Hunter Gatherer Resea rch 1.1 (2015) © Liverpool University Press
ISSN 1476-4261 doi:10.3828 /h gr. 2015. 2
Where goods are free but
Hunter-gatherer ritual economics in Western
Where good s are free but know ledge costs
Anthropology, University College London, 14 Taviton Street, WC1H 0BW, UK
Abstract: Forest hunter-gatherers in Western Central Africa participate in an unusual
economic system that transacts material production in a very dierent way to
intellectual production. While material goods, such as food, tools or clothing, are
generally freely given when demanded, intellectual goods, such as the right to perform
specic rituals or to receive certain remedies, are exchanged for goods and money.
These hunter-gatherer groups trade certain types of knowledge for material goods
with each other, but never trade material goods for other material goods with each
other, despite doing so with neighbouring farmers. They simply demand them from
one another. The distribution of key aspects of this economic system across linguistic
and international frontiers suggests that it is likely to have great antiquity. The hunter-
gatherer ritual system is valued for immediately producing goods. This contrasts with
cult associations among farming societies in Central and West Africa that focus on
ensuring that goods will come in the future.
Keywords: religion, ritual associations, property rights, intellectual property, sharing,
Aka, Baka, BaYaka, Bongo, Luma, Mbendjele, Mikaya, Pygmy
1. What to share and what not to share
1.1 BaYaka and Mbendjele
A wide range of Pygmy hunter-gatherers living in the Western Congo Basin
participate in the ritual economy that is the focus of this article. ese
groups include Aka in Central African Republic (CAR) and Republic of Congo
(RoC); Baka in RoC, Cameroon and Gabon; Bongo in Gabon and RoC; Luma,
1. is article was published Open Access under aCC BY license.
Mbendjele, Mikaya and Ngombe in RoC; and probably others. Participation in
this ritual economy is an important emic marker identifying ‘Bayaka’ (forest
hunter-gatherer) people as opposed to ‘Bilo’ farming people. Bilo also means
‘the uninitiated’ in Mbendjee. Although the Bongo are not BaYaka and not
active in the ritual economy I describe, they show key features of this economy
in their interactions with neighbours.
Mbendjele living in the equatorial forests of northern RoC provide the
ethnographic focus here since I have been working with them since 1994.
Similarly to other Western Pygmies, Mbendjele say that they belong to a
larger group of ‘forest people’ (bisi ndima) generically referred to as Bayaka
(often contracted to Baaka and Baka depending on the speaker). Bayaka people
are said to share the same forest hunter-gatherer ancestors, and to share
the same economic, ritual and musical systems. Mbendjele more often refer
to themselves as Bayaka than Mbendjele. Mbendjele is used to distinguish
themselves from neighbouring Bayaka groups such as the Mikaya, Ngombe or
Luma. Since some of these groups speak Ubangian languages (eg, Ngombe and
Baka) while others speak Bantu languages (eg, Mbendjele), I write this regional
ethnonym as BaYaka.
BaYaka groups still able to access good forest are well described as ‘immediate-
return hunter-gatherers’ (Woodburn 1982). Immediate-return hunter-gatherers
achieve relative equality between camp members by demand-sharing excess
production from each other, through direct individual access to the means
of production and coercion, by valuing mobility and by employing levelling
mechanisms such as teasing and avoidance to deal with attempts by others to
claim status or impose themselves.
Despite language diﬀerences and their dispersal in small camps or settlements
for much of the year, some BaYaka people, most notably young men, visit other
BaYaka groups to explore, ﬁnd work, participate in commemoration ceremonies
(eboka), establish friendships, meet potential spouses or purchase a spirit-play.
eir similar oral tradition, ritual and singing styles, sharing economy, forest
hunting and gathering and egalitarianism make such voyages possible. While
there exists the possibility of marriage relations between BaYaka groups, they
do not trade goods with one another. is contrasts with BaYaka peoples’
relations with Bilo farmers or ‘village people’ (bisi mboka) that are predomi-
nantly based on trading and exchanging goods. Most villagers refuse to marry
BaYaka, many will not eat together with BaYaka nor allow them to stay in
their homes or villages. Rivers divide the territories of diﬀerent BaYaka groups
ensuring they do not overlap, however villagers superimpose their land claims
over parts of BaYaka land.
3WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
ere is much variation in the living conditions of BaYaka groups today.
Industrial, road and market expansion into remote forest areas has drawn
outsiders in to exploit resources. Discrimination by majority groups has led
to no recognition of BaYaka land and resources, violent exclusion from large
areas of forest by conservationists, and persecution for hunting. Many now do
some farming and serve as a labour force for other groups, often in return for
alcohol and food. ese forces combine with aggressive government sedentari-
sation policies since the early 1990s, and earlier in Cameroon and Gabon, to
have a negative impact on the ability of many BaYaka groups to maintain their
autonomy, hunting and gathering lifestyle and culture.
Although there are important cultural diﬀerences between BaYaka groups
(eg, Bahuchet 1996; 2012), diﬀerences also exist within each of the constituent
groups depending on where they live. Some Mbendjele near the Central African
Republic are evangelised and although relatively sedentary do not farm. ose
living in or near logging towns may spend long periods working outside the
forest and practice regular farming. Others further south spend most of the
year in the forest, with some groups not coming out to villages for years at
a time. But in general, most Mbendjele spend about two-thirds of the year
hunting and gathering in forest camps and some part of the year near agricul-
turalists’ villages or the activities of logging companies. Although continuing
to hunt and gather, here they will also trade, labour or perform services for
villagers and others in return for food, goods, alcohol or money.
e ritual and economic system that I discuss here is similar enough, even
across ethnic and linguistic boundaries between BaYaka groups, to justify using
the Mbendjele case to elucidate the system.
1.2 An introduction to Mbendjele economics
Although familiar to hunter-gatherer anthropology, I begin by brieﬂy outlining
the system of sharing that distributes material goods so as to contrast it with
how the ritual economy distributes certain types of knowledge whose use
produces desired goods such as food and drink, and prized social products such
as music, dance and joy.
While I will outline the Mbendjele’s system for distributing material property
through demand sharing, similar practices are well-known from the work of
anthropologists such as Blurton-Jones 1987; Ichikawa 2005; Peterson 1993 and
Woodburn 1982, 1998. Demand sharing is widely recognised as a core value
and practice of egalitarian hunter-gatherers. In contrast to the donor-organised
sharing familiar to most people, where the person owning the resource
dispenses it according to their choice, demand-sharing is recipient controlled.
Potential recipients constantly demand shares of things they suspect may be
around. It is the possessor’s duty to give whatever is requested of them, rather
than being entitled to refuse the request.
