Livelihood trade-offs in the commercialisation of multiple-use NTFP: Lessons from marula (Sclerocarya birrea subsp. caffra) in southern Africa

1 1,Sheona Shackleton , Charlie Shackleton *, Rachel 2 3 4Wynberg , Caroline Sullivan , Roger Leakey , Myles

5 5 6 6Mander , Tania McHardy , Saskia den Adel , Andy Botelle , 6 6 7Pierre du Plessis , Cyril Lombard , Sarah Laird , Tony

7 3Cunningham and Dermot O’Regan1Department of Environmental Science, Rhodes University, Grahamstown, South Africa2Environmental Evaluation Unit, University of Cape Town, Cape Town, South Africa3Centre for Ecology and Hydrology, Wallingford, United Kingdom4Agroforestry and Novel Crops Unit, James Cook University, Cairns, Australia5Institute of Natural Resources, University of KwaZulu-Natal, Pietermaritzburg, South Africa6CRIAA-SADC, Windhoek, Namibia7People and Plants International, Inc., P.O. Box 73, Essex Junction, VT0592, USA*E-mail: c.shackleton@ru.ac.za

Abstract: Commercialisation of non-timber forest products (NTFP), apart from

the multitude of benefits, is often associated with trade-offs in terms of traditional

and cultural livelihoods. This paper presents a holistic assessment of livelihood

trade-offs involved in commercialisation of marula (Sclerocarya birrea subsp.

caffra), a multiple-use NTFP species in southern Africa. The study was conducted

at two sites in South Africa (Bushbuckridge district and Ubombo district) and one

in Namibia (former Ovamboland). Some of the key features of the study include

the household use and trade in marula products, the biological aspects of the

marula resources, marketing and trade of the species and policies associated with

its utilisation. The paper also highlights the important and diverse role that marula

has in local livelihoods and in contributing to the forms of livelihood capital like

human, social, financial, natural and physical capital. Likely trade-offs in terms of

livelihoods with increasing commercialisation of marula are discussed, along with

potential threats and opportunities from commercialisation.

Keywords: ommercialisation, livelihoods, marula, trade-offs.

—

—C

11

Introduction

Many non-timber forest product (NTFP) species have multiple uses and values

(e.g. Salick et al. 1999; Luoga et al. 2000), often forming an integral component of

rural livelihoods and culture (Laird 1999). The marula tree (Sclerocarya birrea

subsp. caffra) is one such species. Widely distributed throughout the dry savannas

of southern and central Africa, it is a frequently dominant species, maintained in

the landscape even after clearance for settlements or agricultural lands. Nearly all

parts of the tree are used for one or more purposes, including medicinal (bark and

leaves), nutritional (raw fruit, kernels), social (local beer/wine and an

internationally traded liqueur), culinary (kernel oil for cooking, roasted kernels

added to food), aesthetic (kernel oil for cosmetics, wood for carving and utensils),

energy (wood for fires), protective (truncheons planted as live fences), and

spiritual and cultural (various local traditions throughout most of its range) (Hall

et al. 2002; Shackleton et al. 2002a; Wynberg et al. 2002b).

140 Sheona Shackleton et al.

Box 1. Marula: a unique multiple use resource—

Sclerocarya birrea subspecies caffra (A. Rich.) Hochst. (marula) is a member of the large Anacardiaceae

family, along with other well-known edible species such as the mango, cashew nut and pistachio. It has a

widespread distribution from tropical east Africa southwards to approximately 31°S in the KwaZulu-

Natal province of South Africa (Peters 1988), and a closely related species, S. birrea subsp. birrea occurs in

Sahelian west Africa. It is a large (8–17 m tall) deciduous tree, and is often a community dominant. It is

dioecious, with only the female trees bearing fruits. Marula occupies a range of habitats from open

savanna to semi-deciduous forest. It has been recorded on a wide variety of soil types, but most

commonly on well-drained soils. Mean annual rainfall regimes from where it has been recorded range

from 200 mm to 1500 mm. It is assumed that its distribution is limited by frost severity. Fruits develop

during December and fall to the ground in January and February where they ripen to a soft yellow colour.

For many thousands of years marula has formed an integral part of the livelihoods, culture and

spirituality of rural households across southern Africa (see Shackleton et al. 2002a for a review). Its

uniqueness lies in the fact that there are so many uses of the tree and its fruits. From the shade of the tree to

its versatile fruit and the burning of the empty nuts for fuel, people are using all of its products, and the

importance of marula stretches from the social, to the cultural, economic, and nutritional aspects of

people's lives. Indeed, there are few wild species that demonstrate such a wide range of uses, or such a

significant position in local culture (Shackleton et al. 2002a). Marula provides vitamin rich fruits (higher in

vitamin C than citrus) that can be eaten whole, made into juice, pulp or jam or brewed into beer; the

kernels form an important food supplement and the oil extracted from them (which has very stable

properties – Burger et al. 1987) has many uses from culinary to cosmetic; the bark and leaves have

medicinal properties; and the wood is used for a variety of purposes including carving and fuelwood. The

tree is also host to a range of edible caterpillars and larvae as well as parasitic mistletoes which produce

outgrowths known as wood roses which are sold in the curio market (Shackleton et al. 2000a; Hall et al.

2002).

Such multi-use species typically present more challenges, opportunities and

constraints under a commercialisation scenario than do NTFP that have more

limited uses. For example, one constraint may be that it is more difficult to

implement and manage sustainable harvesting because a range of different and

possibly competing stakeholders involved; some may require use of the whole

tree (such as woodcarvers) whilst others harvest only a portion (such as fruit

collectors). An opportunity is that multiple-use species are frequently valued and

readily appreciated as socially and economically important and are thus protected

from broader landscape impacts such as land clearing. The complexity intensifies

if the range of uses includes cultural or spiritual dimensions that have, typically,

evolved over millennia and become incorporated into the social fabric and

traditions of rural, subsistence users.

Given the versatility and unique properties of marula, the last two decades have

seen a number of local and international markets established for products derived

from this species; the most well known being 'Amarula' liqueur, produced by

Distell Corporation, which is exported from South Africa to 28 countries around

the world. As with all NTFP commercialisation initiatives, these represent an

opportunity for rural development and social upliftment. However, these new

developments should not occur at the expense of age-old traditional uses of the

species, indigenous rights and resource stocks. Such aspects are seldom afforded

adequate consideration in commercialisation efforts. Ensuring a balance between

the different uses of a multi-use species and its diverse contributions to livelihood

assets is a challenge that varies between species, as well as temporally and

spatially for a single species. Interest and trade in marula products have

accelerated over the last few years, both in terms of local-level trade between rural

communities and adjacent urban centres, as well as international markets via new,

capital-intensive projects (such as canned drinks, juices, cosmetic oils, biscuits,

chocolate and yoghurt). It is, thus, timely to reflect on what this growing

commoditisation might mean for rural livelihoods and the marula resource, as

well as to consider ways in which the trade could be promoted while minimising

any negative impacts on existing uses and benefits.

Because of its many uses and increasing commercialisation, marula is a well-

known species throughout the sub-continent. Yet, until recently there was

surprisingly little quantitative data or information on its abundance,

opportunities for sustainable harvest and its role as an asset in rural livelihoods.

This dearth was addressed to some extent by the publication of a monograph

summarising knowledge and literature on marula (Hall et al. 2002), as well as a

multidisciplinary and multi-institutional project executed at three sites in South

Africa and Namibia (all reports available at http://www.ceh-wallingford.

ac.uk/research/winners/literature.html). This project resulted in a number of

reports and publications on various facets of marula, its local use, and

Commercialisation of multiple-use NTFP 141

commercialisation pathways (see below). Yet there has been no publication that

fully integrates these various facets to evaluate the role of marula in the asset base

of rural households and communities and the trade-offs made at local and other

levels in the commercialisation process. This chapter seeks to do this as a case

model for similar multiple-use NTFPs elsewhere in the world. Specifically this

chapter will (i) summarise the key empirical results of the marula project to lay the

foundation for subsequent integration and interrogation, (ii) present and examine

different commercialisation pathways, (iii) examine the trade-offs involved in use

and commercialisation options, and (iv) interpret these findings in a context of

promoting sustainable livelihoods of rural communities.

Background to the marula project

This chapter synthesises work by several researchers from different disciplines,

and represents a holistic assessment of our current state of knowledge about

marula in southern Africa. The research was conducted at three separate sites in

South Africa and Namibia. These sites were identified as areas where marula was

abundant and widely used, where a large proportion of the local population lived

at or below the poverty line, and where some commercialisation was taking place

or anticipated (Shackleton et al. 2002b). These areas have some geographical

variability, and through the selection of these sites, we hoped to demonstrate the

generic applicability of the findings to other areas where marula is found. The

specific areas in which research was undertaken were:

• the Bushbuckridge district, Limpopo Province, South Africa (mid-point is

31 17´E; 24 45'S) (referred to as Bushbuckridge from here on);

• Ophande ward on the Makhatini flats, Ubombo district, northern KwaZulu-

Natal, South Africa (mid-point is 32°05'E; 27°15'S) (referred to as Makhatini

from here on); and

• former Ovamboland, currently part of the Oshana, Oshikoto and Ohangwena

districts, of North Central Namibia (mid-point is 16°40'E; 17°55'S) (referred to

as Namibia from here on).

At each of these sites, detailed resource inventories were undertaken under

different land-use and tenure systems (Botelle et al. 2002; McHardy 2002a;

Shackleton et al. 2003); fruit yields, mass, kernel size, skin thickness, etc. were all

examined (Leakey 2005; Leakey et al. 2005a,b). Household surveys, group

discussions and individual interviews with key informants yielded information

on the range of uses and social, cultural and economic importance of this species

(den Adel 2002; McHardy 2002b; Shackleton and Shackleton 2005). Market-chain

analysis of five different pathways of marula commercialisation provided insights

into how fruits and other marula products are traded, and how the economic

benefits from this are distributed (du Plessis et al. 2002; Mander et al. 2002;

Shackleton 2004). In an attempt to understand more about governance, tenure and

° °

142 Sheona Shackleton et al.

property rights associated with marula, we investigated the institutional

arrangements surrounding the use of this species, including the range of

customary and statutory regulations that control its use (Wynberg et al. 2002b). We

also considered the gender implications of marula use, and were able to assess the

way in which this resource contributes to a number of livelihood assets

(Shackleton et al. 2002b). As far as possible, the same approaches were applied at

all sites, with research methods and interview schedules being devised through

participation of the whole research team, and piloted at the study sites. These

comprehensive studies provided the empirical basis for more integrative analysis

of the impacts and trade-offs associated with commercialising a multiple-use

resource such as marula. At the time of the study in 2001 the monetary exchange

rate was US$1.00 = Rand 10.00, which is used throughout this chapter. At the time

of writing, the exchange rate was US$1.00 = approximately Rand 6.50.

