernstein, 1992; May, 1996).

The dissertation contributed to fireory building by providing additional
empirical evidence on fire scope of livelihood systems in rural South Africa.
Understanding of fire range of livelihood options and degree of change in such
systems has been strengfirened by fire findings presented herein. Specifically, firis
study showed firat rural households have options from agriculture to petty
commodity production to formal employment to participation in CPAs as
livelihood choices. Scientifically, fire findings presented in chapter four and five
support basic theoretical arguments (Watts, 1981; Bryceson, 1997; Campbell,
1990; Chambers, 1992; Vogel, 1993). Furfirer, livelihood systems approaches have
been broadened by fire integration in firis dissertation of land use change wifir
livelihoods. Investigation of land use change was shown to be compatible wifir

simultaneous investigation of livelihoods, where similar forces shape both.

Land Use Change

Land use change has been used here as an indicator of fire progress of land
reform. Measurement of change caused by land reform can be readily detected in
fire changes in land use in firis study. Transferred land was intended to be used
by fire communities for agricultural production and livestock husbandry and

government planners anticipated that land use would intensify. Identifying fire

202

scope of land use change provides empirical evidence as to what happened on
fire ground.

This dissertation has generated a range of ideas regarding land use change.
It has contributed to fireory on land use change by raising fire issue of ”no land
use change” versus ”land use change”. What has been shown is fire need to
investigate fire ”other side” of land use change - the forces that blodc land use
change. The literature is inundated wifir examples of ”social driving forces” but
no mention of what might be preventing change firat would ofirerwise occur.
Why is ’no land use change’ just as important as ’land use change’? Models
predict what we don’t know based on statistical likelihood of fire replication of
what we do know. One key flaw wifir models, particularly land use change
models, is firat ”unexpected” forces or changes occur, firus invalidating fire
results. Could firese ”unexpected’ forces be the blocked land use changes
becoming unblocked?

This dissertation has shown firat firat knowledge on the social forces
driving land use change can not be broadened and expanded wifirout
identification of fire obstacles firose forces must overcome to be manifest in fire
landscape. For firis case study, those obstacles included social traditions firat
alienated women from full participation in fire land reform program, a skewed
labor market, contradictory tenure relations, and a lack of labor and skills on
CPAs. The point is firat land use change in firis respect doesn’t just happen -

several forces determine which communities/households/ individuals overcome

203

firese forces and which ones do not. There is a suite of social, political,
environmental, and economic factors firat either facilitate or block change.
Understanding firis difference is the real fireoretical challenge and firis
dissertation has begun to address firis question. Land use change is
conceptualized in firis dissertation from a broad perspective, incorporating many
elements of fire global change literature wifir fire political ecology and feminist
political ecology literatures, firus contributing to each. The investigation of land
use change has shown simultaneous intensification and extensification occurring
in these landscapes. As such it challenges the notion of broad, overarching

driving forces driving change in a reproducible and predictable manner.

Land Reform

Land reform forms fire heart of firis study. This dissertation provides
measurement of fire progress of land reform. The firrust of social transformation
of rural areas in post-aparfireid Soufir Africa has been undertaken firrough fire
implementation of land reform projects. Thus, to gauge fire state of rural Soufir
Africa, one must investigate the progress of land reform projects.

The impact of land reform on bofir livelihoods and land use change in the
study areas was shown to be less firan anticipated by project members and
planners. This dissertation has informed fire literature on land reform by clearly
highlighting fire failure of fire Communal Property Association program to

provide meaningful livelihood-enhancing change.

204

The assumptions set out in fire literature that capital injection will lead to
transformation were shown to be erroneous by firis study. In firis respect, fire
dissertation rejects some of fire assumptions set out in land reform fireory and
literature, namely firat 1) redistribution results in increased incomes for
participants, 2) land reform is an impetus to local economic change, and 3)
redistributed farm units become self sustaining and fire involvement of
government will become limited over time (Eicher and Staatz, 1998; Van
Rooyen, 1998; Van Zyl, 1996).

Further, firis study informed fire literature on cooperatives and resetfiement
and many of fire conclusions of firis study are supported in firat literature
(Gyllstrom, 1991; Adeyeye, 1991; Kimario, 1992; Holmen, 1990). The six
constraints facing fire Communal Property Associations in South Africa are ones
that have been encountered across Africa in fire past. This study was able to
show firat fire organizational structure of fire CPAs is inherenfiy flawed. By
comparing fire structure of cooperative farms wifir fire CPAs and firen comparing
fire results, it is clear firat the CPA functions and performs as a cooperative, wifir
fire same limitations. As with many cooperatives, the CPA has failed to create the

level of material benefits necessary to ensure its own survival.

Contributions to Theog
Several contributions to fireory and mefirods were made by firis study. First,

firis dissertation undertook to integrate bofir literatures / fireory and mefirods

205

from several sources. In fire exploration of mefirods, it was found firat fire
combination of traditional social science mefirods, such as interview schedules
and informal interviews, and remote sensing /GIS provided bofir opportunities
and challenges.

What were fire main contributions of fire integration exercise? Regarding
livelihoods and land use change, one methodological approach is not broad
enough to capture fire relationship presented here. Norgaard (1989) maintains
firat mefirodological pluralism can advance understanding in science. By
integrating methodological approaches, this dissertation has answered some
questions and raised ofirers regarding how a one-to—one relationship could be
made between pixels on a satellite image and specific social processes, firus
providing opportunities for future research.

The mefirodological integration here contributed to fire understanding of
the difficulties faced when attempting to mix social theory and geographic
technologies. The complex nature of social systems, often cited in fire land use
and land cover change literature as problematic to modelers and modeling can
be understood at fire household and community level firrough investigation of
the relationship between livelihood systems’ spatial, temporal, and scalar
manifestations.

As reflected in fire publication ”Our Common Journey” (NRC,2000),
livelihoods are a key component in fire understanding of global change, but fire

intersection of fire two topics is not well understood. The role and impact of

206

livelihoods still remain clouded from a global change perspective. This study
contributed understanding of how to begin to address livelihoods together with
and as a key component of land use change. The linking in firis study of livelihood
change with land use change showed firat fire feedbacks between the two are
dynamic, yet difficult to untangle. An action taken to change livelihoods often
results in a land use change, eifirer an intensification or extensification. For
example, when some of the female members of fire CPAs found that firey could
hire other women to garden for firem at fireir home plot, firey were able to affect
land use changes at fire CPA farm itself. When CPA members found firat firey
had to send off fireir children and relatives to work elsewhere, fire resulting lack
of labor led to less activity on fire CPA farm. These findings suggest firat
livelihoods and land use change should not stand alone in fire literature and in
fireory. The common area between fire two will provide fertile ground to
researchers interested in bofir fire social forces driving land use change and fire
interaction of livelihood system wifir fire local environment.

This dissertation has contributed to the political ecology framework by
incorporating gender, focusing on scale issues, and relating fire process of
livelihood change to changes in fire physical environment and fire subsequent
feedbacks. Political ecology was used as a broad framework for analysis but also
included key concepts from fire global change framework, post-modernism, and

feminist political ecology.

207

Rocheleau (1996) noted firat gender was an important component of
political ecology along with class, power relations, and issues of scale. This
dissertation supported Rocheleau’s notion of a feminist political ecology ,
specifically the importance of fire intersection of gender and ecology in
understanding livelihoods. Gender relations influenced CPA performance,
enabling some to participate while blocking others.

A second contribution to science and fire literature centers on scale. lessons
from fire land use and cover change literature are particularly relevant. For
instance, a case study - by - case study approach, common in fire livelihoods
literature, does not evoke firinking and critical response on fire broader patterns
of change (Meyer and Turner, 1994). This dissertation showed the importance of
cross-scale analysis. Here, fire CPA was a mechanism by which households
bridged fireir own production to production at a broader, community scale. The
relationships between the household and fire community scale of production
were detailed in firis study, e.g. women’s sequestering of labor at home as a
means of participating in the CPA. Locating the CPA wifirin the landscape and
relating individual action to broader patterns via fire CPA, firis dissertation
highlighted fire dynamic nature of people~environment interaction across scales
(Blaikie, 1994). It showed how household actions and agents could affect
landscape change firrough group action.

Regional linkages in scale were found, but only in fire sense firat regional

forces impacted fire CPAs (chapter six). The contribution of the CPAs to fire

208

economy of fire Norfirern Province and fireir place in fire regional employment
and livelihood patterns was not shown. This dissertation was also not able to
place fire CPAs in fire national and international scale. There is no indication of
how fire study CPAs might account for total national land use change or if the
CPAs have some ability to impact fire national economy. In firese respects, fire
study faces fire same constraints as previous studies described by Meyer and
Turner.