For most material items, need determines who can clai m them, especially when
they are consumable. Possessing something here is more like a guardianship or
caretaker role until someone else needs it. Certain personal possessions, such
as a woman’s basket, her cooking pots and machete, and a man’s bag, his spear,
knife and axe, are recognised as belonging to named individuals, often the
person who made, found, or bought the item. ese individuals have priority
over the claims of others to the item. But when not in use by them, any of these
objects will be shared on demand with someone who asks.
Certain foods, such as the meat of game animals that may be obtained in large
amounts, must be carefully shared out (bwεdye) among all present according to
very speciﬁc rules called ekila (Lewis 2008). ese determine exactly how each
species should be butchered and to whom diﬀerent parts must be given. So
when a pig is killed, the hunter gets the heart, the men get the liver and kidneys,
a dog that participated would get the lungs, and so on. e remaining meat
must be fairly shared amongst all present or the hunter’s luck will be ruined. If
sharing is not conducted according to these rules it jeopardises future success.
Unlike meat, gathered foods such as wild yams, honey, vegetables, fruit and
small ﬁsh are dependable food sources that regularly provision camp. When
more than can be immediately eaten is gathered, the food is shared among
all present in the forest before returning to camp. Once in the camp women
prepare and cook the food and share it again by sending plates (djalo or gabo) to
the men’s area in the middle of camp, and to their female friends and relatives
at other hearths. By contrast meat is always publically redistributed on arrival
in camp before being cooked and redistributed by the women’s djalo.
In such a society all people are encouraged to contribute according to their
ability, but if you are old or physically or mentally challenged in some way and
only rarely contribute, your entitlement is not diminished. You have just as
much right as anyone else to a share of whatever comes into camp. In this sense
living in such a society is like living in a place where goods are free. If you do
not have what you need, you simply look around to see who might have it and
ask them for it. If it is a tool or object, when you have ﬁnished they or someone
else may ask you for the item again, and so it continues travelling around the
community. If it is food people will politely help themselves to the meal that
you are eating.
5WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
e principle is that if someone has something that you need just ask them
for it, and as the Mbendjele saying translates, ‘we have easy hands so we always
give’. Mbendjele adults should epitomise this quality by being generous to a
fault – so they will give away all that is asked of them even when this results in
them having nothing left for themselves. ey contrast this behaviour with the
‘hard hands’ of their villager neighbours.
Demand-sharing is not a form of tolerated theft (Blurton-Jones 1987) since
hunters voluntarily bring the meat back to camp. Nor is it indirect exchange or
generalised reciprocity (Peterson 1993) since it tends to be a limited number
of individuals that consistently bring meat back to the group most of the time
(Woodburn 1998). Such individuals are forced to share more than others, while
being denied overt recognition of their greater contribution.
eir high valuation of equality between members of the group can result in
some surprising behaviour. Men are very sensitive to who is provisioning the
camp with meat. Rather than men who hunt a lot gaining prestige or girlfriends,
they become the target of teasing and mockery if people think that the group is
eating their production too often. In contrast to models of economic behaviour
where it is assumed that those who are good producers will get recognition,
status and fame, here it is not the case. ey will rather stop hunting for a while;
otherwise they risk being cursed, or exiled if they persist. I observed this occur
in the case of a great elephant hunter (Lewis 2003). Despite repeated calls for
him to stop hunting so often, he continued. Eventually the women of his camp
exiled him by refusing to cook any meat that he produced. He now lives with
the neighbouring Luma Pygmies.
1.3 Intellectual as opposed to material property
While I have traced out the ways that material goods are shared on demand
between Mbendjele, their behaviour towards certain types of knowledge is
quite diﬀerent. Intellectual property, such as the right to sing certain songs
or perform a particular ritual, or certain medicinal knowledge or a mystical
procedure, are not necessarily shared on demand, but may be selectively traded.
ere is a cultural logic to this seeming incongruity between the way people
transact certain intellectual property and the demand sharing of material
property. Komba, the creator, made the forest for all creatures to share. e
rules of ekila that organise sharing are said to originate from this time. No
individual or species has any greater right than any other to the forest and its
resources. So, on being roared at by a silverback gorilla for camping too close,
Mbendjele men got genuinely annoyed and shouted insults at the silverback. To
them it was unacceptable that the gorilla should claim part of the forest as his
own. Similarly, they resent villagers’ claims to own forest and ﬁelds, and often
refer to villagers simply as gorillas because of this likeness.
Since Komba created all material things for all creatures to share, anyone
can take what they need, or demand it from someone who already has it. By
contrast, certain products of our own deductions, inspirations, dreams and
discoveries can belong to us. ey only exist because someone thought or
dreamed them into being. While the material world that Komba brought into
being is shared on demand as Komba wanted, people’s ideas can become subject
to exchange, negotiation and trade. It seems to be that because they are the
product of a particular person’s imagination, their creator can decide on how
they should be distributed.
Most chose to share their herbal remedies freely on demand, but some may
only do so for a fee. It seems to depend on the individual. Certainly in cases
of serious illness, the main concern is to help the patient and they will not be
refused because of a lack of funds. It is similar for certain mystical procedures.
For instance, the elephant hunter who hunted too much had been cursed to
meet gorillas when he went in the forest. I was surprised at how often he was
charged by silverbacks. e Mbendjele healer who knew the remedy to this
curse began by demanding several thousand francs payment to provide it. In
the end he settled for a 1000 CFA and a handful of cigarettes on the day he
made the special liana-string necklace to protect the hunter. Interestingly, once
a payment has been made it is subject to demand sharing by those present just
like any other item. Only by quickly hiding the payment will the recipient have
any hope of keeping some for later.
e most valued and highly appreciated knowledge that BaYaka trade are
the ritual procedures, song repertoire and dance styles associated with a
particular forest spirit (mokondi), generically referred to as mokondi massana,
literally ‘spirit-play’. Only a spirit guardian (konja mokondi) has the right to
call a spirit to play, and if you are not the founder you can inherit or buy this
right. Other types of knowledge that are sometimes traded are herbal remedies,
particular mystical techniques and procedures such as divination or puriﬁ-
cation procedures, but this is not systematic. Only spirit-plays are rigorously
7WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
2. Technologies of enchantment
Each spirit-play (mokondi massana) is in eﬀect a ‘technology of enchantment’
in the sense meant by Alfred Gell (1999). Following a set structure they deploy
speciﬁc techniques whose overall aim is to enchant, seduce or otherwise cast
a spell over the camp and the forest. ere are many diﬀerent ways of playing
with spirits, but each is capable of producing the kind of wonder that we
experience as aesthetic appreciation in a similar manner to that which Gell so
insightfully observed in our reaction to great works of art:
the diﬃculty I have in mentally encompassing their coming into being as objects
in the world accessible to me by a technical process which since it transcends my
understanding, I am forced to construe as magical. (Gell 1999:169)
When the techniques associated with a particular spirit-play are performed
faultlessly, they produce this mystiﬁed sense of delight and wonder. However
they also build on this to go further at peak moments during the ‘play’ to
produce euphoric, trance-like states in participants. It is the euphoria or joy
(bisengo) of these moments that people value so highly and are concerned to
freely share, but not the techniques employed to produce them.