Household use and trade in marula products

Socio-economic profile of households and marula consumption

In both South Africa and Namibia, marula plays a key role in the livelihoods and

culture of local people, many of whom are desperately poor, living well below the

2001 poverty line of about US$ 100 per month for a household of four (den Adel

2002; McHardy 2002b; Shackleton and Shackleton 2005). In all three study sites,

households demonstrated multiple and diverse livelihood strategies relying on

cash and 'in kind' income from a range of sources including agriculture,

remittances, local wage employment, petty businesses, wild resources, state

welfare grants, and in particular old age pensions (Table 1). Average monthly cash

incomes ranged from Rand 448 (US$ 45) to Rand 925 (US$ 93) per household across

sites, with as high as 25–48% of households earning less than Rand 250 per month

(den Adel 2002; McHardy 2002; Shackleton and Shackleton 2005). All the regions

were prone to frequent drought (up to every three years) rendering dryland

agriculture a risky endeavour, whilst land shortage was a major concern in

Bushbuckridge. In all cases a significant proportion of households had no reliable

source of cash income at all (Table 1). Consequently, the majority of households

relied to some extent on natural resources, including marula, to supplement

incomes and as a safety net in times of need.

Extensive use of a wide range of marula products was made, especially the fresh

fruit for the production of the extremely popular beer/wine* (hereafter referred to

as beer) and kernels (Table 2). In Namibia all households interviewed (100%)

brewed beer, whilst in South Africa between 70% and 75% of households

Commercialisation of multiple-use NTFP 143

*Beer and wine are basically the same product, i.e. fermented marula juice, with the term beer commonly used

in South Africa and the term wine used in Namibia. It has been said that fermented marula juice probably

more closely resembles cider than either beer or wine.

Table 1. Socio-economic profile of households (hh) surveyed.

Proportion

of hh with Proportion

Proportion Proportion Proportion at least one of hh with

hh Average of hh earning of hh with of hh with employed regular cash

Site N size hh income <R250/mth fields cattle member income

Namibia 60 11.9 N$799 48% 100% 74% 62% 12%

Bushbuckridge 142 7.2 R925 25% 50% 26% 66% 15%

Makhatini 60 7.5 R448 – 78% 47% 49% 38%

The Nambian dollar is equivalent in value to the South African Rand.

Table 2. Production of the two most popular marula products.

Average amounts produced

% of households (litres per producer household per season);

Product Site N producing/using range in parentheses

Beer/wine* Namibia 60 100 195 (5–1200)

Bushbuckridge 142 75 311 (10–3360)

Makhatini 60 71 253 (25–400)

#Kernels Namibia 60 100 72 (2–250)

Bushbuckridge 142 88 10 (1–100)

Makhatini 60 81 12 (3–84)

*Wine in Namibia is an undiluted, fermented marula drink, whereas South African marula beer is generally diluted

with water.#Most kernels in Namibia are used for the production of a traditional cooking oil, whereas in South Africa they are

eaten whole.

—

—

produced this alcoholic beverage. Kernel consumption was also high in all sites

(81–98% of households) with about 60% of households mixing kernels with wild

leafy vegetables and other foods in South Africa, whilst all households did this in

Namibia. The average quantities of fruit used (over one tonne per household per

season) and beer produced (average between 150 and 350 l per household per

season) were similar in all three sites, but households in Namibia were using up to

ten times the amount of kernels (about 70 l per household per season) relative to

South Africa (den Adel 2002; Shackleton and Shackleton 2005; Table 2).

Namibian households made use of a range of products either infrequently or

never encountered at either of the South African sites (den Adel 2002; Shackleton et

al. 2002b). These included a non-alcoholic juice (100%), a porridge made from the

fruit pulp following juice extraction (78%), kernel oil for cooking (100%) and

cosmetics (28%), and a cake-like residue that remains after oil extraction which is

either eaten (100%) or fed to livestock (68%; Table 3). Jam, a non-traditional

product, was only popular in Bushbuckridge (made by 48% of households) where

it had been promoted by several external agencies and government health

144 Sheona Shackleton et al.

personnel via 'care giver' groups. Producer households in this area were making

approximately five litres of jam each season mainly for home use, with four

producers selling this product on a regular basis, earning Rand 50 (US$ 5) per

household per year. All collection and processing of marula fruit was undertaken

by women, with this being seen to fall clearly within their domain.

In addition to fruit-based products, dry wood and the empty nuts are used for

fuelwood (60–100%), whilst bark (mainly South Africa, 38–80% of households)

and leaves (mainly Namibia, 50% of households) as a source of medicine (Table 2).

Edible emperor moth caterpillars, abundantly found in marula trees early in the

season, were consumed by about one third to one half of households across all

sites, while the beetle larvae found in the rotting wood were less well known and

eaten only by about 5% or less of households in both countries. Additional uses

Commercialisation of multiple-use NTFP 145

Table 3. Use of the full range of marula products by households (% of households) in the different study sites.—

% of households using or making

Product Uses Bushbuckridge Makhatini Namibia

Fruit Eating 94 57 97

Beer 75 71 100

Juice 13 0 100

Jam 31 0 3

Other (fodder, sell, manure) 16 2 –

Kernels Whole 88 81 98

Oil (cooking) 0.7 0 100

Oil (cosmetic) 1.4 (obtained from MDC) 2 28

Add to food 61 68 100

Eat ‘cake’ (residue after oil extraction) 0 0 100

Cake as animal feed 0 0 68

Wood Carving 5 8 0

Utensils 10 11 52

Firewood 94 60 97

Other (cattle yoke, fencing, furniture) 7 0 78 (poles)

Medicine Bark 39 80 2

Roots 8 14 5

Leaves 4 24 52

Moth caterpillars 54 39 30

Beetle larvae 2 14 3

Game pieces 60 0 73

Other (e.g. burn shells, veterinary use, leaf skirts, fodder, rattles 15 4 100

and necklaces, fencing, diviners dice) (burn shells)

MDC, Mhala Development Centre.

that were identified included the use of fruit and skins as fodder for goats and pigs,

leaves as fodder and as a veterinary medicine for goats, poles in live and dead

fencing, leaves to make traditional 'play' skirts, nuts for rattles, beads, necklaces

and game pieces, hair relaxant and lastly the use of a nut as a diviners die. A

number of respondents in Bushbuckridge also listed 'selling' as an additional use

and collected marula primarily for this purpose. Marula is also the preferred host

of several mistletoe species, of which Erianthemum dregei and Pedistylis galpinii are

harvested for the flower-shaped callous formations (locally called wood-roses)

they induce in the host tissue which are sold in local markets (Dzerefos et al. 1999,

2003).

Household trade

In Bushbuckridge and Namibia, a relatively high proportion of households (Table

4) were selling marula products both in local markets and to commercial

producers, although the numbers varied widely both across study sites and

between different villages or areas within study sites (Table 3). Fruit, kernels and

beer were sold in Bushbuckridge, whilst kernels and beer formed the primary

traded products in Namibia. Because of the seasonality of the resource, incomes

earned were, on the whole, modest, varying between Rand 100 and Rand 1000 per

household per year (den Adel 2002; Shackleton and Shackleton 2005). However,

because of high levels of poverty and the timing of the fruiting season, this income

was critical in paying school fees and other schooling costs, and in purchasing

food. By contrast, only two households were selling marula products in

Makhatini, where this activity was generally not supported for cultural reasons by

the community leadership (McHardy 2002b). These households had no other

source of cash income and were selling out of pure economic necessity.

146 Sheona Shackleton et al.

Box 2. Economic value of domestic consumption of marula in Bushbuckridge

—

Not only is the economic value of traded marula products important, but so too is the value of those

products consumed directly by households. The direct use value of household consumption of marula in

Bushbuckridge was calculated by multiplying the average amounts used or produced of each product per

household by the 'farm gate' price. Values are calculated across all 142 households and represent gross

value per household per annum. They exclude the tremendously important cultural and spiritual values

imparted by communities on marula, as well as the indirect values derived from maintaining marula in

the ecosystem.

Gross value per household per annum: Fresh fruit: Rand 314; Beer: Rand 567; Jam: Rand 16; Kernels:

Rand 184.

Total gross annual direct use value of marula in Bushbuckridge (based on fruit): Rand 18 million (US$ 1.8

million).

Social and cultural significance of marula

In addition to its importance for household food security and cash, marula has had

considerable social, spiritual and cultural value for generations (Shackleton et al.

2002a). Of particular significance in all three sites were the neighbourhood marula

'gatherings' where the freshly brewed beer was drunk. These were important in

building social networks and reciprocal relations, and in cementing existing bonds

including with urban relatives who often return to the family home to join in the

festivities. Other socio-cultural activities varied between sites and are discussed

further in the section on social capital. Table 5 indicates the proportion of

households in four villages in Bushbuckridge that participated in a variety of

traditional cultural practices and ceremonies associated with marula.

In South Africa, almost all households using marula harvested the fruit from

communal lands. By contrast, in Namibia, almost all marula trees were 'privately'

tenured (via a usufruct system) and found in people's fields. Consequently, in

Namibia, a number of informal institutions have evolved to ensure equity and

sharing of the marula resource. Beer is brewed under the marula trees in owners'

fields, with this being social event in which other women partake and provide

their labour in order to receive a share of the benefits in the form of the nuts and

beer that they take home. In South Africa, women generally make the beer at

home, sometimes with help from family members. Of significance, is that in South

Africa anyone can access marula and use it at their own discretion, whereas in

Namibia about one quarter of villagers did not have marula trees and depended

on the goodwill of friends and neighbours for marula products.