Furfirer, firis geography of livelihoods and findings on scale issues
contribute to fire study of structure and agency (Giddens, 1984). This study was
informed by structuration fireory as presented in fire sociology and anfirropology
literatures. Those disciplines oufiirred how individuals and groups in society
deal wifir structural obstacles, such as institutions and fire state, firat restrict fireir
ability to express fireir agency. It was shown here firat structure in society and in
fire CPA program itself inhibited fire actions of agents. Agents (members of fire
CPA in firis case) expressed fireir agency in fire area called ”environs” - firose
places outside fire CPA. The CPA as a structure designed to affect change is too
rigid and too closely linked to fire state to be a viable instrument of change. This
finding is supported in fire cooperatives literature (particularly Gyllstrom, 1991).
Agents, even when exercising fireir power firrough mass action, were unable to
surmount fire inequities in fire rural economy and inefficiencies wifir fire CPA.
These inequities and inefficiencies stood as structural barriers to effective action

on fire part of agents to fully utilize fire potential of the CPA.

209

These findings on structure and agency inform fire sociological and
anthropological literature and theory from a spatial perspective. This study
demonstrates firat bofir structure and agency are expressed in fire landscape. The
impediments to social transformation presented in the last chapter have spatial
manifestations, e.g. the lack of change seen in firis study. Structural forces such as
fire Gauteng labor market and laws regarding tenure blocked agents and groups
of agents (fire CPAs) from acting on fireir wishes to transform fireir land.

Structuration theory, in firis sense, has spatial component.

7.2 Limitations of this Study and Future Research

As a baseline study, firis dissertation has inherent limitations built into it as
a result of fire research design. These limitations will clearly affect future
investigation into firis topic. First, livelihoods in firis study were less clearly
differentiated between pre- and post- CPA involvement firan was anticipated.
There were far too few ”before and after” questions asked and firere were few
actual quantifications of income. This will limit future comparisons of fire data
from firis study, specifically investigations will not be able to show precise
income changes.

Second, fire results of integrating satellite remote sensing analysis and
social survey techniques wifir fire intent to understand fire ”social driving forces
of land use change” were ambiguous. Social driving forces were not explicifiy

referenced to instances of landscape change, firus inhibiting future studies in

210

fireir ability to map fire driving forces of change based on firis data. It was
difficult to identify fire specific driving forces of land use change and then relate
firem back to an image. Furfirer, firere was very litfie mefirodological
experimentation in fire land cover analysis, particularly regarding satellite image
processing. In future, exploration of ofirer methods of classifying dryland images
(such as green-up indices and ofirer NDVI based procedures) might prove
insightful.

Future studies will be limited by fire design of firis study firat
overemphasized fire role of agriculture in bofir livelihoods and the dynamics of
land use change. In many places, this dissertation has fallen into fire same trap as
many ofirer aufirors- arguing that unless agricultural production improves a
project in Africa is a failure. The impacts of fire CPA on non-agricultural
components of rural life and livelihoods was not fully investigated. Ofirer
schemes were successful because agriculture was fire central component of fire
livelihood system. This is not fire case in South Africa because non-farm income
is of such great importance, so agriculture-led projects might not be expected to
produce a similar result. Similarly, a clear time sequencing for the recovery of
livelihoods after fire transition into fire CPA was not presented. The study
showed firat the transition on CPAs continues and future studies will need to

consider firis dynamism as firey build upon fire findings of firis dissertation.

211

7.3 Summary

This study undertook to investigate fire impact of a land reform event on
fire livelihoods of rural participants and any subsequent change in land cover
and use. The literature on livelihoods, land use change, land reform, gender
relations, and resetfiement were reviewed to provide insight to the research
questions. These questions focused on how livelihood systems responded to a
land reform change, whefirer or not fire social driving forces of land use change
is clearly evident in satellite imagery, and whether or not land reform in Soufir
Africa result in a land use and / or livelihood change. Recurrent firemes included
the impacts of gender relations and issues of scale.

This study contributed to the livelihoods literature by re-confirming fire
importance of remittances, pensions, and petty commodity production in fire mix
of economic activities. The finding firat land use change can be blocked
contributed to fire land use change literature as well as showing firat land reform
did not meet its intended goals of land use intensification and subsequent
livelihood enhancement. Some assumptions in fire land reform literature were
challenged, specifically those asserting improvements in household and regional
economies as a result of reform.

The exercise of mefirodological integration provided insight into the
difficulties of bringing togefirer social theory and geographic information
technologies. Specifically, linking explicit livelihood activity wifir land use

change was shown to be difficult but possible. The fact that both livelihood

212

change and land use change are often effected by fire same forces informs both
literatures as to fire need to examine fire topics simultaneously rather firan
separately.

Findings of fire study included a lack of clear and compelling evidence that:
the land reform projects direcfiy and unambiguously increased incomes and
strengthened livelihood systems; firat projects have not been abandoned, however,
fire level of activity (production or ofirer use) on fire farms was much lower firan
expected; and very few changes have occurred in land use on fire transferred
lands. Limitations included fire underutilization of available geographic mefirods
and fire convenient sidestepping of complex issues of scale and gender relations.

This is not fire final chapter on land reform in Soufir Africa. Although fire
focus has shifted, reform will still occur. The real question is - how long will fire
rural disenfranchised majority wait for land reform? Clearly, change has been
too narrow and taken too long. One danger facing South Africa is fire same one
firat faces ofirer nations and is now crippling Zimbabwe - fire overt politicization
of land reform for the maintenance of ruling party power. The conditions in
Soufir Africa are very similar to firose in Zimbabwe after 1980, namely a large

number of land hungry people confronted wifir slow and ineffective delivery.

How long will South Africa wait?

213

3:32.. 62:32 05 e. «86 >35 3 cow-80.. a. 2:2“.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

M
(Q0 .
[
E3322 2: o 5%: x
Caucaom 00:30.5 U
283
/r\» «..me MW
\\1 hf \. qllr x/Jrl/rc/L (\f
\\ f A .w m J, /
J. A $455“. fr _r\\ if}.
. m. i.e./16’s o
7 \ 755m * \
//,,,/,mu~//,//,4 . [rt/r
_ /_ mz<z<>zoz V jwaflofimfiam
. A. _, .mmm_oz.>22 O X W
. ._ . e o o e an
F 0 25625.25; e 1e
_ ., 10880203. in _e_ \Lehnz
m . .\ _ x
z m. _ _ radix
sneer: _ x
\\
. 2 \f\

_ ....,.,, \|\r
_ \I\

, s r.

E )Irlkxk

 

 

 

214

 

 

 

 

 

 

 

 

 

N 8235 u:- <mo 32335.23; 3 sac—E 3.00 0...". "u 2:2“.

 

 

28050:! t n N r a

ban—50m 2.9.35
we... 09:82.2... n

2.30..

 

 

 

215

2.235 new (no abacus—egg! - econ .830 new on: new; 6 952".

 

esoeéx e .n a F c

 

.325 33%
recourse I r

1 Ir?

95.320 1 l
2:u_=o_..u<

25.. uoccemeafin r ,, . ; . [Ii/ill 4

ecouon

 

216

$3 32.3w new 55 30.33.522. .0 002:. .200 use“. 6 2:9“.

 

 

is

23050:! v n N r o

Eat—50m 2.9.35
.23.. 8.5.2.5 fl

“Eamon

 

 

217

2.9.35 use (no :>o_ucanEaEa2 . 82. 3:20 .830 new on: new.— 6 952".

fi l‘kll‘tl"-li\illll l l .

"NILE ,
232.55. s. n N F a

 

520m ccm_oEoI Logo“. , r

3523”. I , ., \ _ ’ ;/

canoe; {I\ . r s r r A
9:23ch r u, , . //:. r ,
o..3.:o_._m< 1 1. r 1 5 i r r r w . Z W
,, r > 1‘: \r r\ r 7
new; nachos—mama... D r u 4 h

218

nee—>5 new (no =>o.u:3E-Ea! . 83.82 09.2.0 .300 new 3: 2.3 6 2:2“.

 

 

 

.3»; I

.33... 3 2.5820 I
3.528.... 8 2.2880 I
:88; 2 22820 I
2:._:o_.a< 3 3.2330 E
.225 o. 23.323 B
2.528,. 8 238:9. I

, 2.382.. 8 23.32.? D

. _
.
_

69.2.0 oz ,

9.3 52.9.2.9... D
.23qu

E
9.32.35. v n N w o

 

_
.