From this perspective each spirit-play is a set of techniques which combine
to become a technology of enchantment: a skill-set that once understood and
mastered enable participants to establish a situation in which all experience joy
and communion. Each spirit-play contributes to an economy of joy – a system
distributing practices and knowledge that ensure particular euphoric states are
repeatedly produced and available to all present. Spirit-plays serve the whole
community in this way. Like ‘love’, the English word ‘joy’ covers a wide range of
emotional combinations and conditions. Bisengo does too.
Each spirit-play has its own characteristic style that creates a diﬀerent
quality of joyful experience. During the total darkness of no-moon Malobe, for
instance, ﬁres are put out and torches forbidden, participants huddle together
in the middle of camp with their legs resting on their neighbours’, and their
voices intertwine in complex polyphony until tiny luminous dots ﬂoat into
camp producing a calm, wondrous and expansive joy. In the pitch black partic-
ipants melt into one another and the forest. During Sho by contrast, the joy is
more insular, remaining with the male initiates who make the ground shake
as they stamp and dance in tight formation up and down the camp. Joy is very
masculine, powerful, strong, fearless, earthed and bassey. Although frightened,
women are pleased that their protectors are so impressive and fearsome. Ejεngi
is quite diﬀerent. Ejεngi produces the ‘frisson’ of feeling safe in the presence
of a frightening and dangerous animal, together with an erotically charged joy
generated by sexy symbolism and dancing, gendered exhibitionism, seductive
playfulness and excessive consumption. Other spirit-plays, such as Enyomo or
Monano, produce a relaxed joy by blending clowning humour with virtuoso
singing and dancing. When witnessing these rituals as an uninitiated person,
the wonder so central to enchantment can be strongly experienced.
From the inside, initiates know when to deploy the various techniques
of enchantment once the spirit guardian has called the spirit to play. ese
techniques include speciﬁc secret lore, songs, dances, drum rhythms, choreog-
raphies and costume styles and are jealously guarded. ey are techniques of
enchantment in the way Gell intends because when performed well they leave
the uninitiated wondering how the forest spirit and its attendant rhythms,
songs and dances manage to create such exquisite beauty and joy.
But spirit-plays are also diﬀerent from the paintings and sculptures Gell
discusses. ey go beyond the object of the spirit’s clothing or mask, and seek to
enchant many senses, using strange sounds, stirring sights, beautiful songs and
dance movements, humour and parody, touch and smell, emotions and desires,
of trance and overlapping percussive rhythms. I think the translation ‘spirit-play’
is apt because these rituals seduce non-physical entities (spirits) from the forest
in order to establish something non-physical (spirit) in the sense of an uplifting
or joyful atmosphere. Mbendjele say people, animals and the forest will feel this.
ese technologies of enchantment are by far the most valued and expensive
items that BaYaka will spend money on. Returning from a long journey
in distant parts of the forest as the guardian of a new spirit-play is highly
appreciated. While Toma’s younger brother was working for a logging company
prospecting for trees he bought the spirit-play of Enyomo for an anvil (costing
about 120 Euros) and cash equivalent to about 150 Euros from another
Mbendjele co-worker. When one considers that these men probably earned 70
or 80 Euros a month, the price paid represented four months’ worth of wages.
e only other item of such value that Mbendjele would consider buying
would be a shotgun, then costing around 150 Euros. But since shotguns can be
borrowed from other people fairly easily Mbendjele men prefer to spend their
hard-won earnings on the rights to perform a ritual.
3. A regional ritual economy
Trading in the rights to perform these spirit-plays has set up a ritual economy
that stretches across Mbendjele forest, and beyond it to other Western Pygmy
9WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
groups in Republic of Congo (RoC), Cameroon, Gabon and the Central
African Republic (CAR). Like other BaYaka, Mbendjele love to discuss
which spirit-plays diﬀerent groups have, comment on the speciﬁc techniques
of enchantment employed, and compare each other’s performances. Every
Mbendjele clan dances ancient forest spirits (mokondi) like Ejεngi, Niabula,
Ngoku and Ye le. e majority of BaAka clans dance Ejεngi, Ngoku and Yele,
most Baka clans dance Ejεngi, Niabula and Yele, Mikaya dance Ejεngi and
Yel e . Luma dance Ejεngi.
Among the Southern Mbendjele, I collected information on over 20
diﬀerent spirit-plays, each with its speciﬁc songs, dances and secret knowledge.
Tsuru describes 53 spirit-plays among the Baka living along the Yokadouma
– Mouloundou road (1998:54–5). While the Baka tradition is possibly the
most creative, the Mbendjele system seems the most conservative because it
continues to dance all of the most widespread spirit-plays: Ejεngi, Niabula,
Ngoku and Ye le, and is especially appreciated. In CAR, Kisliuk (2001) describes
how BaAka walk to northern RoC to buy spirit-plays from the Mbendjele.
Louis Sarno (personal communication 2014) explained that Mbendjele in
Lobaye CAR visit Mbendjele in RoC to buy spirit-plays, and loved to watch
footage of spirit-play performances he ﬁlmed in RoC. ey integrate what
they see, and revive forgotten elements to enhance their own performances.
In Cameroon, building on Joiris’s work (1996; 1998), Tsuru (1998; 2001)
describes the key elements of this ritual system among the Baka. Authors
working with Baka refer to the spirit-plays as cult or ritual associations.
ey describe almost identical structures around each spirit-play – the forest
spirit, initiates, their secret path (njanga), the spirit guardians and modes
for transacting spirit-plays by purchase or inheritance. Using their own and
Tsuru’s work, Furniss and Joiris (2011) describe the process by which Baka
re-combine key musical and costume elements in the generation of new spirit-
plays in a constant but structured innovation process. Although Baka in Gabon
participate directly in this ritual economy, Gabonese Bongo are not reported to
do so. However, the ethnography suggests they value these forms of intellectual
property in a similar way to BaYaka. ey copy, adopt and adapt other people’s
ritual and song repertoires while jealously guarding their own repertoires for
themselves (Bonnehomme et al 2012).
e following map shows the minimum area in which forest spirit economies
10 JEROME LEWIS
Figure 1 Map of the area where spirit-plays are traded
Barnard’s (1988) analysis of Khoisan religion in Southern Africa described a
structural similarity across the region despite great diversity in content. As
described here for the Mbendjele, and by others for the Baka and BaAka, the
forest hunter-gatherers’ ritual system becomes apparent at the regional level.
While the rules of participation, dancing and singing styles, songs and the
clothes given to the forest spirit may all vary, the underlying structure of spirit-
guardian, forest spirit, initiates, secret area and musical performance remains
remarkably stable, crossing linguistic and international boundaries. is wide
distribution is testimony to substantial networks of interaction between diverse
BaYaka groups that goes so far back in time that participating groups now speak
diﬀerent languages and interpret the same spirit-plays diﬀerently.
11WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
4. The social organisation of BaYaka spirit-plays
4.1 Spirit-plays (mokondi massana) and spirit guardians (konja mokondi)
I am initiated into the spirit-plays practiced by the Mbendjele I have been
visiting since 1994. In 1997 I became a spirit guardian of Ejεngi, and Ingrid,
my wife, became a spirit guardian of Ngoku and Yele. Since then I periodically
coordinate the initiation of young Mbendjele boys into the Ejεngi spir it-play.
While being a spirit guardian (konja mokondi) is one of the most diﬃcult
things I have ever done, it has given me privileged insight into the way the
Mbendjele speak about and participate in the ritual economy existing between
BaYaka groups. Here I present the ritual economy, as much as possible, in their
own terms since I am familiar with a wide range of their forest spirits, the
diversity and creativity within each spirit-play, and as I participate over decades
in the same spirit-plays, also the durable structures organising them despite
innovation and change.
Every spirit-play is based around a forest spirit (mokondi) who must be called
out of the forest to a secret path and seduced to dance among people. Some
mokondi are given clothing so that they can materialise in public before the
uninitiated (eg Ejεngi), some are simply embodied by the initiates (eg Ngoku),
others only heard (eg Niabula). Each forest spirit is of a particular named type
and associated with speciﬁc ritual procedures, rhythms, dance styles and songs
that it ﬁnds irresistible and serve to draw it to the initiates. e forest spirit
must be cared for appropriately by the initiates. is always involves song and
dance and sometimes preparing clothing for the forest spirit. e initiates must
ensure that the ritual follows the correct procedures in order that the forest
spirit is drawn into the human group and so generates the pleasurable-euphoric
states associated with its spirit-play. Access to the spirit-play’s secret path is
governed by initiation for a fee paid in consumables such as honey or alcohol,
and often just money, though metal items were used in the past.
Each class or type of forest spirit is named. us Ejεngi refers to the class
of forest spirits called Ejεngi, the society of initiates who call and organise
an Ejεngi spirit-play, and the ceremony in which Ejεngi is called to play. Each
clan has its own named Ejεngi and so there are many individual Ejεngi. Each
individual Ejεngi is named and has a recognised spirit-guardian (konja mokondi)
who is responsible for calling it from the forest. For instance, my Ejεngi is called
Mikana and only I can call him from the forest.
e spirit guardian oversees the work of the initiates to create the spirit-play
and sustains its performance over the required period. Tsuru describes the
12 JEROME LEWIS
spirit guardian’s role among Baka as denoting ‘the personal right to keep a
special relationship with me and to organise a ritual association for the spirit.
is right can be shared with other persons through inheritance, gift exchange
and even by purchase’ (1998:54). is applies equally to other BaYaka.
ough the spirit guardian is generally a clan elder she or he has no authority
beyond the immediate task of organising the spirit-play. Even here, their
role is to encourage, never to coerce. Initiates in the spirit-play association
constrain the spirit guardian with mockery and outright rebuttal if she or
he is unreasonable. In addition to their ceremonial tasks, spirit-guardians
organise much of the economic activity around the ritual. ey negotiate for
alcohol, food, tobacco and marijuana from whomever they can on behalf of all
participants, and share it out appropriately. ey oversee practical tasks such
as preparing drums, ritual medicines or costumes. Becoming a spirit guardian,
as I know, is a burdensome and challenging role that is considered more of a
chore than a privilege. Of course when the spirit-play goes just right and the joy
spreads amongst all, it is very satisfying to the spirit guardian and initiates that
made it happen.
Although the spirit guardian receives initiation fees, she or he is obliged to
share these items out with all the initiates present, including the neophytes that
just paid it. So although the spirit guardian gets a fee for the use of his or her
intellectual property, as soon as the fee is received it becomes subject to the
rules of demand sharing and is distributed among all present. In this way the
fees serve to fuel the celebration by being spent immediately on consumables.
is may be similar to Hadza gambling or San xharo gifting which both serve
to circulate valuable objects among the broader community. e past focus
on iron objects suggests it may have served to circulate them among BaYaka.
Today initiation into spirit-play serves to ‘pull out’ money hidden in people’s
pockets so that it can be converted into consumables to fuel the celebration.
4.2 Forest spirits (mokondi)
Mokondi are a type of forest spirit that normally remain in the forest. Human
and other spirits are generally referred to as molimo. Uncaptured mokondi are
wild, temperamental and potentially deadly. ey have mostly humanoid forms,
from tiny pale gangly creatures to weird monstrous beings with feet facing
backwards and wild hair. Mokondi can be found and captured physically in the
forest, or in dreams (Tsuru 2001 reports similar methods among Baka). Once
a mokondi is captured and its songs discovered, it can be shared with other
13WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
Mbendjele through initiation, passed on by inheritance or sale, or even stolen.
e following ethnography will provide examples.
Spirit-guardians buy, ‘capture’ or ‘ﬁ nd’ and make pacts with mokondi that
‘eat’ singing and dancing in return for helping people. Such mokondi can assist
in the food quest, protect their initiates from harm and danger, transform
negative emotions and tension in the camp into laughter and co-operation,
cleanse bad luck, protect from sorcery, and bring people together in extraor-
dinary harmony to generate euphoric states. A lively 40 year-old family man,
Minjembe, explained to me some of the ways that diﬀ erent spirit-plays help the
Women, when they sing Yele and open the camp, they open the camp, yes! When
they open the camp there’s no chance of trouble. When they send it out like that
[sing and dance a beautiful Yele], soon animals will die. ‘Go [women say to men],
you’ll eat food, your camp is open!’
It’s like this with Malobe too. If you see Malobe dancing, then you know when the
men go in the forest, food will die.
Ejεngi is diﬀ erent; he’s not the same. If they dance him, maybe they’ll kill meat.
Figure 2 A painting of the forest spirit Ejεngi by Mongemba, Indongo 1999.
14 JEROME LEWIS
Really its when they have already killed lots of food. en yes, they dance Ejεngi.
When people are joyful because of food … Ejεngi gives thanks for things that
have come, when a joy of joys overwhelms you. When people from far hear Ejεngi
they know things are there at that camp. ‘We’ll eat food today, let’s follow them!’
(Minjembe, 40-year-old Mbendjele man from Ibamba, May 1997)
e initiates ensure the mokondi are treated correctly and in return enchant
participants. All the important preparation for spirit-play takes place on the
secret path (njaηga) where great attention is paid to detail, procedure and best
practice. It is to the njaηga path that the mokondi is called from the forest by
its spirit-guardian to be prepared for the public performance. e secret path
is exclusive to the initiates, and they are prepared to violently protect their
4.2 The neophytes (mboni) and initiates (bangonja)
e initiates are crucial to each spirit-play since they make all the necessary
preparations for the massana. All initiates had ﬁrst to be neophytes (mboni)
and pay to be initiated. Each massana has its own procedures and requirements.
ere is no ﬁxed time for initiations to occur, it depends on an individual’s
circumstances at the time of a ceremony. Since the diﬀerent spirit-plays are
danced at very variable intervals, neophytes may be of varying ages.