The marula resource and fruit characterisation

There were remarkable similarities in the

state of the resource between the different

study sites. Absolute densities of mature

female trees were consistent, with

approximately 3–6 trees/ha (Botelle et al.

2002; Shackleton et al. 2003). The ratio of

males to females decreased with increasing

human impact (Table 6), probably as a

Commercialisation of multiple-use NTFP 147

Table 4. Percentage of households from the household

surveys trading marula products.

Beer/wine Fruit Kernels

Namibia – 4 villages 38% 0 62%

Bushbuckridge – 4 villages 14% 29% 53%

Makhatini – 1 village 1 household 0 1 household

—

Table 5. Marula in traditional ceremonies in Bushbuckridge (source: Shackleton and Shackleton 2005).

Allandale Edinburgh Hokwe Rolle All

% of households practicing or knowing others that practice 80 92 81 83 84

one or more traditional ceremonies

Xikuha (presenting beer to the headman) (% of households) 66 78 58 56 64

Kupahla (an offering of beer to the ancestors) (% of households) 37 36 28 36 32

Xirhwalo (festivities where beer is shared) (% of households) 81 81 82 79 82

—

result of selective cutting of males, although, also

possibly due to increased nurturing and maintenance

of female trees (Shackleton et al. 2003). Overall, size

class profiles across the study sites generally

demonstrated a skewed distribution in favour of large

adult trees in peoples' fields/homesteads, with low

proportions of juveniles, indicating low recruitment

(Shackleton et al. 2003). The opposite applied in

nearby protected areas or inaccessible sites. Across all

study areas, trees in homesteads/fields produced

significantly more and larger fruits than communal

land/protected area trees, although fruit yields were

variable between trees and land uses (Table 7)

(Shackleton et al. 2003). There was a positive relation-

ship between fruit numbers per tree and tree

diameter/crown volume, indicating that larger trees produce greater numbers of

fruit. Lack of a relationship between fruit numbers and fruit size implies that fruit

size is genetically controlled and thus potentially could be improved by

domestication. Noteworthy differences between sites were that, on average,

Namibian trees have bigger fruits (± 20%) than those from South Africa, due to a

greater pulp mass (Leakey et al. 2005a).

Based on 2002 yield data, the current density of adult female trees is sufficient to

meet existing fruit demand for household use and commercialisation (Shackleton

et al. 2003). Thus, the limited natural recruitment is currently not a constraint to

existing demand. Furthermore, the level of nurturing and/or planting in all sites

would appear to be adequate to maintain the current density of producing female

trees (Table 8).

The apparent selective removal of male trees by households is probably resulting

in insufficient numbers of male trees for maximum pollination success. This is

especially pertinent since the loss of male trees coincides with increasing

pollination demand due to the observed higher flower/fruit load of people’s trees.

This results in variable within-tree kernel numbers per nut (0–4), and undesirably

low overall mean kernel mass per nut (Leakey et al. 2005b). Pollinator populations

may also be in decline due to increasing human impacts (deforestation and

removal of dead trees leading to a loss of hive sites), coupled with the lower

density of male trees. Further studies are however needed to determine the

prospects of enhancing pollination through the use of beehives.

Across all sites and land uses there was large and continuous tree-to-tree variation

in fresh mass between all fruit components (skin, flesh/juice, nut, shell and

kernels) and in oil content (Table 9) (Leakey et al. 2005a, b). It was found that large

148 Sheona Shackleton et al.

Table 6. Adult gender ratio in marula trees

under different land-uses in Bushbuckridge.

Homesteads 1 male : 1.39 females

Fields 1 male : 0.73 females

Rangelands 1 male : 0.52 females

Protected areas 1 male : 0.50 females

Table 7. Marula fruit yield under different land-

uses in Bushbuckridge.

Protected Homesteads/

areas fields

No. of fruits per tree < 3 500 > 17 000

Mean fruit mass 20.9 g 24.9 g

—

—

fruits do not necessarily have large kernels.

There was strong evidence of the potential for

selection and cultivar development at all sites.

To guide multiple trait selection, fruit and

kernel ‘ideotypes’ have been identified

(Leakey 2005). One tree in Namibia (’Namibian

Wonder’) particularly illustrates the domesti-

cation potential of marula (Table 9). The

consistent mass of individual kernels per nut

suggests that selection for nuts with four

kernels is a better strategy than selection for a

nut with a single kernel. Evidence that farmers

are already starting to domesticate marula

comes from some degree of selection in trees

planted by households in terms of higher fruit

yields in close proximity to homesteads/fields,

frequency distribution of fruit and kernel

mass, and higher mass of traits in fields. This

selection is continued into the market place

where the worst fruits (15–20%) are discarded.

The commercial markets and market analysis

Economic value and benefits from the commercial tradeThere were five commercial market channels or subsectors in the marula industry

when the study was conducted (see Figure 1 and du Plessis et al. 2002; Mander et al.

2002; Shackleton 2004) with the first four being the most important:

• Liqueur: Amarula cream is produced by Distell Pvt Ltd (a large South African

liquor-producing corporation) and marketed nationally and internationally. The

marula pulping factory in Phalaborwa is owned by Distell, Mirma and a

Community Trust, with the distillation factory owned by Distell in Stellenbosch.

• Juice: Marula juice produced at the Mhala Development Centre (MDC) in

Thulamahashe, under the Mine Workers' Development Agency Marula Project, is

attempting to source markets for juice. The MDC was initiated as a job creation

programme in an area of high rural unemployment. Since the initial study, this

project has been established as an independent enterprise known as 'Marula

Natural Products' (Pvt Ltd). Its main purpose and objectives, i.e. to create jobs and

income for rural producers, have not changed.

• Oil: Marula oil is produced by Eudafano Women's Cooperative in Namibia

(facilitated by an NGO-CRIAA SA-DC) and by the Mhala Development Centre in

South Africa. Both the Namibian and South African channels are trading to The

Commercialisation of multiple-use NTFP 149

Table 8. Proportion of households planting and

nurturing of marula trees within homesteads and fields.

% planting % nurturing

Bushbuckridge 33 53

Makhatini 2 19

Namibia 22 35

Table 9. Range in mean fresh mass in marula fruits and

nuts.

Mass/%

Fruit part All sites Namibian wonder fruit

Fruit mass 10.2–41.8 g 69.9 g

Pulp mass 7.2–36.0 g 59.1 g

Skin mass 4.5–19.1 g 20.1 g

Flesh mass 1.3–22.6 g 39.0 g

Nut mass 2.5–6.9 g 10.9 g

Kernel mass 0.09–0.56 g 1.14 g

Shell mass 2.2–6.8 g 9.7 g

Oil content 44.7–72.3% 62.2%

—

—

Body Shop, which is an international environmentally conscious and fair trade

cosmetics retailer. The French company Aldiva S.A. is also involved in commercial

development of the oil in partnership with CRIAA SA-DC and PhytoTrade Africa.

• Beer: Traditional marula beer produced by rural communities in South Africa

and Namibia and actively traded within producer communities.

• Marula kernels/traditional cooking oil: Mirma is presently exploring the

production of kernels for the local market. Kernels are sold at informal and

pension markets in South Africa and on a request basis from people's homes.

Kernels and oil are traded locally and within the main towns in Namibia. This

channel is not explored further due to the ad hoc nature of the local trade and the

fact that Mirma has not yet commenced commercial production.

In 2001/2002 the total value of the commercial marula product trade to local rural

communities was estimated to be worth Rand 1.1 million (US$ 110,000) and N$

350,000 (US$ 35,000) a year in South Africa and Namibia, respectively. This is in

comparison with the Rand 18 million (US$ 1.8 million) annual direct-use value in

Bushbuckridge alone. The marula trade in South Africa resulted in an average

annual income of Rand 578 (US$ 58) per trader for an estimated 1,900 trading

households over an average period of 29 days worked during the year. Approxi-

mately 1,100 households earned this income over eight months, whilst the

remaining 800 households earned this over a two-month period. In Namibia,

about 3,500 rural women have benefited from the sales of kernels. In South Africa,

the bulk of the income arrived at a critical time in the rural households’ calendar,

i.e. in January/February when cash is in very short supply. The cash earned from

marula product trade is approximately 10% of total annual cash income earned by

households and consequently has limited potential for investment. The signifi-

cance of cash benefits was limited due to the large number of people trading

marula products and the relatively small volumes bought by processors.

However, there is considerable opportunity to significantly increase the

commercial value of marula products through market growth and species

domestication, and to use this as a strategy for poverty alleviation and the

achievement of the UN Millennium Development Goals (Leakey et al., in press).

The marula trade generated a profit for all participating households when the cost

of their labour was excluded. The independent marula fruit traders supplying

fruits to the Distell marula pulp factory made the greatest profit (Table 10), while

traders of kernels and marula fruit to MDC made the lowest profit. Of interest was

that, although traders selling home-brewed beer earned greater than average

income, the addition of value did not earn them greater profits than the trade of

raw fruit to Distell. The substantial profit made by the suppliers of marula fruit to

– –

– –

150 Sheona Shackleton et al.

Distell was the result of the economies of scale and efficiency of transport. The

traders were able to coordinate their harvesting and transport of fruit, thereby

reducing spoilage. In comparison to prevailing wage labour rate in the region for

unskilled farm labour (Rand 12.50/day; US$ 1.30), the beer traders and the fruit

suppliers to Distell earned between 1.3 and 2.1 times more income for their effort

than the average farm worker. MDC suppliers earned between 0.3 and 1.2 times

the wage rate. The incomes earned from fresh marula products were however for

only a few months of the year. Noteworthy is the fact that there are significant

inefficiencies in the marula trade, with the greatest losses to local communities

incurred as a result of poor coordination of transport with the fruit buying

companies (in South Africa), and by not processing kernels immediately after

extraction (in Namibia).

MDC and Distell bought an estimated 2,200 tonnes of marula fruit for the

production of juice and pulp in 2002, while MDC bought 12 tonnes of processed

marula kernels in 2001. It was estimated that Distell produces 800 tonnes of pulp,

with MDC producing up to 10 tonnes of juice and 4 tonnes of oil per annum.

Eudafano Women's Cooperative in Namibia trades 20–42 tonnes of kernel per year

to CRIAA SA-DC.