 

 

 

219

9.32.55. 1 n N w o

Ewucaom 2.0.3.5
9.0.. 02.03:; n

2.000..

83 2.0.35 new 5.0 0:053:02 ..0 00!... .200 02-... K 2:0...

 

 

 

 

 

220

2.0.35 new 5.0 0583.5! . SON .300 new 00: 0:3 6 0.50.".

 

 

/, 23050:: e n N .. o

 

I
as... I
2.5890

.
. recourse
.
.
7 23.8.5...

. .
2.3 02.220... D

A 9.03..

 

"ND“

 

221

5

nausea. cw N F o

Emu—50m 3235
3:3 $2.232... D
ucwmm4

agar 2.2.2.5 3

an (to 2.222302 “.0 man-E 3.00 «an... "a 2:9".

  

 

222

2.235 new (no ecu—=35! . 821.030 .23 on: 2.0.. "3 2:2".

5 7
/ 23050:! v n N w o
\ 7

 

       

3.2.8.8". I
7 .3325 I
, 25.320
7 2332.? H“

7 9:3 tots—25F D
7 2.0qu

2.235 23 (no 235->5! . 88.82. 09.2.0 3: 2:... "2‘ 2:2“.

 

7

.5528.“ 0. canoe! I
22.8.6 8 :88;
2330:? 8 33oz
32.2.8 2 2.2396
2.825 2 23.8.5
9.2.3:? a. 2:396
3.6.281 8 2328...? I
38°; 8 2330...? I
23.380 2 23.32.: D
8:26 oz m L‘

23 3.8.2...» B
283

 

 

”nun
£32.55. 1 n N _. o

 

 

224

88 2.235 23 (no :osnotcom ho ounE. 3.00 02-“. "NF 2:9".

 

 

E
Eon—0:35. n N _. o
Emu—50m 3235
3a.. 02.83%... D
Egg
7
7
7
7
7 z

 

 

225

2.2.2—m use 5.0 592.35”. . 83 .830 .23 3... 3:3 "2 2:9".

 

 

        

 

.835 Egg 2825.5. n N F a
3:523. I
8:82. I
25.829 D
2332.? H7

95.. token—.2... U
2550..

226

 

 

82 0.3.35 0:0 (00 30000000”. .0 0005. 3.00 00.0". "3 0.50.".

1 l

33322 n N F

 

, Emu—50m u:0...>cm
03.. 00.3322... D

0:003

227

n:0...>:m 0:0 (“.0 :08000:0¢ . 000.. .030 0:0 00... 0:1. 6.. 0.50.".

 

 

          

.803 g. 28080.5. 0 n N r o
3.528: I
.8082. Flu.
25820 B
23.:0...0<

.23.. 0252.5 U
0:003

______

 

 

228

0:0...>:m 0:0 (:0 5000003”. . 83.03.. 00:25 .300 0:0 00: 0:3 6.. 0.00.".

 

fl
0:03:35. 0 n N

5.2... I

.5528: o. 508.... I
25.8.20 8 2.08.5 I
2.23:? 3 508.... will
.3528: 8 05.820 I
508.... 2 05.8.6 I
23.:0..5< 8 05.0020 n
.3525: 8 9.3.8:? I
0.082. 8 2.25:3 I!
05.820 8 2332.2 WI
35:0 oz U

0:1. 023.05; B

0:003

v

 

 

003 (:0 000.05?! 3 0005. 3.00 00.0“. "t 0.50.".

 

 

 

2805.0. 0

0:04 00:230.... D
0:000;

 

 

5.0 5202.5... . 080 5.60 05 8: 2.0.. "2 2:2".

 

 

508.5 ”H—
0:0.0005

0:000;

 

28022.: F ..

 

 

 

231

8.: <3 5202?... .o 002:. .28 8.0: a. 2:2“.

 

7 5522.: F

7 0:0._ 00:220.... n
7 0:000.

 

 

232

5.0 5202?: - 82 .260 05 on: 05.. .8 950....

 

 

 

. ;

25.5.0. F

5.82... n“.
0:0.0005 7

0:000;

 

 

 

233

 

(no E3553: . 83-32. aucazo .030 new 3: 2.3 "K 2:9“.

 

I 1
_ 25.820 8 38oz. D

.8825 2 2.232“. I
, 3:20 oz E

I 9.30..

 

23325. a...

\\\ /

 

/,/,

 

234

Sou 2.2.2—m u... 5...... .9950 3 09E. 50.00 o..."— "NN 2:9“.

 

 

E
88050:! v n N _. o

Educ—Sm 2.9.35
Ea“. 93.... n
9.33

 

235

 

    

Eggs: v. m N w

. ascoguam I

25.380 TD
. 2.3.8:? U
scan. 22...... n
. ucou¢4

2.9.35 Una 8.3“. .2200 2:32... . Sow .830 tea on: .25.. "nu 2:9".

 

 

236

8a.. aco._>..m E.- .Euu. .2230 3 ounE. .200 can“. .3 2:2“.

 

E
9.32.55. 1 n N r o

bauczom 2.9.35
6...“. 2...“... n
9.30..

 

 

 

 

237

.2232: ¢\m «2. o

. 3.2.8.8”. I
2.2380 [1
23.3.5... FL

Ea“. 2.52... D

.
.
venom;
.

oco._>..m E... E..."— 3550 223:. . 89 .260 E... 3: 0...... "mm 2:0...

 

 

 

238

2.2.2.0 .2... E... .228 2...... . 88-82 3:26 .260 2... 8.. 2.3 a... 2.5..

 

5
9.30.5.5. 0 n N ., o

.3528. 8 05.820 I
2.5.3:? 8 9.0.30.0 ‘_
33:00.00”. 8 0.3.32.2 E
.8083 8 2332.? .II.
.055 oz U

2.... 2...... D .H.
9.000.. a.

 

 

 

239

530050030 0003.09.33 0:0 0:30.20 :00.0m.:o :0050: 00:20.20 KN 0.30...

 

T

E

0.30505. v n N —. o
22.3.? I

20.320. H

. 2.0.. 8.83:5... I
_ 0.9... .0 8:85.30 08.22.35

.820. 82.3.0 330.8. D
0803

 

240

BIBLIOGRAPHY

ACARTSOD.1990. Understanding Africa's Food Problems : Social Policy Perspectives .
London: Hanz Zell

Adams, L. 1993. " A Rural Voice : Strategies for the Poor” In South African
Outlook 122: no pg. numbers

Adeniyi, P. 1986. ”Agricultural Land Use Inventory and Mapping In Nigeria: the
Application of Remote Sensing” in Remote Sensing and Tropical Land
Management Eds. Eden, M. and Parry, I. Barbosa P.M. , New York:
Wiley 6: Sons

African National Congress. 1994. The Reconstruction and Development Programme:
A Policy Framework. Johannesburg: Umanyano Publications.

African National Congress. 1955. The Freedom Charter. Johannesburg: ANC

Ahmed, I. 1985. Technology and Rural Women: Conceptual and Empirical Issues.
London: George Allen & UNWIN

Alamgir,M. and Arora, P. 1991. Providing Food Security for All . New York: New
York University Press

Amin,S. 1974. Accumulation on a World Scale. New York: Monthly Review Press

Anderson, M. and Woodrow, P. 1991. " Reducing Vulnerability to Drought and
Famine: Developmental Approaches to Relief " In Disasters 15: 43-54

Apethorpe, R. 1968. Land Settlement and Rural Development in Eastern Africa.
Kampala: Nkanga

Aspass, R. 1998. "Heading Households and Heading Businesses: Rural Kenyan
Women in the Informal Sector” in Professional Geographer 50(2).

Baber, R. 1996. " Current livelihoods in Semi-Arid Rural Areas of South Africa”.
in Land, labour, and Livelihoods in Rural South Africa. eds. Lipton, et al.
Durban: Indicator Press.

Babu, S. and Mthindi, G. 1995."Developing Decentralized Capacity for Disaster
Prevention: Lessons from Food Security and Nutrition Monitoring in
Malawi” In Disasters 19: 127-139

Barbosa,P. et.al. 1996. ”Performance of Several Landsat 5 Thematic Mapper image
Classification Methods for Crop Extent Estimates in an Irrigation
District” in International Iournal of Remote Sensing 17(18) p.3665-3674

Barrows, H. 1923 " Geography as Human Ecology” in Annals of the Association of
American Geographers (13)1 p.1-13.

Bassett, T. 1988. " The Political Ecology of Peasant-Herder Conflicts in the
Northern Ivory Coast". in Annals of the Association of American

Geographers 78(3) p.453-472
Bay, E. 1982. Women and Work in Africa Boulder: Westview.