Initiation into spirit-plays is seen as necessary for learning about men’s or
women’s speciﬁc powers and abilities in relation to society, spirits and the
forest. us Ejεngi initiation, among other things, gives the neophytes Ejεngi’s
eyes (diso ua Ejεngi) that help initiated men to see trails, tracks and animals in
the forest, and gives them the ability to avoid large charging animals such as
buﬀalo or elephant. Similarly young women who get initiated into Ngoku will
begin learning about sexuality, feminine power and how to control Mbendjele
Initiations can occur whenever a spirit-play is performed. is will always
involve a fee to be paid to the spirit-guardian and generally an initiate to vouch
for the neophyte’s ability to handle the initiation. Fees can be paid in kind
(alcohol, meat or honey for men, and stingless bee honey (koma) and wild yams
for women) or in cash. Part of the fee is given to the forest spirit and the rest
is shared and/or consumed by all initiates present. e Mbendjele are proud of
their ability to make such fun produce so many desirable goods.
To ensure the correct atmosphere is created for the forest spirit to come
among them, the initiates begin by vivaciously animating the singing and
15WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
dancing to build enthusiasm among participants. is is a skilled task and
people good at it are appreciated. e mokondi will not come until its preferred
rhythms and songs are being performed faultlessly. e polyphonic Mbendjele
singing style demands a high degree of co-ordination. If one section falls away
or loses time it immediately spoils the song.
When done well, the music takes on a life of its own as the co-ordination
between the singers reaches astounding synchronicity. e eﬀect is synergistic
and euphoric. People dance out from the group and trance-like states begin to
engulf participants. It is at this stage that the forest begins to yield its charms
and small leafy parts of it begin to enter into the horseshoe-shaped group of
singers in camp. As the forest spirits repeatedly move into the horseshoe of
singers and back out into the forest again it is as if the forest is making love
to the camp. It is at this stage that the distinctive euphoria of the spirit-play is
most strongly experienced.
e trance states of mokondi massana do not produce the violent shaking
or loss of control reported among !Kung healers (eg Katz 1982; Low this
issue:27–57). Instead they are more euphoric or ecstatic, producing exuberant
and spectacular dancing or singing that can last many hours or even days.
is is referred to as ‘bisengo’, which I translate as ‘joy’. Initiates enable the joy
produced by the forest spirits to be shared with the whole community in this
way. With so many diﬀerent forest spirits each with the potential to provoke
euphoric states, each spirit-play constructs a particular economy of joy – a
unique means to produce and then distribute joy among participants.
5. Mokaba and Monano
Although the most important and widespread spirit-plays, such as Ejεngi, are
said to be as old as society, many minor ones have been more recently acquired.
e history of Monano, the most recently captured mokondi in my research
area, illustrates the life-cycle of a spirit-play.
Mokaba was a ﬁfty-year-old man from Ibamba. As a small boy he got
poliomyelitis that severely wasted both his legs. He walked on his hands, with
his stunted and bony legs crossed tightly together dragging under him. Despite
this, Mokaba travelled widely in the forest, swimming across the numerous
marshes, and speared trapped animals. He was widowed three times in his
short life and is survived by one remaining daughter. Mokaba was a great
singer and song composer (kombo), as well as the most respected Mbendjele
craftsman and blacksmith in the area.
16 JEROME LEWIS
Whilst camping in deep forest around 1985, Mokaba captured a forest spirit
called Monano using a special ﬁbre string (mokodi). He showed Monano to the
other men in camp. Monano was an instant success. When the uninitiated see
Monano dance his body is covered in cloth, he wears socks on his feet, has no
arms and his face is a wooden mask. Like all spirit-plays, Monano has his own
secret njaηga path and access is governed by initiation.
Over the next few years Monano was often danced and many were initiated,
including Mbendjele visiting from other regions. One such man, Samba, stayed
in Ibamba forest for several years doing brideservice. When Samba returned to
his home area of Minganga his good friend Njulle, the elder of a big group of
Mbendjele from Minganga, had to organise a large commemoration ceremony
(eboka). As the guests arrived and time went on, Njulle (the spirit guardian)
found it diﬃcult to produce enough drink and food to fuel the ceremonies.
People were complaining. Samba came to Njulle’s aid by announcing that he
Figure 3 Mokaba
17WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
had captured a new forest spirit called Ekatambili. But Ekatambili was exactly
the same as Monano despite its changed name.
By launching Ekatambili during a major commemoration ceremony Samba
generated enough money to buy food and drink to sustain many days of spirit-
play. Every eligible man present who wanted to participate needed initiating.
is ‘pulled out money’ (ulua mbongo) hidden in peoples’ pockets. is is the
key method by which spirit-plays generate goods.
When the Mbendjele of Ibamba heard the news they were furious. Samba
and Njulle had stolen their mokondi! As mokondi thieves, any insult or wrong
done to Mbendjele of Minganga was now justiﬁed. Insults ﬂew! When Njulle
heard these he composed a Malobe song to insult Mokaba: ‘Mokaba a mu toηga
yε, a pia mambi na mabo!’ (Mokaba got pierced [by sorcery], he takes faeces
in his hands!). Since Mokaba walks on his hands, so he must put his hands in
Mokaba was on a long journey in deep forest when he heard the song during
a Malobe spirit-play. Mokaba was shocked ‘He sings about me like this! We are
both invalids! So, they shall sing further!’ Since Njulle was born with a tiny
malformed right arm attached to a full hand Mokaba responded with another
song. ‘Njulle a bukia obo na mitambo a bunjia’ (Njulle broke his arm in a ﬁbre
Njulle was infuriated and threatened Mokaba and his group with violence.
e song battle continued and the two groups insulted and oﬀended each other
whenever possible. In 1995 a large brawl took place in the main logging town,
and another potentially more serious aﬀair, involving dozens of men on either
side, was narrowly averted by my intervention a year later. Some of Mokaba’s
kinsmen were travelling with me while I did censuses in the Minganga area.
ey used Njulle’s secret njaηga path as a toilet. Njulle demanded they pay a
ﬁne. ey refused and explained the problem to me. As aggressive shouting
of claims and counter-claims increased the tension, mass violence seemed
imminent, so I paid the ﬁne.
When I returned to Mokaba’s home area in 2000 I was very sad to learn of
his death from food poisoning in 1999. His two brothers inherited Monano’s
leather bag (ngamata). As they moved between forest camps they carried the
bag with them and carefully kept it out of the rain in their huts.