In 2002, Distell invested Rand 1.5 million (US$ 150,000) in the local economy

through expenditure on processing at the Phalaborwa pulping plant. The

expenditure at the Stellenbosch distillery was unknown. MDC invested about

Rand 700,000 (US$ 70,000) in 2001 within the local economy. In total, the formal

marula industry in South Africa (excluding traditional beer) contributed to an

investment of Rand 2.2 million (US$ 220,000) within the local economy, with 50%

accruing to rural suppliers. In 2002, Distell produced pulp at an estimated Rand

1,900 (US$ 190) per tonne, and MDC produced juice (not pulp) at Rand 23,000 (US$

2,300) per tonne. In 2001, MDC produced marula oil at an estimated Rand 99,000

(US$ 9,900) per tonne. The Distell pulping operation was estimated to make a

profit of Rand 3,880,000 (US$ 38,800) per annum. The MDC oil operation made a

Commercialisation of multiple-use NTFP 151

Table 10. Annual economic benefits of the marula trade to localised economies in Namibia and South Africa.

Income to marula producing Average annual Total investment

Communities profit per Number into local

Channel (Rand) trader (Rand) of traders economy (Rand)

CRIAA SA-DC oil 250,000 3,500 250,000

MDC oil 290,000 150 1,100 394,000

MDC juice 64,000 360 220 140,000

Distell Amarula liquer 1,100,000 1,100–720 420 1,532,000

Bushbuckridge community beer 136,000 500 200 136,000

Total 1,838,000 5,440 2,452,000

All figures rounded off. Exchange rate is US$1.00 = Rand 10.00.

—

152 Sheona Shackleton et al.

Fig

ure

1.

Gen

eral

ised

co

mm

erci

al m

ark

et c

han

nel

s in

th

e m

aru

la i

nd

ust

ry i

n N

amib

ia a

nd

So

uth

Afr

ica.

—

profit of Rand 58,000 (US$ 5,800) during the 2001 season. The MDC juice operation

in 2002 was estimated to have made a loss of Rand 92,000 (US$ 9,200). In Namibia,

the Katutura Artisans Project and village associations' investment adds up to

about Rand 250,000 (US$ 25,000). The international returns to Amarula liquor are

unknown, but are believed to be significant with exports to 28 countries around

the world.

At an industry level, the supply of marula fruit exceeded demand with the result

that suppliers of fruit and kernels were price takers having no leverage to increase

price. Suppliers were thus vulnerable to the pricing strategies of the buying

companies. There were few barriers to selling marula fruit and kernels to the

processing firms. However, the current buyers – Distell, CRIAA SA-DC and MDC –

face significant barriers to enter the consumer market. Distell engages a multi-

million Rand marketing drive to develop international consumer demand. CRIAA

SA-DC has invested considerably in locating high-value international markets for

marula oil. At the time of the project, it had just concluded an agreement with The

Body Shop to supply marula oil for a range of cosmetic products. Subsequently, it

registered a patent in collaboration with PhytoTrade Africa and Aldiva S.A. for

Maruline, a marula oil derivative. This is also being registered as a Fair Trade and

Environmentally Sustainable product. MDC has benefited from the CRIAA SA-DC

marketing investment by supplying oil to The Body Shop in association with

CRIAA SA-DC, but is struggling to find its own markets.

There is no significant current consumer demand for beer and kernels outside

local markets. The market still has to be developed despite widespread knowledge

of the fruit. This may, however, be influenced by effective but expensive marketing

strategies, as demonstrated by the marketers of Amarula Cream, who have created

wide awareness of this product.

Pursuing models of commercialisation that meet the needs of the rural poor

While the above section reports the seemingly low economic returns from marula

commercialisation, a question that is frequently asked is – how are these returns

distributed, and are rural producers obtaining a fair share of the benefits? The

latter also needs to be considered in relation to the potential costs to different

community members that might arise as a result of commercialising this locally

important, multi-purpose resource. A wide variety of results can emerge from

different NTFP commercialisation models that may be adopted by the state, non-

governmental organisations, producer communities and the private sector. Four

main models of marula commercialisation currently operate in southern Africa:

the ‘Local Entrepreneur’ (local beer and kernel traders), the ‘Altruist’ (MDC), the

‘Honest Broker’ (CRIAA SA-DC) and the ‘Corporate Buyer’ (Distell) (Wynberg et

Commercialisation of multiple-use NTFP 153

al. 2003; Wynberg 2005). The benefits and risks involved in each are summarised in

Table 11, and illustrate the importance of developing models that are based on

partnerships between producer communities, NGOs and the private sector. All

these models also point towards ‘winning’ conditions that must be in place to

ensure that local people capture a larger portion of benefits and commercialisation

154 Sheona Shackleton et al.

Table 11. Marula commercialisation models.

Commerciali- Key players Financial Positive Limitations

sation model and activities benefits elements and risks

The local Women trading Annual income of Women retain Greater income than fruit

entrepreneur beer/wine/kernels Rand 500 to some 200 ownership and sellers, but on an effort basis only

in local markets households (BBR); control of the 1.3 times more than average farm

0–87% of women sell enterprise; worker; Seasonality of income;

wine and earn Low barriers to Market saturation as more women

Rand 50–4000 annually entry; Easy to initiate engage in the trade; Wastage due

(Namibia); trade without outside to poor shelf-life of beer.

40% of women sell actors; Value adding

kernels earning occurs; Entrepreneurial

Rand 30–400 each per skills are built; Evolves

season (Namibia) locally so minimal

impacts on other uses.

The altruist Mineworker's Annual income of Possibility exists for Reliance on external funding

Development Rand 127 to 197 community-based (although this is limited to the

Agency, donor- households involved ownership of development phase until a

sponsored job in selling fruit; the enterprise; viable business is developed);

creation project Annual income of Some linkages exist Generates the least income

producing marula Rand 1440 to 27 with foreign export per effort for suppliers,

pulp and oil. households involved markets; between 0.3 and 1.2 times

in processing Strengthening of that of the average farmworker;

236 tonnes of fruit organisational capacity Producers are price takers;

occurs at village level Difficulties in meeting

through setting quality and quantity

up of committees; required by corporate buyers

Low barriers to entry; based on low-technology,

Cost recovery community-based processing.

gradually being

established.

The honest Partnership Average annual Community-based Reliance on public monies

broker between Eudafano income of Rand 450 to ownership of the (although this is limited to

Women's Co- 750 producers (2000); enterprise; the development phase for

operative Ltd. and Average annual Linkages with institutional capacity building,

Katutura Artisans’ income of Rand 200 to foreign export research and development, and

Contd...

—

Commercialisation of multiple-use NTFP 155

efforts. These conditions include a wide range of management, institutional,

financial and market factors such as:

• the quality of information that the producer and/or agent has about markets,

and the efforts placed on creating and promoting markets;

• the extent to which individuals are organised as a group;

Commerciali- Key players Financial Positive Limitations

sation model and activities benefits elements and risks

Project, and the 1483 producers (2001). markets; High levels marketing); Difficulties in

NGO CRIAA of organisation meeting quality and quantity

SA-DC, centred within Cooperative; required by corporate buyers

on the production Involvement of based on low-technology,

of cold-pressed organisations community-based processing.

marula oil for with processing,

the cosmetic market management

and marketing skills

and capacity;

Minimises intrusion

into traditional use

of marula for beer/

wine and leaves marula

fruit in households

for consumption and

local trade;

Kernel production

and trade fits well

with agricultural and

other obligations;

Cost recovery

gradually being

established; all

marketing done

under Eudafano.

The corporate Production of Average annual Greatest profit for Model based on 'trickle down'

buyer Amarula liqueur income of Rand marula fruit traders benefits rather than community-

by Distell 1340 to 419 who earn 1.3–2.1 based ownership of the enterprise;

households; more than farmworker; Producers are price takers;

Rand 100,000 Company has the Larger profits hinge on traders

for development ability to invest in transporting fruit at their own cost;

projects in the area marketing, image- Skewed power relations between

building and producers and company;

product development. High levels of secrecy on the

part of the company, which

disenfranchises producers from

business; Availability of synthetic

replacements may displace

need for raw material.

Source: Wynberg et al. 2003.

• coordinated production within and between communities;

• skills in bargaining;

• the extent to which a producer and/or agent is 'networked' and linked into

partnerships;

• access to transport by the producer;

• the quality and consistency of products received;

• the existence of a diversity of products, markets, players, and industries; and

• awareness among consumers as to the product or brand.

The policy environment

As is the case for all NTFP, a range of laws and policies impact the management,

use and commercialisation of marula (Wynberg et al. 2002b). These include laws

and policies that directly relate to the resource base, such as natural resource,

agriculture, forestry and environment laws relating to land tenure and resource

rights; and a variety of economic and financial measures such as trade and

taxation. Relevant laws and policies are manifested at the international, regional,

national, provincial and local level – the latter in southern Africa including

customary laws and regulations, as well as a well-established (albeit fragmented)

statutory framework to protect trees.

In both South Africa and Namibia the dominant land tenure system is communal,

with land allocation and management falling under the jurisdiction of the

traditional authorities. However, with respect to the marula resource, marula fruit

is largely harvested from open access communal grazing lands in South Africa,

while in Namibia most trees are found in people’s fields where they are de facto

‘privately’ owned (by the families to whom the fields have been allocated). In both

these situations statutory laws and an additional layer of customary law apply,

with the latter being the system best understood and implemented by

communities living in the area. Additionally, the effective individual tenure over

trees in Namibia has led to a range of informal institutional mechanisms that

operate largely at the household level.

Marula tree felling

Customary laws form a central component of marula use and strong cultural

taboos have long existed to protect wild fruit trees on communal lands in South

Africa and Namibia. The results of the research indicated that marula cutting is

still strongly prohibited in all areas surveyed – by both customary and govern-

ment laws (Wynberg et al. 2002b). However, the practical application and enforce-

ment of this ban varies considerably across communities. In Bushbuckridge, for

example, the cutting of marula and other trees is common, largely as a result of

increased pressures for land and weakening of the traditional authority structures

due to changes in local government arrangements since 1994. By contrast, the

156 Sheona Shackleton et al.

Commercialisation of multiple-use NTFP 157

felling of marula trees in Namibia is not widespread. Tree felling in general in

Namibia has decreased since independence, attributed mostly to enhanced edu-

cation and awareness about the importance of trees, and increased cooperation

between government and traditional authorities reinforcing customary norms.