241

Bebbington, A. 2000. ”Reencountering Development: Livelihood Transitions and
Place Transformations in the Andes” in Annals of the Association of
American Geographers. 90(3): 495-520.

Beneria, L. 1982. Women and Development: The Sexual Division of Labor in Rural
Societies. New York: Praeger

Berck,P. and Bigman, D. 1993. Food Security and Food Inventories in Developing
Countries. Oxon: Cab International.

Bernstein, H. 1999. ” Social Change in the South African Countryside? Land and
Production, Poverty and Power” Paper presented at the Conference
”Land And Agrarian Reform in South Africa: Successes, Problems, and the
Way Forward” July 26-28, 1999. Pretoria.

Bernstein, H. 1996. The Agrarian Question in South Africa. London: Frank Cass

Bernstein, H. 1994 ”Food Security in a Democratic South Africa". in
Transjbrmation (24) p.3-25.

Bernstein,H. 1993. “The Maize Industry: Business as Usual” In South African
Outlook 110: no pps.

Berstein, I-I., et.al. 1992. Rural Livelihoods: Crises and Responses. Oxford: Oxford
University Press.

Bernstein, H. 1990. ”Agricultural Modernization and the Era of Structural
Adjustment" In The Journal of Peasant Studies. 8: 3—35

Bilsborrow, R. and Okoth-Ogendo, H. 1992. ”Population -driven Change in Land
Use in Developing Countries" in Ambio 21(1): 37-45.

Binswanger, H. and Deininger, K. 1996. “South African Land Policy: The Legacy
of History and Current Options” in Agricultural Land Refin'm in South
Africa: Policies, Markets, and Mechanisms” eds. Van Zyl, et.al. Cape
Town: Oxford University Press.

Blaikie, P. 1985. The Political Economy of Soil Erosion in Developing Countries.
London: Longman.

Blaikie,P. and Brookfield, H. (1987. Land Degradation and Society. London:
Metheun

Brisco, B.et.al. 1989. ”Early Season Crop Discrimination with Combined SAR and
TM Data” in Canadian Journal of Remote Sensing 15(1) p.44-54

Brondizio, E., et al. 1994. “Land Use Change in the Amazon Estuary: Patterns of
Caboclo Settlement and Landscape Management” in Human Ecology
22(3) p. 249-278

Brown, M., et.al. 1998. Land Restitution in South Africa: A Long Way Home. Cape
Town: IDASA.

Bryant,C. 1988. Poverty, Policy, and Food Security in Southern Africa Boulder: Lynne
Rienner

Bryceson, D. 1997. Farewell to Farms: De-agrarianisation and Employment in Africa.
Hampshire, UK: Ashgate.

Bryceson, D. 1995. Women Wielding the Hoe: Lessons from Rural Africa jbr Feminist
Theory and Development Practice. Oxford: Berg

242

Bryceson, D. 1990. Food Insecurity and the Social Division of labour in Tanzania,
1919-85. Oxford: Macmillan
Buchanan-Smith, M. and Davies, S. 1995. Famine Emly Warning and Response: The
Missing Link. London: Intermediate Technology Publications
Buchanan-Smith, M. et al 1991. A Guide to Famine Early Warning and Food
Information Systems in the Sahel and Horn of Africa. Brighton: University
of Sussex
Bundy, C. 1988. The Rise and Fall of South African Peasantry. Cape Town: David
Philip
Bush, J. 1995. "The Role of Food Aid in Drought and Recovery : Oxfam's North ?
Turkana (Kenya) Drought Relief Programme, 1992-4" In Disasters 19: l‘
247-259
Campbell, D. 1998. "Towards and analytical framework for land-use change". in
Carbon and Nutrient Dynamics in Natural and Agricultural Tropical
Ecosystems, ed. L. Bergstrom and H. Kirschmann, New York: CAB 1)
International. . ‘
Campbell, D. 1990. ”Strategies for Coping with Severe Food Deficits in Rural
Africa: A Review of the Literature" In Food and Foodways 4(2): 143-162
Campbell, D. and Olson, J. 1986. "The Prospect for Desertification in Kajiado
District, Kenya” In The Geographical Journal 152: 44-55
Campbell, D. 1990. ”Community based Strategies for Coping with Food Security:
A Role in African Famine Early Warning Systems” In GeoJournal
Campbell, D. and Axinn, G. 1980 " Pastoralism in Kenya” In Ti?
Campbell, D. 1977. “Trends in Man-Land Interaction in the West African Sahel"
In Drought in Africa, 2. eds. Dalby, D. , Harrison-Church, P.J., and
Bezzaz, F. London: International African Institute
Carpio, J. 1988. ”Settlement Policies in the Forest Highlands of Peru” in Land
Settlement Policies and Population Distribution in Developing Countries:
Achievements, Problems, and Prospects. New York: Praeger
Carr,M. 1991. Women and Food Security: The Experience of SADCC Countries.
London: lTP.
Carter, M. and May, J. 1999. “Poverty, Livelihood, and Class in Rural South
Africa’ in World Development 27(1): 1-20.
Casterad, M. and Herrera, J. 1996. Performance of several Landsat 5thematic
mapper (TM) images classification methods for crop extent estimates
in an irrigation district International Journal of Remote Sensing 17: 3665-
3674.
Cekan, J. 1992 ”Seasonal Coping Strategies in Central Mali: Five Villages During
the 'Soudure' In Disasters 16:66-73.
Cemea, M. and Guggenheim, S. 1993. Anthropological Approaches to Resettlement.
Boulder: Westview Press.
Chambers, R. 1969. Settlement Schemes in Tropical Africa: A Study of Organizations
and Development. London: Routledge

 

243

Chambers, R. and Moris, J. 1973 MWEA: an Irrigated Rice Settlement in Kenya.
Munich: Weltforum Verlag
Chambers, R. and Conway, G. 1992. Sustainable Rural Livelihoods: Practical
Concepts for the 21“ Century. Sussex : Institute of Development Studies.
Chambers, R. 1989. " Editorial Introduction: Vulnerability, Coping, and Policy"
in Institute for Development Studies Bulletin 20(3) p.1-7
Chazan,N. and Shaw, T. 1988. Coping with Africa's Food Crisis . Boulder: Lynne
Rienner
Chole, E. 1990 Food Crisis in Afiica : Policy and Management Issues. New Delhi:
Vikas Publishing House. I’-
Chole, E. and Mulat, T. 1988. ”Land Settlement in Ethiopia: A Review of
Developments” in land Settlement Policies and Population Redistribution
in Developing Countries. Ed. Oberai, A. New York Praeger.
Chopak, C. 1991. Determinants of Rural Incomes in Communal Areas of Zimbabwe: ,3
Household Food Security Implications. Ph.D. Dissertation. East Lansing: l"
Michigan State University
Claassens, A. 1999. ”Land Rights and Local Decision Making Processes:
Proposals for Tenure Reform” Paper presented at the Conference ”Land
And Agrarian Reform in South Afiica: Successes, Problems, and the Way
Forward” July 26-28, 1999. Pretoria.
Cloud, K. 1977. Sex Roles in Food Production and Food Distribution Systems in the
Sahel. Tucson: University of Arizona
Colson, E. 1971. The Social Consequences of Resettlement. Manchester: Manchester
University Press.
Commanders. 1989. Structural Adjustment and Agriculture. London: Overseas
Development Institute
Corbett]. 1988. "Famine and Household Coping Strategies” In World Development
16: 1099-1112.
Cristiansen, R. 1996. ”Overview of Land Reform Issues” in Agricultural land
Reform in South Africa: Policies, Markets, and Mechanisms” eds. Van Zyl,
et.al. Cape Town: Oxford University Press.
Cross, C. 1995. ”Losing the Land: Securing Tenure in Tribal Areas” in Indicator
SA 12(2): 23-28.
Cross.C. 1988. " Freehold in the Homelands: What are the Real Constraints?". in
Towards Freehold? : options for land and Development in South Africa's
Black Rural Areas. eds. Cross, C. and Haines, R. Cape Town: Juta and
Co.
Cross.C. and Haines, R. 1988. " An Historical Overview of Land Policy and
Tenure in South Africa’s Black Areas" in Towards Freehold? : Options jbr
Land and Development in South Africa's Black Rural Areas. eds. Cross, C.
and Haines, R. Cape Town: Juta and Co.

 

244

Cunningham, T. 1984. Indigenous Plant Resources :A Baffler against Rural Poverty.
From fire Second Carnegie Inquiry in to Poverty and Development in
Southern Africa. Cape Town.