Returning again in 2003, we performed Ejεngi and people came from all
around. On the third day, towards the end of the ceremony, the men summoned
Mokaba’s brothers onto the njaηga path. News had come in from the forest that
Monano’s leather bag (ngamata) had been left in an abandoned camp. A party
of logging workers came through the area, found the camp and opened the
18 JEROME LEWIS
bag. Its contents were found on the ground. Monano had escaped back into the
forest. e men considered this a great tragedy. It took many years to recover
the situation. In 2011 I saw Mokaba’s brothers again and asked about Monano.
ey explained that they had recaptured him and now Monano danced again.
6. Mokondi forest spirit economies
e movement and dancing of spirit-plays connect, identify and distinguish
Mbendjele and other BaYaka over a wide area. Old spirit-plays, such as Ejεngi
and Yele, are widely distributed among BaYaka groups and so rarely traded.
However, newer spirit-plays are unevenly distributed between clans living in
the same forest area. ese newer spirit-plays have a variety of histories. Some
were traded from other BaYaka (Mbendjele say they got Malobe and Niabula
from Baka), some came from other Mbendjele clans or communities and
others, such as Monano, were captured in the forest.
Introducing a new spirit-play to one’s clan after a long journey (molongo) is,
like smoked meat or ﬁsh, one of the spoils of the journey brought back for others
to enjoy. Like Samba who stole Monano, a man who travels to do brideservice
or to work outside his traditional forest may learn about new spirit-plays. If
accepted by the original spirit-guardian, he can legitimately become a spirit-
guardian. He must give the original spirit-guardian all he demands, principally
metal goods such as small anvils, coils or spear blades, but also wine, food,
cloth and money. If he satisﬁes the spirit-guardian, he will be initiated and
becomes a legitimate spirit-guardian himself. en he can begin producing joy
by initiating others and using the spirit-play to ‘pull-out/bring into the open’
(ulua) more goods and money.
While also responsible for procedures, spirit-guardians are expected to
‘pull-out money’ (ulua mbongo) by ﬁnding initiates for their spirit-play. Money,
unlike meat or honey, can be well hidden and its existence denied. By obliging
or pressuring eligible young men to become neophytes, their elder siblings and
parents are forced to ‘pull-out’ their money to pay the fees. Since the money or
goods received are shared among all participants to fuel the spirit-play, no one
individual will beneﬁt more than others.
Bringing money and goods out into the open is an explicit objective of spirit-
plays and one of the main ways Mbendjele have made ritual an immediately
productive activity. Spirit-plays are also used in this way to pull-out goods from
neighbouring villagers. Due to their status as ‘ﬁrst people’ BaYaka perform
spirit-play rituals at all of the villagers’ most important ceremonies. ese
19WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
ceremonies are the main arena for villagers’ inter-clan status competitions. It is
important for inter- and intra-villager claims to prestige and status that large
numbers of Mbendjele perform spirit-plays during their rites, especially during
weeklong commemoration ceremonies (matanga).
Between villager clans, status is demonstrated by conspicuous consumption
and by sustaining large numbers of dependents on the ‘wealth-in-people’
model described by Jane Guyer (1993). A well-attended villager commemo-
ration ceremony, with impressive spirit-plays and abundant supplies of food
and drink is the mark of a strong clan. Villagers organising such ceremonies
have to provide whatever the Mbendjele demand in return for performing
their spirit-plays. Mbendjele use this dependence to extract vast quantities of
goods from villagers. Villagers often complain about this. ey said that other
villagers tease them if Mbendjele stop a ritual too early because all the alcohol
and tobacco had been consumed.
Since villagers will never be able to buy the rights to perform the spirit-plays
or initiate neophytes² themselves, the Mbendjele maintain a monopoly on these
proﬁtable performances. e Mbendjele are proud that the villagers depend
entirely on them for the proper performance of their major rites.
Mbendjele also do commemoration ceremonies (eboka) for themselves.
Due to their immediate-return economy, like Njulle mentioned earlier, it is
challenging to provide for their guests. Although many clans will organise
net-hunting expeditions and tap palm trees for wine in order to have at least
something to oﬀer, the requirement to give whatever guests demand is always
diﬃcult to fulﬁll. In this context having some new spirit-play rituals to perform
will reduce the burden by generating desirable goods such as alcohol, or money.
Introducing a new spirit-play involves initiating as many eligible people as
possible. Neophytes will pay an initiation fee. If not in kind, fees are immediately
spent on alcohol and other consumables to fuel the performance. During
large commemoration ceremonies several dozen neophytes can be initiated
generating signiﬁcant amounts of consumables for several days of performance.
is is an important source of sustenance during these long ceremonies.
Another way new spirit-plays generate goods and money is that they will
attract curious villagers to come and watch. Mbendjele deftly play up villagers’
2. ere are two exceptions: Mabonga and Lota. Lota is an Ejεngi from Minganga and the
story can be read in Lewis 2002:173 fn 148. Sangha-Sangha villagers stole Mabonga from the
Mbendjele of Ibamba by force sometime before the 1960s, and have suﬀered the mystical
consequences ever since. When some neophytes went mad after initiation in the late 1980s, the
villagers stopped performing the ritual until 2005 when a young man usurped the chieftaincy
and began holding Mabonga ceremonies to intimidate his opponents.
20 JEROME LEWIS
claims to superiority by drawing them into status competitions with one
another. Spirit-guardians must manipulate villager ‘big-men’ to contribute
something to festivities by telling them how much another villager big-man has
already given. If done with charm and skill, a smart spirit-guardian will extract
huge amounts of alcohol and manioc.
If the hosting clan entertains their guests with a wide variety of spirit-plays
this is appreciated and will be a favourite topic of conversation for some time
to come. However, if the hosting clan does not own the spirit-play that their
guests are expecting, they will be obliged to ﬁnd and possibly pay a spirit-
guardian to call the spirit-play for them. is can become a source of conﬂict
since Mbendjele have so little they are often tempted to do the ritual without
following this protocol. is may simply involve doing the spirit-play without
an owner present, or more seriously, as Samba did with Monano on Njulle’s
behalf, by changing its name and claiming to be its discoverer so as to initiate
Past conﬂicts over non-payments when one clan was discovered to have
danced another’s spirit-play led each clan to get its own mokondi. Emeka said
that the ancestors have bought and sold spirit-plays since ancient times. He
knows which spirit-plays his ancestors bought and from whom, as he knows
which people his ancestors initiated as spirit-guardians. Some of these transfers
occurred several generations ago. Like Emeka, most elders know who legiti-
mately owns what.
is ritual economy of forest spirit-centred relationships are carefully
remembered and form a key part of the way Mbendjele identify, discuss and
judge the extent to which other Pygmy groups are real ‘forest people’. Spirit-play
rituals are discussed in terms of the accomplishment of various key aspects of
the performance – the quality of the singing and dancing, the appropriateness
of the forest spirit’s behaviour and the success of the dance formations charac-
teristic of the particular ritual whose performance is being appraised.