Marula fruit harvesting

In principle, fruit may be harvested by anyone, although it is typically women who

are involved in this task. A traditional rule that applies across all study areas

requires fruit to fall to the ground before harvesting is permitted. At all sites

visitors are required to seek permission to harvest fruit in the homesteads or

farms, or in close proximity to these, of individual households, but the communal

areas of Bushbuckridge are typically regarded as open access areas from which

both villagers and outsiders may collect. However, in some villages, there are

indications that growing commercialisation is leading to intolerance towards

outsiders. At Makhatini, more control is exerted over communal lands, and a

greater distinction is made between outsiders – who require permission from the

headman to harvest fruit in these areas – and community members, who do not. In

Namibia, land and resource tenure were most defined, and individuals must seek

permission before harvesting fruit from individually tenured fields or the

common lands. In all cases, however, the interpretation of ‘ownership’ is highly

debatable and usually translates as a user right only, carrying no legal status.

Across all sites there was a lack of clarity as to the rules governing access to marula

fruits within conservation areas, titled private lands, and state and muni-

cipal lands – from communities, traditional authorities and government officials

alike.

Monitoring and enforcement

Despite widespread awareness of rules governing marula use, practical imple-

mentation and enforcement remain weak. Just how significant an issue this is

depends to a large degree on the extent to which rules are being infringed, the type

of tenure system, the levels of cooperation between traditional authorities and

government, acceptance of the rules by user groups, and the levels of capacity that

exist within authorities. In Namibia, an emphasis is placed on self-monitoring by

the community and the village headmen, and this is an approach considered to be

efficient. In South Africa, and especially in Bushbuckridge – an area where levels

of marula cutting and rule violation appear highest – monitoring and enforcement

is particularly weak, and is exacerbated by conflict and confusion with regard to

jurisdictions. A major difference between South Africa and Namibia lies in the

tenurial systems that apply to trees, and it is suggested that self-policing is likely

easier for trees tenured to households, as in Namibia and to a lesser extent in South

Africa, than for trees in the commons surrounding villages.

Across all study sites, and in South Africa in particular, marula conservation is

considered a relatively low priority, and only two cases were reported for the

infringement of marula rules over the past five years (Wynberg et al. 2002b).

Interviews with villagers, however, suggest that this number does not reflect the

actual extent of cases.

Legal systems and institutions

Cooperation between government and traditional authorities emerged repeatedly

as an issue affecting marula conservation and use (Wynberg et al. 2002b).

Respondents at all study sites reported consistency in the rules of both

government and traditional authorities with regard to marula, but differences in

the implementation, enforcement and coordination of these approaches. In

general, government laws were perceived to be both more stringent in their

enforcement, and more general in their application, pertaining broadly to all trees.

Coordination between the traditional and official governance systems was

regarded as an important factor for the successful implementation of laws.

A crucial role is played by customary law, which has greater influence than

provincial or national law over marula tree management, both in terms of local

knowledge of rules and regulations, and enforcement. In Namibia, these were

maintained and upheld strongly; and this, combined with increased private

tenure over trees, seems to have led to greater levels of resource control and

management. In South Africa, however, the changing political and social context

have exacerbated resource management problems and led to an erosion of the role

of traditional authorities. Customary laws governing marula use have thus

become less rather than more effective in recent years in South Africa. The

weakening of the traditional authorities, and the policy vacuum in which they

operate is cause for concern, particularly in the face of dwindling resources from

government.

Intellectual property rights and traditional knowledge

A rich traditional knowledge exists about marula, ranging from knowledge of its

medicinal and insecticidal properties, to that of the special properties of marula oil

and the domestication of the tree by communities and their selection of desirable

traits across generations. Marula commercialisation demonstrates both the

potentially positive and negative roles that intellectual property rights can play in

protecting the interests of primary producers. Trademarks and geographical

indications, for example, can assist in protecting southern African marula

producers – and their suppliers – from competitors selling products that do not

contain any marula, or which are derived from sources outside the natural

distribution area of marula. But through Plant Breeders Rights, intellectual

property rights can also play a negative role in preventing rural communities from

158 Sheona Shackleton et al.

Commercialisation of multiple-use NTFP 159

using material that has been ‘improved’ by commercial breeders, and in facilitat-

ing the unfair use by commercial breeders of material that was originally

‘improved’ and domesticated by communities.

There are several international laws regarding intellectual property rights

including the TRIPS (Trade-Related Intellectual Property Rights) Agreement of

the World Trade Organisation which requires member states to protect intellectual

property rights, and the UPOV (International Union for the Protection of New

Plant Varieties) Convention which protects new varieties of plants. However, the

UPOV system is geared towards industrial application and is poorly suited to

protect the interests of rural producers. The monopoly nature of UPOV and Plant

Breeders' Rights makes its application to marula particularly inappropriate given

the strong social conditions and culture of marula, which are community rather

than individually based. The 1998 African Model Law for Community Rights

provides an alternative to protecting the rights of holders of knowledge, and to

ensuring that they receive a fair share of benefits arising from commercialisation

of protected varieties (for example, those communities involved in domesticating

the 'Namibian Wonder'). Currently, both a 'community rights' model (viz. the

African Model Law for Community Rights and the World Agroforestry Centre)

and a ‘privatisation’ model (through UPOV) for domestication exist for marula,

and urgent efforts need to be made to develop systems of protection for

communities that do not involve monopoly rights and which promote poverty

alleviation, food security and sustainable agriculture.

Contribution of marula to livelihood assets and trade-offs with commercialisation

One of the key objectives of the marula project was to assess the effects of marula

commercialisation on rural livelihood systems, household income, socio-cultural

values and norms, subsistence use, local markets and resource sustainability. In

this section we highlight the important and diverse role that marula has in local

livelihoods and in contributing to the five forms of livelihood capital outlined in

the sustainable livelihoods framework (Carney 1998). We then consider the

existing and projected positive and negative impacts of commercialisation on

livelihoods.

We strongly feel that commercialisation should not disrupt the crucially important

subsistence and social benefits that marula has bought to local households for

generations. Respect for traditional non-monetary values such as reciprocity,

resource sharing and communal processing is vitally important, if the full benefits

from monetary markets are to be reaped, and if the commercialisation process is to

bring equity and benefits to the poorest of rural households. If valuable and

unaccounted for socially or ecologically important attributes are lost in the

commercialisation process, overall livelihoods may be adversely affected. All too

often these values are forgotten in the drive to maximise economic returns. Trade-

offs between different forms of livelihood capital are, of course, inevitable, but

more often, it is social, cultural or natural capital that loses out to financial capital

given the overriding need for cash income. If due consideration is not given to

these non-financial benefits, commercialisation could have negative outcomes

and make some people worse off, instead of having the desired effect of enhancing

livelihoods.

Human capital

Health, nutrition and food security. The consumption of fruit, beer, juice, jam and

kernels makes an important contribution to the diet and nutrition of most of the

rural households at the different project sites. The marula fruit is particularly rich

in vitamin C, and the kernels offer a good source of protein, micronutrients and oil

during the dry season when food is scarce (Thiong’o et al. 2002; Shackleton et al.

2002a for a review). This nutritional role of marula may assume added significance

when viewed in the context of the HIV/AIDS epidemic in southern Africa, in

which nutrition and food security are crucial issues. The emperor moth cater-

pillars found on marula during early summer also provide an important source of

protein.

In Namibia, so far, domestic consumption appears to have been unaffected by

commercialisation. People use the same variety and quantities of marula products

as they did previously and rather expend more time and effort to produce

sufficient quantities to participate in the commercial trade. Up until recently, the

marula resource was considered abundant enough to cater for all needs, but a

majority of households (75%) now feel that supply is insufficient to meet demand,

potentially impinging on domestic requirements. The situation in Bushbuckridge

is much the same, with domestic use currently far exceeding commercial use. At

the moment, less than 20% of total fruit collected is used for commercial purposes.

In this area, supply appears abundant enough to satisfy current traditional and

commercial use. In Makhatini cultural taboos currently ensure that the resource is

used primarily for domestic consumption.

Processing approaches used during commercial production, which increase yield

and commercial gains, can at the same time negatively affect the range of marula

products available at household level. For example, the use of fruit presses reduces

the amount of flesh remaining on the nuts after pressing, and leaves the skins on,

which in turn decreases the availability, quantity and quality of the juice made by

agitating these nuts with water. This could have potentially negative nutritional

consequences for children, since they are the main consumers of this product.

Similarly, retention of kernels for commercial use by tree-owning households

160 Sheona Shackleton et al.

Commercialisation of multiple-use NTFP 161

limits their availability to households that do not own trees, as currently practiced

through shared processing of the fruit (see below).

Overall, however, marula commercialisation could have a positive effect on health

and nutrition if some of the income from the sales of marula products is directed

towards other critical household needs. This is particularly the case if women

remain in control of the income derived from marula trade as they tend to spend

this directly on immediate family needs. Moreover, the importance of home

consumption of nutritionally rich indigenous fruits for boosting the immune

systems of HIV/AIDS sufferers is currently being highlighted (Barany et al. 2001;

Hemrich 2003), especially within the context of southern Africa, which has the

highest HIV/AIDS infection rates in the world.

Local skills and knowledge. Knowledge of where to find the best fruits and skills

on the uses and processing of marula are shared amongst all women in the areas

where marula occurs, with this usually being acquired at a young age. In recent

years, however, there has been reduced interest amongst the younger generation.

The cash benefits brought by commercialisation could play a role in reversing this

trend by providing an incentive to retain this traditional knowledge and skills. In

Bushbuckridge, a number of young girls were encountered who were selling

marula products for pocket money. Commercialisation has also built

entrepreneurial skills. For example, women selling beer are now extending their

trade to other products. The income earned from commercialisation is invariably

invested into the good of the household including the education of children via

school fees, books and uniforms, hence contributing to the human capital and

empowerment of rural households and communities in general.