Dadhwal, V., et.al. 1996. Comparative Performance of Thematic Mapper Middle-
infrared Bands in Crop Discrimination. International Journal of Remote
Sensing 17: 1727-1734.

Davies, S., et.al;. 1991. Early Warning in the Sahel and Horn of Africa: The State of the
Art Brighton: University of Sussex

Davison, J. 1997. Gender, Lineage, and Ethnicity in Southern Africa. Boulder:
Westview

Derman, P. and Poultney, C. 1984. The Politics of Production and Community
Development in Rural South Africa. From the Second Inquiry into
Poverty and Development in South Africa. Cape Town

de Klerk,M. 1991. A Harvest of Discontent : The land Question in South Africa. Cape

Town: IDASA
Department of Land Affairs, 1999a. The National Land Committee’s Contribution to

land Reform, 1994-1998. Pretoria: Department of Land Affairs.

Department of Land Affairs, 1999b. Annual Report 1999. Pretoria: department of
Land Affairs.

Department of Land Affairs, 1998. Annual Quality of Life Report 1998. Pretoria:
Department of Land Affairs.

Department of Land Affairs. 1997. White Paper on South African land Policy. Cape
Town: CI'P Book Printers.

de Villiers, A. and Critchley, W. 1997. Rural land Reform Issues in Southern Afiica:
lessons for South Africa’s Northern Province. Sovenga, SA: University of
the North Press.

de Villiers, A. 1995. land Management and Rural Development Research Programme.
Sovenga, SA: University of the North.

Dias, J. 1981. ”Famine and Disease in the History of Angola" In Journal of African
History. 22:349-378

Diaw, K. 1990. Effizcts of Volta Lake Resettlement in Ghana- A Reappraisal After 25
Years. Hamburg: Institut fur Afrika-Kunde.

Dirks, R. 1980. ”Social Responses during Severe Food Shortages and Famine" In
Current Anthropology 22:21-44

Downing,T.,Gitu,K. and Kaumau,C.1989. Coping with Drought in Kenya. Boulder:
Lynne Rienner

Downing,T. "Monitoring and Responding to Famine: Lessons from the 1984—
85 Food Crisis in Kenya" In Disasters 14:204-225

Dreze, J. and Sen, A. 1991. The Political Economy of Hunger. New York: Oxford
University Press.

Ehrlich, D. et.al. 1997. ”Biomass Burning and Broad-Scale Land-Cover Changes in
West Africa” in Remote Sensing of the Environment 61: 201-209.

 

245

Eicher, C. and Staatz, J. 1998. International Agricultural Development. Baltimore:
John Hopkins Press.
Ellis, F. 1998. ”Household Strategies and Rural Livelihood Diversification” in The
Journal of Development Studies 35(1): 1-38.
Ellis, F. 1993. Peasant Economics: Farm Households and Agrarian Development.
Cambridge: Cambridge University Press.
Escobar, A. 1996. "Construction Nature: Elements for a Post-Structuralist Political
Ecology". in Futures 28(4) p.325-343
Fairhead, J. and Leach, M. 1996. Misreading the African Landscape: Society and
Ecology in a Forest-Savanna Mosaic. Cambridge: Cambridge University
Press
Feldman, S. 1995. "Feminist Knowledge Claims, Local Knowledge, and Gender
Divisions of Agricultural Labour: Constructing a Successor Science" in
Rural Sociology 60(1).
Findley, S. 1988. ”Colonist Constraints, Strategies, and Mobility: Recent Trends in ' i“
Latin American Frontier Zones” in land Settlement Policies and
Population Distribution in Developing Countries: Achievements, Problems,
and Prospects. New York: Praeger
Fuggle,R., and Rabie,M. 1992 Environmental Management in South Africa. Cape
Town: Juta and Co.
Gabre—Madhin, E. and Reardon, T. 1989. "The Importance and Determinants of
Women's Non-Agricultural Income Strategies in Rural Burkina Faso"
in American Journal of Agricultural Economics. Abstract
Geldenhuys,G. 1994. Food Security in South Africa. Johannesburg: South African
Institute of International Affairs
Geoghegan, J., et.al. 1998. “Socializing the Pixel and Pixelizing the Social in Land-
Use and Land-Cover Change” in People and Pixels: Linking Remote
Sensing and Social Science. eds. Liverman, et.al. Washington, DC:
National Academy Press
Gibbon,P. 1992. ”African Agriculture under Structural Adjustment with Special
Reference to Kenya and Ghana.” in The Journal of Peasant Studies. 20:
50-95
Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration.
Berkeley: University of California Press
Giorgis, D. 1989. Red Tears: War Famine and Revolution in Ethiopia. Trenton: The
Red Sea Press, Inc.
Golino, C., et.al. 1996. ” Providing Agricultural Support” in Agricultural land
Reform in South Africa: Policies, Markets, and Mechanisms. Eds. Van Zyl,
J, Kirsten, J., and Binswanger, H. Cape Town: Oxford University Press.
Groenewald, J. 1998. ”Land Tenure” in Agricultural Policy Reform in South Afiica
Eds. Van Rooyen, et.al. Cape Town: AIPA.

I _ ‘r ‘7--.-.1-1— . ‘
v "

 

246

Grolle, J. 1995. Famine, Migration, Resettlement, and Recovery: Case Studies
from Northwestern Nigeria. Ph.D. Dissertation, Department of
Geography, Michigan State University, East Lansing, MI.

Grubler, A. 1994. ”Technology” in Changes in land Use and Land Cover: A Global
Perspective. Eds. Meyer, W. and Turner, B. Cambridge: Cambridge
University Press.

Guyer, J. and Peters, P. 1987. ”Introduction” in Development and Orange. 18(2):
197-214.

Haggblade, S. et.al. 1989. "Farm-Nonfarm Linkages in Rural Sub-Saharan Africa”
in World Development 17(8) p.1173-1201.

Hansen, A. and Oliver-Smith, A. 1982. Involuntary Migration and Resettlement: The
Problems and Responses of Dislocated Peoples. Boulder: Westview Press

Hay, R. 1988. "Famine Incomes and Employment: Has Botswana Anything to
Teach Africa?“ In World Development 16: 1113-1125

Hay, R. and Rikuni, M.1988. 'SADCC Food Security Strategies: Evolution and
Role" In World Development. 16: 1013-1024

Heilig, G. 1993. Neglected Dimensions of Global land-Use Change” Reflections and
Data. Laxemburg, Austria: IIASA.

Henriques, M. 1988. ”The Colonization Experience in Brazil” in land Settlement
Policies and Population Distribution in Developing Countries:
Achievements, Problems, and Prospects. New York: Praeger

Hinchey, M. 1978. Proceedings of the Symposium on Drought in Botswana.
Worchester: Clark University Press

Hjort Af Ornas, A. "Pastoral and Environmental Security in East Africa.” In
Disasters 14: 115-122

HollistW., and Tullis,F. 1987. Pursuing Food Security: Strategies and Obstacles in
Africa, Asia, latin America, and the Middle East. Boulder: Lynne Rienner

Houghton, R. 1994. " The Worldwide Extent of Land-use Change” in BioScience
44(5) p.305-313.

Huss-Ashmore, R. and Katz, S. 1989) African Food Systems in Crisis . New York:
Gordon and Breach Science Publishers.

Hyden, G. 1980. Beyond Ujamaa in Tanzania. Los Angeles: University of California
Press

Hyden, G. 1983. No Shortcuts to Progress. Los Angeles: University of California
Press

Imbernon, Jacques. 1997. Properties, Changes and Links of Land Uses in the
Kenyan Highlands: Upper and Lower Embu Zones. AHI Technical
Report Series No. 7. Nairobi: International Centre for Agroforestry.

Isaacman, A. 1990. ”Peasants and Rural Social Protest in Africa.“ In African
Studies Review 33:1-120

Jackson, C. 1993. " Doing What Comes Naturally? Women and Environment in
Development" in World Development 21 (12).