Figure 4 illustrates these for Malobe and Enyomo. Malobe has become the
most appreciated spirit-play ritual that the Mbendjele bought from the Ngombe
(Baka). e Ibamba Mbendjele got Malobe when Dito Dzelle bought it from the
Ngombe ﬁve generations ago. Since then it went from Ibamba up the Sangha
River north of Ouesso, and into the interior to the Terres des Kaboungas and
further north to Mbandza. A spirit-play called Malobe is danced on the Motaba
River, but with very diﬀerent dance formations and melodies.
e mokondi Enyomo is not at Ibamba but travelled South from Mbandza
only as far as Terres des Kaboungas and North to the upper Motaba River and
some parts of the Ibenga River. In 1997 I participated inadvertently in part of
21WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
Figure 4 Map of forest spirit mokondi journeys
22 JEROME LEWIS
this process because the elder, Toma, who was the ﬁrst Enyomo spirit guardian
in Makao on the Motaba River, travelled with our party back to his home area
with Enyomo. Upon arrival he began initiating his clan’s people, and many
others too. A great Enyomo was danced for three days. Toma was perceived as
exploitative by some, since he cashed in on all the new initiates when it was
his younger brother who had bought Enyomo. Toma, being older, claimed the
right to be its guardian for the clan. Lewis 2002 (Appendix 1) provides some
of the oral history concerning mokondi trading between the Ngombe and the
Mbendjele over the past century and possibly longer.
7. Elephant hunting and spirit-play
In contrast to the eboka or matanga (Lingala) commemoration ceremonies just
described, elephant hunting combines many types of spirit-play rituals in a way
that illustrates the main dimensions of these economies.
Here the great social capital of the elephant’s meat is carefully surrounded
by spirit-play so as to undermine any claims to status by the hunter by dividing
responsibility for his success with the women. is follows the rule that the
person who spears an animal ﬁrst is named it’s hunter, someone who speared
it second is called mokobia even though it may have been their blow that killed.
In the context of big game hunting women are hunters, men are mokobia.
Occasionally an elephant or its fresh tracks are encountered by chance,
but in ideal circumstances Yele, a women’s spirit-play, precedes the departure
of men to go elephant hunting. Women sing Yele songs, drink a herbal brew
and entering trance ﬂy over the forest. When they see large game such as an
elephant they ‘tie up’ the elephant’s spirit. In the early morning they tell the
men where to ﬁnd it. In eﬀect women catch the elephant ﬁrst. is is why an
elephant hunting journey is called ‘mwaka ya baito’ – ‘a women’s hunting trip’,
although no women accompany the men.
Men then depend on using both mystical and physical skills cultivated
and reﬁned during spirit-plays such as Niabula and Ejεngi to supplement the
great skill they have developed in tracking and stalking to get to the elephant.
Approaching close enough to shoot or spear the elephant requires secret
knowledge acquired on the njaηga path in addition to an intimate knowledge
of one’s fellow hunters and the prey’s habits, precision in aim and movement
and a signiﬁcant dose of courage. In line with their egalitarianism, when in
conversation men emphasise the help they get from these forest spirits rather
than their skill.
23WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
While the men are out hunting the elephant the women may continue in the
characteristic Yele deep trance posture. Women rest their folded raised arm on
the top of their crocked head, which bows forward as if weighted down by the
arm. I was told that they continue to rock rhythmically back and forth as they
sing the Yele songs until the forest spirit Moshunde ﬂies through the forest to
tell them the men have killed, and leads them to the place.
e spirit-play called Malimbe occurs at the site of the kill. Malimbe uses
children as his emissaries to demand whatever he needs (normally meat) from
the wife of the elephant hunter (mwito ya tuma). Eya is called in after dusk to
mark the death of the elephant. e Eya forest spirits begin a raunchy, sexually
explicit and sometimes simply rude conversation with the camp in squeaky,
screechy voices, which the singing women saucily reply to.
During feasting Yolo will be sung, celebrating the abundance of meat, and
later Eya again for some more raunchy provocation. Feasting, dancing and
sex go well together. Until the majority of the meat is consumed, spirit-plays
and feasting continue, sometimes involving the men in all night sessions of
Niabula, the elephant hunters’ spirit-play.
ese spirit-play rituals serve to circulate the meat as widely as possible within
the camp as well as with other camps. When people in neighbouring camps hear
the drumming drifting over to them at night, they know there is something
there for them and come to visit. Any extended feasting is accompanied by
spirit-play. If suﬃcient people are present Ejεngi is the spirit-play of choice.
To enjoy the great abundance provided by killing an elephant many types of
spirit-play come together.
8. Conclusion: a regional system
Hunter-gatherers’ spirit-plays are sophisticated, many dimensioned aesthetic
achievements. ey oﬀer music and dance to establish a dialogue with the forest
mediated by the mokondi forest spirits. Participants share and are stimulated
by the humour, poetry and beauty of the songs, wowed by the aesthetics of the
performance and the insight and wisdom of the teachings of the njaηga path,
and irresistibly drawn to experience the solidarity and community created
among people normally widely dispersed in small camps across the forest.
When meeting together like this, spirit-plays ‘pull out’ desirable goods hiding
in peoples’ baskets or pockets, while creating and sharing out joy right away.
ey allow groups in society to communicate with each other as groups rather
than as individuals, and cultivate attraction between the gender groups while
24 JEROME LEWIS
re-aﬃrming mythical agreements. If the camp is getting on well then it is ‘open’,
food will come and life is good.
ere is huge creativity within this form, with constant innovations to the
songs and dances of established spirit-plays, as well as new spirit-plays being
discovered, or purchased from neighbours and then brought home, to be learnt
and enjoyed, and through the attraction of initiates generate desirable goods to
fuel further singing and dancing. e economic achievement of spirit-play is
to extract hidden production, distribute actual production and teach skills and
knowledge that assure future production.
e trading networks that evolve around the movement of spirit plays
across language boundaries and international frontiers unite Mbendjele with
each other and with other BaYaka groups. ey judge participation in this
ritual economy to be a key indicator of their shared identity as autoch-
thonous, egalitarian forest hunter-gatherers. is marks them out as a distinct
component of the larger West and Central African cultural area proposed by
Herskovits (1926), Vansina (1990) and others.
Despite the BaYaka sometimes being considered small-scale, isolated
and mutually independent groups with closer ties to their agriculturalist
neighbours than with each other, the ubiquity of spirit play among these
diﬀerent groups in the western Congo Basin hints at a broader shared identity
and culture. Approaching from a diﬀerent angle, Moise labels this the ‘autoch-
thonous tradition of Central Africa’ (2014). An important reason not to refer
to spirit plays as ritual associations is to emphasise the diﬀerence between the
agriculturalists’ practices and those of the autochthonous tradition.