Empowerment of women. In both South Africa and Namibia, all aspects of the

marula supply and marketing chain are dominated by women. Participation in the

marula trade has provided women, particularly those heading their own

households, with independent income and stature. There is evidence

internationally that as cash earnings increase and enterprises grow, men begin to

take up activities previously seen to be the province of women (e.g. fuelwood

trading). In many instances, the role of men may be increased through

introduction of processing technologies, for example, fruit presses in Namibia

(Box 3), overriding some of the social stigma associated with performing women's

tasks. The possibility then exists that they take over and dominate the trade to the

disadvantage of the women who were first involved, although is not an issue as yet

in the marula trade.

In Namibia, traditional group processing of marula is important for women as an

opportunity to gossip, cement bonds, talk about their problems, and look for

solutions. While such interaction also happens at other times of the year, the

marula season is seen particularly as a time for prolonged social contact,

communal activity and sharing, often with distant households. Thus, women’s

cooperatives were established for the selling of kernels. The survey results showed

that over one third of households interviewed processed marula fruits as often as

5–6 times per week during the season. The increased contacts they have with the

'outside world', the recognition that they now receive from the rest of community,

their ability to 'help themselves', and the negotiation and bargaining power they

have gained have all added to their self-esteem and range of skills. However, there

is a possibility that rising incomes may increase selfishness and individualism and

slowly erode traditional cooperative activities and the sharing of the marula

resource.

Social capital

The cultural, spiritual and social significance of marula has already been

mentioned, and can be seen in the numerous traditional songs, dances and stories

around marula, its role in rituals and celebrations and the customary regulations

surrounding it.

In southern Africa, the marula season brings people together, and is a time of great

festivity. In most areas people still present beer to the headman in celebration of

162 Sheona Shackleton et al.

Box 3. Consequences of introducing a fruit press in Namibia (den Adel 2002)

—

Some indications about future changes can be learnt from the experiences with a marula juice press that

was introduced at the beginning of the 2002 season in Namibia. The press was made on request of the

Eudafano Women's Co-operative (EWC), which felt that the processing of fruits should be smoothened

and increased. The press was introduced in the ten villages where EWC associations are active. The

associations decided to charge a small fee for the use of the pressing machine, for purposes of

maintenance, and as an income-generating activity. Technically, the press showed little problems, and all

women were excited and impressed with the amount of juice processed in a short time. Upon questioning

though, some concerns were raised. One complaint was that the press, because of its usage fee, was

accessible only to the relatively wealthy households. Additionally, some wealthy households who owned

a lot of trees were now pressing their marula with little involvement of their neighbours and friends.

Those without access to trees on their land especially felt left out on the action and of their fair share of

beer, juice, and kernels. Because of their neighbours' complaints, the 'wealthy households' decided to try

out the fruit press only on a limited number of trees, but it is not clear whether this neighbourliness will

prevail when and if marula becomes 'big business'. Another possible outcome of marula becoming a

larger and more profitable enterprise relates to the gendered nature of marula processing. The processing

of marula has always been a women's activity, and money from sale of marula products has, up to now,

been in the hands of women only. There is, however, considerable male interest in the marula fruit press,

and the possibility exists that men may become more involved and even dominate the trade as processing

becomes more accessible to them, and greater profits.

Commercialisation of multiple-use NTFP 163

the season, although this tends to be much less formal and ritualistic than in the

past. In all regions, the giving, sharing and drinking of marula beer with friends,

neighbours and relatives are crucial for building social bonds, networks and

reciprocal obligations (Shackleton et al. 2002a). Indeed, some women make beer

solely for the purpose of giving it away, as no one in their household drinks it. This

significant social function of marula extends way beyond the actual fruiting

season and is a form of ‘insurance’ to households who might need to call on their

neighbours for support (labour, assistance, food, a loan, etc.) later in the year. The

principle is simply one of ‘give today and you may receive tomorrow’. Many

family members also return home from distant urban areas during the season to

share in the beer, reinforcing family bonds and ensuring the continuation of the

important urban–rural linkages upon which many households depend. These

visitors also often arrive with numerous gifts for the household. In Namibia, the

marula season brings a change in the traditional rules as well; people are not

permitted to carry any weapons, and the traditional court closes for those few

months.

While numerous local people feel that economic necessity overrides all other

aspects of life, others fear that large-scale marula commercialisation will lead to a

loss of social and cultural capital. Some individuals were concerned that

commercialisation of, especially, fruit and beer would ultimately result in

increased individualism and selfishness, theft (as evidenced in Bushbuckridge)

and the loss of traditional ways and community structure. Certainly the findings

at Edinburgh (see Box 4) tend to support this. There was also concern amongst

headmen, particularly in Namibia, that the practice of people offering them beer

would be lost completely if commercialisation developed further. Such offerings

have the role of acknowledging and reinforcing traditional institutions and

structures. A similar concern relates to the potential decrease or disappearance of

the much-praised informal marula gatherings, and the accompanying

strengthening of social contacts. In Makhatini, where marula commercialisation is

virtually absent, many respondents indicated that, although they would be

interested in selling fresh fruit or nuts, they were reluctant to sell beer due to its

social implications. In Bushbuckridge this concern is addressed somewhat by the

fact that the traders only sell beer in the towns and not their own villages.

In Namibia, the potential negative effects of commercialisation, particularly of

beer and fruit, are possibly more complex than in South Africa because of the

sources of fruit and social setting within which beer production takes place. In this

country, marula beer production is an institution in itself. Neighbours help each

other brew this drink, and expect assistance in return. Women without trees in

their own farmland, also aid in processing marula beer, even if they get a little or

no beer to take home. They do so because it provides them with access to nuts,

reciprocal assistance, social networks and often an invitation to share the beer later

(Box 5). These social aspects of processing marula in Namibia, combined with the

tradition of giving away marula products to neighbours, friends, visitors and

relatives, are mechanisms that make marula a resource accessible to all in a

situation where most trees are found in individual fields rather than in the grazing

commons. Apart from the tradition of ‘giving away’ becoming somewhat less, the

commercialisation process has so far not had a significant negative impact on the

social aspects of marula. It is however likely that large-scale commercialisation

will change certain aspects of the marula social culture in the future. It is difficult to

predict in what ways changes will take place. It is likely that households that have

no trees (some 22% of respondents interviewed), and those with little or no cash

income, will lose relative to the current situation. Some indications of what further

commercialisation of fruit products might bring is illustrated in the narrative

around the introduction of a fruit press (see Box 3).

Compared to beer and fruit trading, there was generally far less concern around

the sales of the kernel resource. Although of social and cultural importance, the

main difference is that kernels are generally not produced and consumed within

social settings other than the household. Thus, they do not have the same role in

building social networks and cohesion, although kernels are often exchanged in

friendship (Shackleton et al. 2002a; Shackleton and Shackleton 2005). There is also

less concern amongst the traditional leadership about this resource. For example,

the headman of Hokwe in Bushbuckridge was opposed to the sales of fruit and

beer for the reasons mentioned, but he did not prevent villagers from selling

kernels.

164 Sheona Shackleton et al.

Box 4. Negative commercialisation impacts in Edinburgh village, Bushbuckridge, South Africa

—

Although there appear to be no strongly negative consequences of commercialisation at this stage, closer

analysis demonstrated some subtle dynamics around the marula resource in Edinburgh – the village with

the highest number of households (62%) involved in supplying the two commercial producers. We

observed that Edinburgh had the lowest percentage of households giving away fruit or nuts/kernels (6%

compared to 20–30% in the other villages), and lowest proportion of households allowing others to collect

from 'their' trees (less than 20% compared to about half in the other villages). In addition, fewer households

than in other villages indicated that people from outside of the village could collect freely from the

communal lands around Edinburgh. More said that permission from the headman was required – which

suggests a degree of 'privatisation' of the resource at village level, or at least a move from the marula

resource being open access to it becoming the common property of the village. The highest number of

households reporting theft occurrences was also recorded in this village. On the other hand, it was

interesting to note that more households in Edinburgh practised xikuha (presentation of beer to the

headman), and certainly as many households as in the other villages were sharing their marula beer.

Commercialisation of multiple-use NTFP 165

The contribution of marula to social capital should not be underestimated. Social

assets are often more central to rural livelihoods than is generally acknowledged

or appreciated, and are an important safety net in times of hardship. There is

increasing reference in the literature to the breakdown of social capital and

reciprocal relations and the emergence of greater individualism, often to the

detriment of local farming and natural resource management systems, and to the

poorest members of society (Campbell et al. 2001; Paton 2002). Any practices that

help maintain or build community cohesion and cooperation should therefore be

supported and encouraged rather than compromised for other livelihood assets.

Such issues demand careful consideration in the commercialisation process, with

some approaches, e.g. The Honest Broker (Table 11), requiring cooperative work

in collecting, processing and marketing of marula that builds on and fosters

existing social relations.

Financial capitalThere is no doubt that marula makes an important contribution to financial capital

of households selling marula products, albeit extremely seasonal (Shackleton

2004). The trade provides women, in particular, with access to an independent

source of income (i.e. income that they control), which is especially important for

women-headed households (Shackleton 2004). Although, on average, incomes

earned from trading either beer, kernels and fruit were only a few hundred Rands

over the season, in areas of high unemployment and poverty, traders mentioned

that it all counted towards them being able to put food on the table, pay school fees,

clothe their children or purchase household goods. Many people mentioned how

they were 'suffering', and how the sales of marula products have helped to ease

this, albeit a few months of the year. The extremes in gross income earned, e.g.

from between Rand 8 and Rand 899 (US$ 1–US$ 90) per household for fruit, and

Rand 11–Rand 1,900 (US$ 1–US$ 190) for kernels in Bushbuckridge, indicate that

Box 5. Benefits of group processing of beer in Namibia (den Adel 2002)

—

Women work together to make beer/wine because:

• They enjoy the social event. While processing the fruit they catch up with the latest news and gossip,

make jokes, and sing their favourite marula songs. Young girls are also often initiated into the skills of

brewing.

• They build up their ‘social bank account’, strengthen their social bonds, and may expect support in any

form in return.

• They have access to kernels and juice, as they can take home the nuts with the remaining flesh from

fruits that they have processed. They will have kernels for subsistence use, and maybe even enough to

sell.

• They will be invited to parties at their neighbours' houses to enjoy the beer they have made.

households that apply themselves seriously to the production and sales of marula

products are able to earn an income of some significance, particularly to extremely

poor people (up to ± US$ 190 over the season).