Joekes,S. et al. 1996. Integrating Gender into Environmental Research and Policy

247

 

Jowkar, F. et al. 1991. Gender Relations of Pastoral and Agropastoral Production: A
Bibliography with Annotations. Binghamton, NY: Institute for
Development Anthropology

Jones, S. 1994. “Food Security Reserve Policy in Ethiopia: A Case Study of
Experience and Implications." In Disasters 18: 140-151

Kelly, M. 1992 " Entitlements, Coping Mechanisms and Indicators of Access to
Food: Wollo Region, Ethiopia,1987-88." In Disasters 16: 322-338

Kelly,M. and Buchanan-Smith, M. 1994. "Northern Sudan in 1991: Food Crisis
and the International Relief Response." In Disasters 18: 16-34

Kennedy,E. and Cogill,B. 1988. "The Commercialization of Agriculture and .
Household Level Food Security: The Case of Southwestern Kenya.” In if
World Development 16: 1075-1081

Khan,M., et al. 1992. “Composite Indicators for Famine Early Warning Systems."
In Disasters16: 195-206

Kinsey, B. and Binswanger, H. 1996. "Characteristics and Performance of
Settlement Programmes: A Review” in Agricultural Land Reform in
South Africa: Policies, Markets, and Mechanisms. Eds. Van Zyl, J, Kirsten,
J., and Binswanger, H. Cape Town: Oxford University Press

Kirsten, J. 1996. " The Potential for Creating Additional Rural Livelihoods in
Agriculture and the Rural Non-Farm Sector in Semi-Arid Areas: A
Case Study in the Northern Province". in Land, labour, and Livelihoods
in Rural South Afiica. eds. Lipton, et al. Durban: Indicator Press.

Koenig, D. 1987. After the First Years’ Move: Trip Report - Manantali Resettlement.
Binghampton, NY: IDA.

Kowet, D. 1978. land, Labour Migration, and Politics in Southern Afiica: Botswana,
Lesotho and Swaziland. Uppsala: The Scandinavian Institute for African
Studies

Kriger,N. 1992. Zimbabwe's Guerrilla War: Peasant Voices

Krueger, A. et al. 1991. The Political Economy of Agricultural Pricing Policy.
Baltimore: John Hopkins Press

Lambert, R. 1994. ”Monitoring Local Food Security and Coping Strategies:
Lessons from Information Collection and Analysis in Mopti, Mali.” In
Disasters 18:333-343

Lambin, E. and Ehrlich,D. 1997. ”Land-Cover Changes in Sub-Saharan Africa
(1982-1991): Application of a Change Index Based on Remotely Sensed
Surface Temperature and Vegetation Indices at a Continental Scale” in
Remote Sensing of the Environment 61: 181-200

Lambin, E. and Guyer, J. 1994 " The Complementarity of Remote Sensing and
Anthropology in the Study of Complex Human Ecology”, Working
Paper No. 175, African Studies Center, Boston University

Lambin, E. 1992. ”Spatial Scales, Desertification, and Environmental Perception
in the Bourgouriba Region (Burkina Faso)" Working Paper No. 167,
African Studies Center, Boston University

'57-.

 

Wm.a mam-—
l

248

Lambin, E. 1986. ”Agricultural Inventory and Agrarian System Survey by
Remote Sensing”. in Proceedings of the Symposium on Remote Sensing of
Environment.
Larson, T. 1966. ”The Significance of Rainmaking for the Mbukushu." In African
Studies. 25: 23-36
Lemon, A. 1995. The Geography of Change in South Africa. New York: Wiley and
Sons
Lemon, A. 1987. Apartheid in Transition. Hants, UK: Gower
Leow, L. 1988. " Towards Resolving the New Confusion about Land Tenure”. in
Towards Freehold? : options jbr Land and Development in South Africa's 'FT'
Black Rural Areas. eds. Cross, C. and Haines, R. Cape Town: Juta and
Co.
Letsoalo, E. 1987. land Reform in South Africa. Johannesburg: Skotaville.
Levin, R. and Weiner, D. 1997. No More Tears: Struggles for land in Mpumalanga,
South Africa Trenton: Africa World Press * ‘
Levin, R. 1996. " Politics and Land Reform in the Northern Province: A Case
Study of the Mojapelo Land Claim”. in land, labour, and Livelihoods in
Rural South Africa. eds. Lipton, et al. Durban: Indicator Press.
Leys,C. 1975. Underdevelopment in Kenya. London: Heineman
Lindblade, K., et.al. 1998. ”Mitigating the Relationship between Population
Growth and Land Degradation" in Ambio 27(7): 565-571.
Lipton, M., et al. 1996. land, labour, and Livelihoods in Rural South Africa Volume
Two: KwaZulu-Natal and Northern Province Durban: Indicator Press
Lipton, M. and Lipton, M. 1993. " Creating Rural Livelihoods: Some Lessons for
South Africa from Experience Elsewhere" in World Development, 21(9),
p.1515—1548.
Liverman, D., et.al. 1998. People and Pixels: Linking Remote Sensing and Social
Science. Washington DC: National Research Council.
Liverman, D. 1990. “Drought Impacts in Mexico: Climate, Agriculture,
Technology, and Land tenure in Sonora and Puebla" in Annals of the
Association of American Geographers 80(1) p. 49—72
Mackintosh, M. 1989. Gender, Class, and Rural Transition: Agribusiness and the Food
Crisis in Senegal. London: Zed.
MacIntosh, A. and Vaughan, A. ” Experiences of Agrarian Reform in South
Africa: The Limits of Intervention” Paper presented at the Conference
”land And Agrarian Reform in South Africa: Successes, Problems, and the
Way Forward” July 26-28, 1999. Pretoria.
Makopi, S. 1999. ”Awards to Provide Security of Tenure and Comparable
Redress” Paper presented at the Conference ”Land And Agrarian Reform in
South Africa: Successes, Problems, and the Way Forward” July 26-28, 1999.
Pretoria.
Mamdani, M. 1976. Politics and Class Formation in Uganda. New York: Monthly
Review Press

fii
O I C

 

249

Manetsch, T. 1979. "On Strategies and Programs for Coping with Large Scale
Food Shortages" In Ecology of Food and Nutrition. 8: 209-218

Manshard, W. and Morgan, W. 1988. Agricultural Expansion and Pioneer
Settlements in the Humid Tropics. Tokyo: UN University Press

Marsh, S. , et al. 1994. "Development of a Computer Workstation for Famine
Early Warning and Food Security." In Disasters 18: 117-129.

May, J. 1999. ”The Structure and Composition of Rural Poverty and Livelihoods
in South Africa” Paper presented at the Conference ”Land And Agrarian
Reform in South Afiica: Successes, Problems, and the Way Forward” July
26-28, 1999. Pretoria.

May, J. 1996. " Assets, Income, and Livelihoods in KwaZulu-Natal” in Land,
labour, and Livelihoods in Rural South Africa. eds. Lipton, et al. Durban:
Indicator Press.

May, J. 1994. Rural Poverty and Institutions. Wits, SA: LAPC.

May, J. and Peters, A. 1984 a. A Study of Income and Expenditure in Rural KwaZulu
- Mbongalwane. Ulundi: KwaZulu Finance and Investment
Corporation limited

May, J. and Peters, A. 1984 b. A Study of Income and Expenditure in Rural KwaZulu
- Madadeni. Ulundi: KwaZulu Finance and Investment Corporation
Limited

May, J. and Peters, A. 1984 c. A Study of Income and Expenditure in Rural KwaZulu
-Nqutu. Ulundi: KwaZulu Finance and Investment Corporation
Limited

Mbithi, P. and Barnes, C. 1975. The Spontaneous Settlement Problem in Kenya.
Nairobi: East African Literature Bureau.

McCorkle, C. 1987. "Foodgrain Disposals as Early Warning Famine Signals: A
Case from Burkina Faso.” In Disasters.

McCusker, B. 1997. Coping Mechanisms to Drought Induced Food Shortages in
KwaZulu-Natal, South Africa. Unpublished Master’ s Thesis: Michigan
State University.

Mertens, B. and Lambin, E. 2000. ”Land Cover Change Trajectories in Southern
Cameroon” in Annals of the Association of American Geographers. 90(3):
467-494.

Meyer, W. and Turner, 8., 1994. Oranges in Land Use and land Cover: A Global
Perspective Cambridge: Cambridge University Press.

Meyer, W. and Turner, B. 1992. ”Human Population Growth and Global Land-
Use/ Cover Change” in Annual Review of Ecological Systems 23: 39-61.

Middleton, S. 1997. ”Women's Land Rights and Needs: The Case of Thornhill
and Merino Walk” in Women, Land , and Authority: Perspectives from
South Afiica. Cape Town: Creda Press.

Minnaar, A. 1994. Access to and Aflbrdability of Land in South Africa: The Orallenge of
Land Reform in the 1990’s. HSRC: Pretoria

250

db‘ ‘3
r _ l'.
a--l A!

' " E . "hr db

 

 

Mini, S. 1994. “Gender Relations of Production in the Eastern Cape and the
Restructuring of Rural Apartheid" in Africa Insight 24(4).