Despite surface resemblances with spirit-play economies, the agricul-
turalist ritual associations and their ritual economies embody fundamental
diﬀerences that resemble those between ‘delayed return’ farming societies and
‘immediate-return’ hunter-gatherer ones. As with BaYaka spirit play, villager
ritual association members are organised in a cult whose access is governed by
initiation. Some cults have masked dancers or other performers who display
publicly. Others only allow their music or sounds to be heard.
Röschenthaler’s 2011 comparative study of the diversity of these cults
in the Cross River Region spanning Cameroon and Nigeria allows certain
characterisations to be made. Agriculturalist ritual associations are based
around ancestral spirit shrines, earth shrines and masquerade plays and only
transacted between land-holding, high status men and women for high status
goods such as slaves, though now iron and money. e associated perfor-
mances are intended to be visible manifestations of power and status based
around places or objects revealed by the ancestors as powerful. Some cults
25WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
perform blood sacriﬁces, including human sacriﬁce, others only pour libation
to ancestors to gain their blessings.
In village ritual associations, ancestors are the central mediating spirits
between initiates and the invisible world. rough the revelation of power
objects ancestors become the focus of these cults, and when oﬀered sacriﬁces,
mediate for the living to ensure good harvests, fertility and success in warfare.
In other words, ritual work is like farm work; invest now to harvest later.
ough the ritual eﬀort is made now, its yield will only be enjoyed in the
future. In this sense the ritual associations of villagers, like their economy, are
By contrast, hunter-gatherers’ spirit-plays are concerned with immediately
producing valued goods for all to freely enjoy, not with waiting for them to
come in the future. e hunter-gatherer spirit plays mediate between the camp
and the forest, through mokondi forest spirits not ancestral spirits, and it is
singing that feeds them, not blood. e performance of spirit plays emphasises
the equality of all, rather than the status of some.
e diﬀerent modes BaYaka groups employ to distribute knowledge and
material goods are almost an inversion of the scientiﬁc ideal that knowledge
should be shared on demand, while material property remains exclusive. As I
am doing with you now, we academics are expected to share our insights and
knowledge freely, but not our shirts, computers or bicycles.
PhD research was supported by the Wenner-Gren Foundation, an Emslie
Horniman Scholarship and the Swan Fund (1994–1997). I return regularly
to the Mbendjele area. I thank seminar audiences at Heidelberg University,
London School of Economics, Kyoto University and Vienna University and
two anonymous reviewers for their critically constructive comments on earlier
versions of this paper.
Bahuchet, S 1996. Fragments pour une histoire de la Forêt Africaine et de son
peuplement: les données linguistiques et culturelles. In Hladik, CM, Hladik, A,
Pagezy, H, Linares, OF, Koppert, GJA & Froment, A (eds) L’alimentation en forêt
tropicale: interactions bioculturelles et perspectives de développement. Paris:
26 JEROME LEWIS
Bahuchet, S 2012. Changing language, remaining Pygmy. Human Biology 84(1):11–43.
Barnard, A 1988. Structure and ﬂuidity in Khoisan religious ideas. Journal of Religion
in Africa 18(3):216 –236.
Blurton-Jones, N 1987. Tolerated theft, suggestions about the evolution of sharing,
hoarding and scrounging. Social Science Information 26(1):31–54.
Bonhomme, J, De Ruyter, M & Moussavou, J 2012. Blurring the lines
ritual and relationships between Babongo Pygmies and their neighbours (Gabon).
Furniss, S and Joiris, V 2011. A dynamic culture: ritual and musical creation in the
Baka context. Before Farming 4(3).
Joiris, DV 1996. A comparative approach to hunting rituals among the Baka Pygmies.
In Kent, S (ed) Cultural Diversity among 20th Century Foragers. Cambridge:
Cambridge University Press:245–275.
Joiris, DV 1998. La chasse, la chance, le chant. Aspects du système rituel des Baka du
Cameroun. PhD thesis, Université Libre de Bruxelles.
Gell, A. 1999. e art of anthropology: essays and diagrams. In Hirsch, E (ed) LSE
Monongraphs in Social Anthropology. London: Athlone Press.
Guyer, J 1993. Wealth in People and self-realization in equatorial Africa. Man, New
Herskovits, M 1926. e culture areas of Africa. Africa: Journal of the International
African Institute 3 (1): 59 –7 7.
Ichikawa, M 2005. Food sharing and ownership among Central African hunter-
gatherers: an evolutionary perspective. In Widlok, T & Tadesse, W (eds) Property
and equality. Oxford: Berghan Books:151–164.
Katz, Richard 1982. Boiling energy: community healing among the Kalahari Kung.
London: Harvard University Press.
Lewis, J 2002. Forest hunter-gatherers and their world. PhD thesis, University of
Lewis, J 2003. e hunter’s curse. Film in the ‘What’s going on?’ video and documen-
tation tool. London School of Economics. 7 minutes. http://elearning.lse.ac.uk/
Lewis, J 2008. Ekila: Blood, bodies and egalitarian societies. Journal of the Royal
Anthropological Institute 14:297–315.
Low, C 2015. e role of the body in Kalahari San healing dances. Hunter Gatherer
Kisliu k, M 2001. Seize the dance. Ba Aka musical life and the ethnography of per formance.
Oxford: Oxford University Press.
Moise, R 2014. Do Pygmies have a history? revisited: e autochthonous tradition in
the history of equatorial Africa. In Hewlett, B (ed) Hunter-Gatherers of the Congo
Basin. Cultures, Histories and Biology of African Pygmies. New Brunswick and
London: Transaction Publishers:85–116.
Peterson, N 1993. Demand sharing: reciprocity and the pressure for generosity among
foragers. American Anthropologist 95(4):860 – 874.
Röschenthaler, U 2011. Purchasing culture: the dissemination of cult associations in
the Cross River region of Cameroon and Nigeria. New Jersey: Africa World Press.
27WHERE GOODS ARE FREE BUT KNOWLEDGE COSTS
Tsuru, D 1998. Diversity of spirit ritual performances among the Baka Pygmies
in south-eastern Cameroon. African Study Monographs, Supplementary Issue
Tsuru, D 2001. Generation and transaction processes in the spirit ritual of the Baka
Pygmies in southeast Cameroon. African Study Monographs, Supplementary
Vansina, J 1990. Paths in the rainforest. toward a history of political tradition in
equatorial Africa. London: James Currey.
Woodburn, J 1982. Egalitarian societies. Man, the Journal of the Royal Anthropological
Woodburn, J 1998. Sharing is not a Form of exchange: an analysis of property sharing in
immediate-return hunter-gatherer societies. In Hann, C (ed) Property Relations:
Renewing the Anthropological Tradition. Cambridge: Cambridge University