Due to its highly seasonal nature and the relatively low incomes earned, the

financial contribution marula makes needs to be seen in the light of a diverse

livelihood base in which each component would, most likely, not be able to

provide a sustainable livelihood on its own, but when pooled can help ensure that

people are able to meet their basic living requirements. People can survive on

relatively small budgets in rural areas, but cash is needed for paying school fees,

health services, basic goods and supplementary sources of food, especially during

years with inadequate rainfall. A large proportion of households lived on less than

Rand 250 per month (US$ 25), and therefore any additional sources of cash,

however limited, are imperative in helping households pay for their basic needs.

On average, the trade contributed an additional 10–15% to the cash income earned

annually by trading households. Amongst some particularly destitute

households, the sales of marula products provided one of the few sources of cash

income. In 2 χ and

that individuals participating in the trade tend to come from less well-off

households than those not trading (t = 2.08, P < 0.05) (Shackleton and Shackleton

2005). 'Freely' available natural resources can therefore perform an important

safety net function when few other options exist. It was in this context that the first

commercialisation of marula occurred in Makhatini by two households that

desperately required cash to buy food.

One of the important features of a natural resource such as marula is that there are

few economic barriers to entry to the trade: the fruit is freely available and

accessible, and no start-up finance is required. Some of the marula beer traders

said that their traditional leaders 'turned a blind eye' to the fact they were

infringing customary taboos by selling beer and fruit because they recognised the

hardship many people were facing and their need to earn income. Preventing a

woman, especially a single mother, from selling marula fruit, beer or kernels

would be akin to denying her children their education. Similarly, in Namibia, even

the more traditional of headmen acknowledged that times were changing, and

people were forced to sell products that were traditionally exchanged as gifts,

though most of them make a clear distinction between the selling of marula beer,

which has important cultural values attached to it, and the sale of marula kernels,

which is generally more accepted. Nevertheless, beer is sold both at a local level (in

contrast to Busbuckridge where it is only sold in the towns) and in the towns in

Namibia.

Bushbuckridge, it was found that all but three households with no

regular sources of income were selling marula products ( = 4.63, P = <0.05),

166 Sheona Shackleton et al.

Commercialisation of multiple-use NTFP 167

Of importance is that, in Bushbuckridge, despite a mainly positive attitude to

commercialisation, traders who sold marula products to commercial producers

felt that they were not receiving a fair price for the fruit. The price was said to be

inadequate in terms of the time and effort that went into collecting. The comment

'they are robbing us' was expressed frequently in the interviews. There was,

however, less dissension around the price obtained for kernels. Another fre-

quently raised problem related to the poor reliability of the commercial producers’

buying system, i.e. the trucks sent out to buy often did not arrive. Similarly,

producers delivering kernels to MDC were often turned away either because the

daily quota was met, or because they had no cash to purchase. Consequently, both

fruit (which quickly over ripens) and people’s time were wasted. Whilst

appreciative of a new opportunity to earn income, local people thus felt that they

were the ‘losers’ in some aspects of the process.

Natural capital

In South Africa, tree density is declining in communal lands, largely as a result of

human impacts including deforestation for farming and energy, and livestock

browsing of regenerating vegetation, exacerbated by poor enforcement of

traditional and formal regulations. In homesteads and fields, in Namibia and

South Africa, planting (truncheons and seeds) and nurturing of seedlings has

maintained and in some instances increased densities of marula trees. It is

probable that commercialisation is more likely to have a positive impact on the

marula resource base, through increased cultivation and protection of trees, than a

negative one, especially if linked to a participatory tree domestication programme

in key villages/communities. In both South Africa and Namibia, respondents felt

that increased value of marula products would stimulate greater protection of the

resource. This will be a result of not just the economic benefits, but also their pride

in this unique resource, appreciated both locally and by the outside world.

Domestication is likely to greatly enhance the quality of the genetic resource in

areas where it is planted with improved germplasm. Evidence from West Africa,

indicates that during the early stages of domestication by local communities, the

genetic base is diversified (Leakey et al. 2004), even without deliberate conser-

vation measures. However, it is important that the domestication strategy adopted

is one that deliberately enhances and maximises the genetic diversity in planted

populations, by ensuring that the wild resource is protected, that a number of

cultivars (20–30) are grown, and that they originate from geographically distinct

populations. Over time, new selections from the wild and from a breeding

programme should further diversify the domesticated stock. Throughout the

domestication process, components of the wild resource must also be conserved.

The integration of cultivars within agroforestry practices based on marula and

other fruit and medicinal species could result in landscapes that are biologically

more diverse than is currently found (Leakey 1999), so enhancing the natural

capital at the agroecosystem level.

Physical capital

The contribution of marula to physical capital at a household level has been

limited with most income earned being spent on day-to-day needs. However, at

community level, producer groups in Namibia now have access to technology and

equipment in the form of fruit presses, scales, decortication blades, storage

facilities and improved communications through the acquisition of cell phones.

MDC in Bushbuckridge has provided similar facilities, although these are not

widely available for community members to use. Distill Corporation has set up a

community trust fund which provides grants and donations of infrastructure to

participating communities. Increased commercialisation is unlikely to have any

negative effects on physical capital, although as mentioned above, one potential

unintended consequence may be that technological advancement and new equip-

ment could result in women being sidelined and men taking over processing.

Conclusions

In response to concerns that commercialisation of natural resources often leads to

over exploitation and may have unintended impacts on livelihoods, this analytical

review was initiated to examine the potential of different NTFP commercialisation

pathways to reduce poverty and sustain the natural resource base, while also

supporting other aspects of rural livelihoods. This was considered important vis-à-

vis recent initiatives to domesticate indigenous fruits and nuts (Leakey et al. in

press). The integration of the different components of the work on marula offers

some insights into the trade-offs made, or to be considered, during the commer-

cialisation and domestication of a multiple-use NTFP. There is little doubt that the

multiple-use characteristic poses some particular challenges, and opportunities,

not evident for single use species, and the multiple uses inevitably result in some

explicit or implicit trade-offs.

As a start, resource use and commercialisation involves trade-offs at different

levels and between different actors. A few examples from this case study include

(i) removal of male trees to increase the proportion of fruit-bearing females results

in lower pollination success and hence number of kernels per fruit, (ii) the

introduction of fruit presses to increase juice yield can alter social dynamics

around processing, such as changing the group approach, increasing male

'interest' in what is a traditionally female domain, and altering the quality of the

juice, and (iii) capturing the potential financial value of marula resources involved

trade-offs in terms of possible privatisation of the resources and reduced sharing.

Understanding how these trade-offs are made, the scale at which they operate and

how they influence adherence to norms and rules is key in predicting the possible

168 Sheona Shackleton et al.

range of outcomes from commercialisation initiatives. This is particularly so when

subsistence use, local-level commercialisation and externally driven

commercialisation all need to be accommodated in the same geographic regions

and based on the same resource. A balance needs to be sought and maintained

between them in order to optimise asset accumulation within total livelihoods and

not just one dimension. Before commercialising an NTFP, especially an important

multiple-use resource like marula, it is imperative to have an understanding of the

existing uses of the species and its importance for different livelihood assets.

A particularly robust finding in this study across the three sites was the

significance of marula in local traditions, ceremonies and in building social

capital. The role of social capital is frequently difficult to quantify or analyse on the

same basis as other, more tangible, forms of capital, but it is no less important.

Commercialisation ventures seemingly threatened this, and were hence resisted

to some degree by the traditional leadership or by major role players or users. Yet,

the need to access cash incomes in areas of high poverty posed constant tensions in

this regard. It is vital that externally facilitated commercialisation efforts pay heed

to this and minimise the loss of social capital.

Of particular interest were the different models and scales of commercialisation,

ranging from local-level, self-initiated entrepreneurs, through to large-scale,

export-orientated corporations. Seemingly, economic efficiencies were not

necessarily related to capital outlay. Local-level traders without any external or

capital investment received better returns per unit effort than some capital-

intensive approaches. Nonetheless, the magnitude of total returns is clearly

greater for large-scale, export-orientated initiatives. Such scaling up, however,

may result in small-scale users losing access to either the resource or to markets

(Wynberg et al. 2003), although that is not currently the situation at the three

project sites. Monopolisation and privatisation of the resource, through a reduced

number of participants, need to be guarded against, whilst at the same time

seizing upon opportunities for local upliftment. In the study areas, there was a

diversity of commercialisation initiatives that ensured benefit sharing across all

scales. Multiple models based on partnerships between producer communities,

NGOs and the private sector are most likely to succeed. While it may not make

sense for producer groups to export, distribute or market their own products,

partnerships between different players can allow for mutually profitable

arrangements, as evidenced in the CRIAA SA-DC marula oil model. Most

important is the retention of ownership and control of the enterprise at producer

level.

Central to the notion of multiple models, especially for multipurpose species, is

the fundamental tenet of diversification. This should apply across the species

used, products produced, markets traded and players involved, as an important

169Commercialisation of multiple-use NTFP

strategy to minimise the potential risks associated with NTFP commercialisation

for rural communities. This is especially true for marula, given its highly seasonal

nature and relatively low income-earning potential. Diversity in end-products

(e.g. fruit, beer, kernels, oil, jam) and industries (e.g. local beer traders to kernel oil)

is equally important to reduce overall risks and deliver greater benefits for

primary producers in terms of absolute income and the period over which this is

earned. However, it is best to build on what exists at the local level rather

than aiming for high value, specialised markets. Diversification also allows

new avenues when the costs of trade-offs become too great for one or more role

players.

Lastly, it is useful to reiterate that rural communities are not just passive

consumers or bystanders. They engage in a diverse portfolio of activities to secure

a livelihood, in which NTFP play a meaningful role. In terms of building natural

capital, inhabitants at the three study sites actively selected for and promoted

beneficial traits of marula. In terms of social capital, communities had developed

an array of taboos, customs and local rules, supported by local institutions. And

with respect to financial capital, they were role players, participants and lead

agents in an assortment of commercialisation models. They recognise potential

opportunities and mobilise resources and make trade-offs to capture such

opportunities, often having to overcome numerous constraints along the way. In a

similar way, when communities are involved in the participatory domestication

models for indigenous fruits and nuts (Leakey et al. 2003), it can enhance

commercialisation, with positive impacts on natural capital and livelihoods.