Mkandawire, R. and Matlosa, K. 1993. Food Policy and Agriculture in Southern
Africa. Harare: Sapes Books

Moller, V. ,et al. 1984. Poverty and Quality of Life among Blacks in South Africa.
From Second Carnegie Inquiry into Poverty and Development in
Southern Africa. Cape Town

Moock, J. 1986. Understanding Africa’s Rural Households and Farming Systems.
Boulder,CO: Westview Press

Moore, D. 1993 ”Contesting Terrain in Zimbabwe’s Eastern Highlands” in U]
Economic Geography, 69(4), p.380—402

Moran, M., et.al. 1996. Opportunities and limitations for image-based remote
sensing in precision crop management. Remote Sensing of Environment .

61: 319-346. v_

 

Natrass, J. 1986. The Anatomy of Rural Black Poverty Durham University of Natal-
Durban

Neocosmos, M. 1993. The Agrarian Question in Southern Africa and Accumulation
from Below. Uppsala: Nordiska Afrikainstitutet

Nesman, E. 1981. Peasant Mobilization and Rural Development. Cambridge, MA:
Schenkman

Ngqaleni, M. and Makhura, M. 1996. " An Analysis of Women's Status in
Agricultural Development in the Northern Province” in Land, labour,
and Livelihoods in Rural South Africa. eds. Iipton, et al. Durban:
Indicator Press.
Ngqaleni, M. 1996. The Role of Women in the reconstruction of Agriculture in
Developing Areas . University of the North Technical Paper.
Ngqangweni, S. and van Rooyen, C. 1998. ” Farmworker Participation Schemes
as a Mechanism for Agricultural Transformation” in Agricultural land
Reform in South Africa: Policies, Markets, and Mechanisms” eds. Van Zyl,
et.al. Cape Town: Oxford University Press.

Norgaard, R. 1989. ”The Case for Methodological Pluralism”. in Ecological
Economics (1) p.37-57

Oberai, A. 1988. land Settlement Policies and Population Distribution in Developing
Countries: Achievements, Problems, and Prospects. New York: Praeger

Ojima, D., et al. 1994 ”The Global Impact of Land-use Change” in BioScience 44(5)
p.300-304.

Olomola,A. and Nwoso,A. 1993. Perspectives on Food Security in Nigeria. Lagos:
Nigerian Rural Sociological Association

Olson, J. 1990. The Impact of Oranging SocioEconomic Factors on Migration Patterns
in Rwanda. East Lansing : Michigan State University.

Oyedipe, F. 1983. Adjustment to Resettlement: A Study of the Resettled Peoples in the
Kainji lake Basin. Ibadan: University Press limited.

251

 

 

 

 

Pankhurst, A. 1992. Resettlement and Famine in Ethiopia: The Villager’s Experience.
Manchester: Manchester University Press.

Patnaik, U. 1991. “Food Availability and Famine: A Longer View.” In The Journal
of Peasant Studies. 19: 1-25

Pax-Lenny, M. and Woodcock, C. 1997. ”Monitoring Agricultural Lands in Egypt
with Multi-temporal Landsat TM Imagery: How Many Images are
Needed? ” in Remote Sensing of the Environment 59: 522-529

Fax-Lenny, M. and Woodcock, C. 1997. ”The Effect of Spatial Resolution on the
Ability to Monitor the Status of Agricultural Lands” in Remote Sensing
of the Environment 61: 210-220

Fax-Lenny, M. et.al.1997. ”The Status of Agricultural Lands in Egypt: The Use of
Multi-temporal NDVI Features Derived from Landsat TM ” in Remote
Sensing of the Environment 56: 8-20

Peet, R. and Watts, M. 1996. Liberation Ecologies: Environment, Development, Social
Movements. London: Routledge

Peters, A. and May, J. 1984. A Study of Income and Expenditure in Rural KwaZulu -
Mapumulo. Ulundi: KwaZulu Finance and Investment Corporation
limited

Pickles, J. 1995. Ground Truth: The Social Implications of Geographic Information
Systems. New York: Guilford Press

Pinckney,T. and Valdes,A. 1988. ”Short Run Supply Management and Food
Security: Results From Pakistan and Kenya." In World Development 16:
1025-1034

Pottier, J. 1985. Food Systems in Central and Southern Aflica. London: School of
Oriental and African Studies

Pottier,J. 19 ”The Politics of Famine Prevention: Ecology, Regional Production,
and Food Complementary in Western Rwanda.” In Afiican Affairs

Prah,K. 1987. Food Security Issues in Southern Africa. Maseru: Institute of Southern
African Studies.

Ragsdale, T. and Ali, A. 1988. ”A Case Study of the Resettlement of Nomads in
Somalia” in Land Settlement Policies and Population Redistribution in
Developing Countries. Ed. Oberai, A. New York Praeger.

Rahmato, D. 19 ”Peasant Survival Strategies in Ethiopia.” In Disasters 12: 326-
344

Razavi,S. and Miller,C. 1995. ”From WID to GAD: Conceptual Shifts in the
Women and Development Discourse” UN Research Institute for Social
Development.

Ranger, T. 1983. "The Invention of Tradition in Colonial Africa.” In The Invention
of Tradition. Eds. Hobsbawn and Ranger. London: Cambridge
University Press

Reardon, T. 1997. ”Using Evidence of Household Income Diversification to
Inform Study of the Rural N on-farm Labor Market in Africa,” World
Development, 25 (5), 735-748.

252

 

Reardon, T.et al. 1988. ”Coping with Household Level Food Insecurity in Drought
Affected Areas of Burkina Faso." In World Development. 16: 1065-1074

Reid, R., et.al. 2000. ”Land-use and Land-cover Dynamics in Response to
Changes in Climatic, Biological, and Socio-political Forces: the Case of
Southwestern Ethiopia” in landscape Ecology 15: 339-355.

Rikuni, M. and Wyckoff, J. 1991. Market Reforms, Research Policies and SADCC Food
Security. Harare: University of Zimbabwe

Rikuni, M. et al .1989. Food Security Policies in the SADCC Region. Harare:
University of Zimbabwe.

Rikuni,M. and Bernsten, R. 1987. Southern Africa: Food Security Policy Options.
Harare: University of Zimbabwe

Rikuni,M. and Eicher, C. 1987. Food Security for Southern Afiica. Harare:
University of Zimbabwe

Robertson, C. and Berger,I. 1986. Women and Class in Africa. New York: Africana
Press.

Rocheleau, D. et al. 1996. Feminist Political Ecology: Global Issues and Local
Experiences. London: Routledge

Rockwell, R. 1994. ”Culture and Cultural Change” in Oranges in land Use and
land Cover: A Global Perspective. Eds. Meyer, W. and Turner, B.
Cambridge: Cambridge University Press.

Roux, A. and Gilmour, D. 1991. Economic Conditions among Blacks in Rural Ciskei
Grahamstown: Rhodes University

Sage, C. 1994. ”Population and Income” in Oranges in Land Use and Land Cover: A
Global Perspective. Eds. Meyer, W. and Turner, B. Cambridge:
Cambridge University Press.

Salih, M. 1994. Inducing Food Insecurity: Perspectives on Food Policies in Eastern and
Southern Africa. Stockholm: Gothab

Sanderson, S. 1994 ”Political-Economic Institutions” in Changes in land Use and
land Cover: A Global Perspective. Eds. Meyer, W. and Turner, B.
Cambridge: Cambridge University Press.

Sauer, C. 1925. "The Morphology of Landscape". in University of California
Publications in Geography. Berkeley: University of California Press.

Scatena, F., et.al. 1996. ”Cropping and Fallowing Sequences of Small Farms in the
Terra Firme Landscape of the Brazilian Amazon: A Case Study from
Santarem, Para” in Ecological Economirs 18: 29-40.

Sen, A. 1981. Poverty and Famines. Oxford: Clarendon Press.

Sibanda, S. 1999. ”Proposals for the Management of Land rights in Rural South
Africa” Paper presented at the Conference ”land And Agrarian Reform in
South Afiica: Successes, Problems, and the Way Forward” July 26-28, 1999.
Pretoria.

Showers, K. 1994. ”Historical Environmental Impact Assessment: Connecting
Land Use History to Environmental Impact Assessment” Working
Paper No. 174, African Studies Center, Boston University

253

 

Skole, D, et al. 1994 " Physical and Human Dimensions of Deforestation in
Amazonia" in BioScience 44(5) p. 314-322.

Smith, D. 1990. Apartheid in South Afiica. Cambridge: Cambridge University
Press.

Snyder, K. 1996. " Agrarian Change and Land-Use Strategies Among Iraqw
Farmers in Northern Tanzania" in Human Ecology 24(3) p.315-340

South Africa, Republic of. 2000a. The Commercial Farmer Programme: A Component
of an Integrated land Redistribution Programme of the Ministry of
Agriculture and land Afiairs. Unpublished Government Green Paper.

South Africa, Republic of. 2000b. land Affairs General Amendment Act of 2000.
Pretoria: Government Printer.

South Africa, Republic of. 1999a. land Restitution and Reform Laws Amendment Act
of 1999. Pretoria: Government Printer.

South Africa, Republic of. 1999b. South Africa Yearbook 1999. Pretoria: GCIS. ;

South Africa, Republic of. 1998a. Extension of Security of Tenure Act (No. 62 of ‘J’
1997). Pretoria: Government Printer.

South Africa, Republic of. 1998b. Alienation of land Amendment Act (No.1538 of
1998). Pretoria: Government Printer.

South Africa, Republic of. 1998c. Agricultural Duos Rationalization Act of 1998.
Pretoria: Government Printer.

South Africa, Republic of. 1998d. Subdivision of Agricultural land Act Repeal Act of
1998. Pretoria: Government Printer.

South Africa, Republic of. 1998e. Land Aflairs General Amendment Act of 1998.
Pretoria: Government Printer.

South Africa, Republic of. 1998f. Prevention of Illegal Eviction From and Unlawful
Occupation of Land Act of 1998. Pretoria: Government Printer.

South Africa, Republic of. 1997a. Land Restitution and Reform laws Amendment
Act of 1997. Pretoria: Government Printer.

South Africa, Republic of. 1997b. Extension of Security of Tenure Act of 1997.
Pretoria: Government Printer.

South Africa, Republic of. 1996a. Communal Property Associations Act (No.849 of
1996). Pretoria : Government Printer.

South Africa, Republic of. 1996b. Interim Protection of Informal Land Rights Act (No
1057). Pretoria: Government Printer

South Africa, Republic of. 1996c. Upgrading of Land tenure rights Amendment Act
(No. 1099 of 1996). Pretoria: Government Printer.

South Africa, Republic of. 1996d. Development Facilitation Act (No.1412 of 1996).
Pretoria: Government Printer.

South Africa, Republic of. 1996e. land Restitution and Reform laws Amendment Act
(No. 1878 of 1996). Pretoria: Government Printer.

South Africa, Republic of. 1995a. land Reform (Labour Tenants) Bill (No. 446 of
1995). Pretoria: Government Printer

 

254

 

 

 

South Africa, Republic of. 1995b land Administration Act (No. 569 of 1995).
Pretoria: Government Printer.

South Africa, Republic of. 1995c. Restitution of Land Rights Amendment Act (No. 84
of1995). Pretoria: Government Printer.

South Africa, Republic of. 1994. Restitution of Land Rights Act (No. 2011 of 1994).
Pretoria: Government Printer.

South African Outlook, 1994. ”How to Beat the Drought." In South Afiican Outlook
?: 100-102

Spencer, D. 1976. Afiican women in Agricultural Development: A Case Study in
Sierra Leone. East Lansing: Michigan State University Press

Sperling, L. 1987. ”Food Acquisition during the African Drought of 1983-1984: A
Study of Kenyan Herders." In Disasters ?:263-272

Srivastava, S., and Panda, B. 1990. Evaluation of characteristic profile models for
crop discrimination. Remote Sensing for Agricultural Applications.
Proceedings of the National Symposium on Remote Sensing for Agricultural
Applications, December l6-8 1990, New Delhi.

Staatz, J. and Eicher, C. 1998. ” Agricultural Development Ideas in Historical
Perspective" in International Agricultural Development. Eds. Eicher, C.
and Staatz, J. Baltimore: John Hopkins University Press.

Staudt, K. 1987. ”Uncaptured or Unmotivated? Women and the Food Crisis in
Africa” in Rural Sociology 52(1).

Sussman, R., et.al. 1994. ”Satellite Imagery, Human Ecology, Anthropology, and
Deforestation in Madagascar” in Human Ecology 22(3) p.333-354

Swift, J. 1989. "Why are Rural People Vulnerable to Famines? " in Institute for
Development Studies Bulletin 20(3) p.8-15

Tapson, D. 1988. " Freehold vs. Leasehold in the Homelands". in Towards
Freehold? : options for land and Development in South Afiica's Black Rural
Areas. eds. Cross, C. and Haines, R. Cape Town: Juta and Co.

Taylor, J. 1992. Food Science in the New South Africa : Fuelling the Economy and
Feeding the Nation. Pretoria: The University of Pretoria.

Teklu, T. 1994. "The Prevention and Mitigation of Famine: Policy Lessons from
Botswana and Sudan." In Disasters 18:35-47

Thenkabail,P. and Nolte,C. 1996. ”Capabilities of Landsat 5 Thematic Mapper
(TM) data in regional mapping and Characterization of Inland Valley
Agra-ecosystems in West Africa” in International Journal of Remote
Sensing 17(8) p.1505-1538

Thimm, Het al. 1993 Regional Food Security and Rural Infrastructure. Hamburg: Lit
Verlag.

Thobane, T. 1997. ”Delivery of Agricultural Extension Services to Small-Scale
Farmers” in Rural Land Reform Issues in Southern Africa: lessons for
South Africa’s Northern Province. Eds. de Villiers, A. and Critchley, J.
Sovenga, SA: University of the North Press.

255

 

Turner, B., et al. 1990. The Earth as Transformed by Human Action. Cambridge:
Cambridge University Press.

Tyson,P. 1987. Climatic Orange and Variability in Southern Africa. Cape Town:
Oxford University Press

Van den Brink, R., et.al. 1996. ” Rural Livelihoods, Fiscal Costs, and Financing
Options: A First Attempt at Quantifying the Implications of
Redistributive Land Reform" in Agricultural Land Reform in South
Africa: Policies, Markets, and Mechanisms” eds. Van Zyl, et.al. Vcape
Town: Oxford University Press. .

Van Rooyen, J. 1998. ”Agricultural Policy Issues and Reform for Growth with
Equity in South Africa” in Agricultural Land Reform in South Africa:
Policies, Markets, and Mechanisms” eds. Van Zyl, et.al. Cape Town:
Oxford University Press.

Van Rooyen, J. and Njobe-Mbuli, B. 1996. ”Access to Land: Selecting the
Beneficiaries” in Agricultural land Reform in South Africa: Policies,
Markets, and Mechanisms” eds. Van Zyl, et.al. Cape Town: Oxford
University Press.

Van Zyl, J. 1996. ”The Farm Size-Efficiency Relationship" in Agricultural land
Reform in South Africa: Policies, Markets, and Mechanisms. Cape Town:
Oxford.

Van Zyl, J. 1994. "The Last Straw: Drought and the Economy." In South Afiican
Outlook

Van Zyl, J. 1991. "The Effects of Increased Earnings from Traditional Agriculture
in Lebowa" in Agrekon 30(4) p.276-278.

Vogel, C. 1994. "A Prayer for Rain in Boputhatswana.” In South African Outlook.

Vogel, C. 1993 ” Dimensions of Drought : South African Case Studies” in
GeoJournal. 39: 51-58.

Voh, J. 1980. Resettlement Adjustment Patterns to Rural Development Programs: The
Case of Tiga Dam in Kano State Of Nigeria. Ann Arbor: UMI.

Walker, P. 1989. Famine Early Warning Systems: Victims and Destitution. London:
Earthscan Publications

Walker, R. 1996. ”Land Use and Land Cover Dynamics in The Brazilian Amazon:
An Overview” in Ecological Economics 18: 67-80.

Watts,M. 1983. Silent Violence: Food, Famine and Peasantry in Northern
Nigeria. Berkeley: University of California Press

Webb,P.,et al. 1992. Famine in Ethiopia: Policy Implications of Coping Failure at
National and Household Levels. Washington: International Food Policy
Research Institute

Webb,P. 1993. "Coping with Drought and Food Insecurity in Ethiopia." In
Disasters 17: 34-47

Wilkie, D. 1994. ”Remote Sensing Imagery for Resource Inventories in Central
Africa: The Importance of Detailed Field Data” in Human Ecology 22(3)
p.379-403.

~

7: "

256

Wuyts, M. 1994. "Accumulation, Industrialization and the Peasantry: A
Reinterpretation of the Tanzanian Experience." In The Journal of Peasant
Studies. 21: 159-193

Young, H. and Jaspars,S. 1995. "Nutritional Assessments , Food Security, and
Famine." In Disasters 19:26-36

Zimmerer, K. 1994. ”Human Geography and the 'New Ecology' : The Prospect
and Promise of Integration". in Annals of the Association of American
Geographers 84(1) p.108-125.

257