In conclusion, this study has highlighted the complex network of interacting

factors affecting the livelihoods of the rural poor and their natural assets, and it is

clear that, while there is potential for commercialisation of indigenous fruits to

have negative impacts on rural livelihoods and natural capital, there is equal

potential for positive impacts. Whilst commercialisation of marula has resulted in

changes in local livelihoods and structures, it has also brought many benefits, with

different commercialisation pathways yielding different profiles of beneficiaries.

It is therefore important that programmes promoting the domestication and

commercialisation of NTFP recognise that the approach taken will determine who

and what are the 'winners' and the 'losers', especially for a multiple-use resource.

In all instances a key challenge is to find the balance between providing much

needed new options and opportunities whilst minimising the impacts on existing

systems of resource use and benefits

.

ReferencesBarany, M., A.L. Hammett, A. Sene and B. Amichev. 2001. Non-timber forest benefits and

HIV/AIDS in sub-Saharan Africa. Journal of Forestry 99: 36–41.

170 Sheona Shackleton et al.

Botelle, A., du Plessis, P., Pate, K. and R. Laamanen. 2002. A survey of marula fruit yields in

North-Central Namibia. Unpublished report, CRIAA SA-DC, Windhoek. 22 pp.

(available at http://www.ceh-wallingford.ac.uk/research/winners/literature.html).

Burger A.E.C., J.B.M. de Villiers and L.M. du Plessis. 1987. Composition of the kernel oil and

protein of the marula seed. South African Journal of Science 83: 733–735.

Campbell, B., W. De Jong, M. Luckert, A. Mandondo, F. Matose, N. Nemarundwe and B.

Sithole. 2001. Challenges to proponents of CPR systems: Despairing voices from the

social forests of Zimbabwe. World Development 29: 589–600.

Carney, D. 1998. Implementing the Sustainable Livelihoods Framework, ODI, London.

den Adel, S. 2002. Use of marula products for domestic and commercial purposes by

households in North-Central Namibia, Unpublished report, CRIAA SA-DC,

Windhoek. 35 pp. (available at http://www.ceh-wallingford.ac.uk/research/winners/

literature.html).

du Plessis, P., C. Lombard and S. den Adel. 2002. Marula in Namibia: Commercial chain

analysis. Unpublished report, CRIAA SA-DC, Windhoek. 28 pp. (available at

http://www.ceh-wallingford.ac.uk/research/winners/literature.html).

Dzerefos, C.M., E.T.F. Witkowski and C.M. Shackleton. 2003. Host preference and density

of woodrose-forming mistletoes (Loranthaceae) on savanna vegetation, South Africa.

Plant Ecology 167: 163–177.

Dzerefos, C.M., C.M. Shackleton and E.T. Witkowski. 1999. Sustainable utilisation of

woodrose-producing mistletoes (Loranthaceae) in South Africa. Economic Botany 53:

439–447.

Hall, J.B., E.M. O’Brien and F.L. Sinclair (eds). 2002. Sclerocarya birrea: a monograph. School

of Agricultural and Forest Sciences, University of Wales, Bangor.

Hemrich, G. 2003. Food Security and HIV/AIIDS: An Update, FAO, Rome.

Laird, S. 1999. Forests, culture and conservation. In: Cultural and Spiritual Values of Biodiversity.

(Ed. D.A. Posey), UNEP.

Leakey, R. 2005. Domestication potential of marula (Sclerocarya birrea subspecies caffra) in

South Africa and Namibia: 3. Multiple trait selection. Agroforestry Systems 64: 51–59.

Leakey, R.R.B. 1999. Agroforestry for biodiversity in farming systems. In: Biodiversity in

Agroecosystems (eds W.W. Collins and C.O. Qualset). pp. 127–145. CRC Press, New

York.

Leakey, R., S. Shackleton and P. du Plessis. 2005a. Domestication potential of marula

(Sclerocarya birrea subspecies caffra) in South Africa and Namibia: 1. Phenotypic

variation in fruit traits. Agroforestry Systems 64: 25–35.

Leakey, R., K. Pate and C. Lombard. 2005b. Domestication potential of marula (Sclerocarya

birrea subspecies caffra) in South Africa and Namibia: 2. Phenotypic variation in nut

and kernel traits. Agroforestry Systems 64: 37–49.

Leakey, R.R.B., K. Schreckenberg and Z. Tchoundjeu. 2003. The participatory domestication

of West African indigenous fruits. International Forestry Review 5: 338–347.

Leakey, R.R., Z. Tchoundjeu, K. Schreckenberg, S.E. Shackleton and C.M. Shackleton. 2005.

Agroforestry tree products: targeting poverty reduction and enhanced livelihoods.

171Commercialisation of multiple-use NTFP

International Journal for Sustainable Agriculture 3: 1–23.

Leakey, R.R.B., Z. Tchoundjeu, R.I. Smith, R.C. Munro, J-M. Fondoun, J. Kengue, P.O.

Anegbeh, A.R. Atangana, A.N. Waruhiu, E. Asaah, C. Usoro and V. Ukafor. 2004.

Evidence that subsistence farmers have domesticated indigenous fruits (Dacryoides

edulis and Irvingia gabonensis) in Cameroon and Nigeria. Agroforestry Systems 60:

101–111.

Luoga, E.J., E.T.F. Witkowski and K. Balkwill. 2000. Differential utilization and ethnobotany

of trees in Kitulanghalo forest reserve and surrounding communal lands, eastern

Tanzania. Economic Botany 54: 328–343.

Mander, M., J. Cribbins, S.E. Shackleton and F. Lewis. 2002. The commercial marula

industry in South Africa: A sub-sector analysis. Unpublished report, Institute of

Natural Resources, Pietermartizburg. 71 pp. reports available at http://www.ceh-

wallingford.ac.uk/research/winners/ literature.html).

McHardy, T. 2002a. Inventory of available marula resources on the Makhatini Flats,

Maputaland, in the fruiting season of 2002. pp. 15. Unpublished report, Institute of

Natural Resources, Pietermaritzburg. (available at http://www.ceh-wallingford.

Ac.uk/research/winners/ literature.html).

McHardy, T. 2002b. Use of marula products for domestic and commercial purposes by

households in the Ophande district, Maputaland, South Africa. pp. 29. Unpublished

report, Institute of Natural Resources, Pietermartizburg. (available at

http://www.ceh-wallingford.ac.uk/research/ winners/literature.html).

Paton, C. 2002. Ties to the land no longer bind. Sunday Times Monitor. Sunday Times

Newspaper, 4 August 2002.

Peters, C.R. 1988. Notes on the distribution and relative abundance of Sclerocarya birrea

(A.Rich.) Hochst. (Anacardiaceae). Monograph in Systematic Botany of the Missouri

Botanical Garden 25: 403–410.

Salick, J., A. Biun, G. Martin, L. Apin and R. Beaman. 1999. Whence useful plants? A direct

relationship between biodiversity and useful plants among the Dusun of Mt Kinabalu.

Biodiversity & Conservation 8: 797–818.

Shackleton, C.M., J. Botha and P.L. Emanuel. 2003. Productivity and abundance of

Sclerocarya birrea subsp. caffra in and around rural settlements and protected areas of

the Bushbuckridge Lowveld, South Africa. Forests, Trees & Livelihoods 13: 217–232.

Shackleton, S.E. and C.M. Shackleton. 2005. The contribution of marula (Sclerocarya birrea)

fruit and fruit products to rural livelihoods in the Bushbuckridge district, South Africa:

Balancing domestic needs and commercialisation. Forest, Trees and Livelihoods 15: 3–24.

Shackleton, S.E., C.M. Shackleton, A.B. Cunningham, C. Lombard, C.A. Sullivan and T.R.

Netshiluvhi. 2002a. Knowledge on Sclerocarya birrea subsp. caffra with emphasis on its

importance as a non-timber forest product in South and southern Africa: A summary.

Part 1: Taxonomy, ecology and role in rural livelihoods. Southern African Forestry

Journal 194: 27–41.

Shackleton, S.E., S. den Adel, T. McHardy and C.M. Shackleton. 2002b. Use of marula

products for domestic and commercial purposes: Synthesis of key findings from three

sites in southern Africa. pp. 57. Unpublished report, Rhodes University,

172 Sheona Shackleton et al.

Grahamstown. (available at http://www. ceh-wallingford.ac.uk/research/winners/).

Shackleton, S E. 2004. Livelihood benefits from the local level commercialisation of savanna

resources: A case study of the new and expanding trade in marula (Sclerocarya birrea)

beer in Bushbuckridge, South Africa. South African Journal of Science 100: 651–657.

Thiong’o, M.K., S. Kingori and H. Jaenicke. 2002. The taste of the wild: variation in the

nutritional quality of marula fruits and opportunities for domestication. Acta

Horticulturae 575: 237–244.

Wynberg, R., J. Cribbins, C. Lombard, M. Manders, S. Shackleton and C. Sullivan. 2002a.

Knowledge on Sclerocarya birrea subsp. caffra with emphasis on its importance as a non-

timber forest product in South and southern Africa: A summary. Part 2: Commer-

cialisation, cultivation and property rights. Southern African Forestry Journal 196: 67–77.

Wynberg, R.P., S.A. Laird, Botha, J., S. den Adel and T. McHardy. 2002b. The management,

use and commercialisation of marula: Policy issues. pp. 62. Unpublished report, Centre

for Ecology and Hydrology, Wallingford. (available at http://www.ceh-wallingford.

Ac.uk/research/ winners/).

Wynberg, R., S. Laird, S. Shackleton, M. Mander, C. Shackleton, P. du Plessis, S. den Adel,

R.R.B. Leakey, A. Botelle, C. Lombard, C. Sullivan, T. Cunningham and D. O’Regan.

2003. Marula commercialisation for sustainable and equitable livelihoods. Forests,

Trees and Livelihoods 13: 203–215.

Wynberg, R. 2005. Pro-poor models of commercialisation for non-timber forest products in

southern Africa. Ph.D. thesis, University of Strathclyde, Glasgow.

173Commercialisation of multiple-use NTFP