-
International African Institute
Brideservice, Residence, and Authority among the Goba (N. Shona)
of the Zambezi ValleyAuthor(s): C. S. LancasterSource: Africa:
Journal of the International African Institute, Vol. 44, No. 1
(Jan., 1974), pp.46-64Published by: Cambridge University Press on
behalf of the International African InstituteStable URL:
http://www.jstor.org/stable/1158566 .Accessed: 09/12/2013 04:50
Your use of the JSTOR archive indicates your acceptance of the
Terms & Conditions of Use, available at
.http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars,
researchers, and students discover, use, and build upon a wide
range ofcontent in a trusted digital archive. We use information
technology and tools to increase productivity and facilitate new
formsof scholarship. For more information about JSTOR, please
contact [email protected].
.
Cambridge University Press and International African Institute
are collaborating with JSTOR to digitize,preserve and extend access
to Africa: Journal of the International African Institute.
http://www.jstor.org
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
[46]
BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG THE GOBA (N. SHONA)
OF THE
ZAMBEZI VALLEY' C. S. LANCASTER
INTRODUCTION
THE Shona-speaking complex of tribes covers most of present day
Southern Rhodesia, stretching to the Zambian border along the
Zambezi River and into
Mozambique in the north and to the Indian Ocean in the east. The
Korekore group is the major northern division of the Shona-speaking
peoples. The original Korekore appear to have been a band of
Karanga invaders from the south who entered the middle Zambezi
Valley in the fifteenth century and established the Mwene Mutapa
confederacy. Today the Korekore group occupies an area extending
south from the Zambezi between longitude 28? and longitude 33?
east, including the low-lying valley floor and the middle and high
veldt or plains to the south where most of the population is
concentrated (Garbett, I966). The valley floor in this region has
been considered a poor habitat by African and White settlers alike.
The climate is generally poor, there are few cattle because of the
tsetse fly, the area is economically unde- veloped, and the
population is sparse and scattered. Because of their poorer habitat
the valley dwelling Korekore have come to be distinguished from
their more for- tunate highland neighbours to the south and are
disparagingly known as Goba or poor lowlanders. In times past they
have also been known as Banyai or vassals to the stronger Shona
chiefs on the more populous cattle-rich highlands, and also as Chi-
kunda or followers of the Portuguese and half-caste traders
operating along the Zambezi in the nineteenth and earlier
centuries. This historical background of chang- ing political
relations symbolized by changing ethnic identities will be
described elsewhere (Lancaster, n.d.).
Neighbouring highland Shona have traditionally followed a
patrilineal ideology indicated by patrilocal residence for a man,
virilocal residence for a woman, and the agnatic inheritance of
social roles and most property. As compared with this, the absence
of bridewealth cattle in the valley has in time led to distinctive
changes in social organization brought about by the practice of a
form of uxorilocal service marriage (kugarira) sometimes referred
to by the younger men who must endure it as 'slave' marriage
because it is considered fit only for poor men or slaves. Kugarira
has been a standard alternate form of marriage and residence among
Shona-speaking peoples practised when the preferred bridewealth
cattle are unavailable (see Bullock, I928: 214, 355; Holleman,
I969: I24; Schapera, I929). Normally relatively rare among
cattle-keeping Shona on the tsetse-free highlands, kugarira appears
to have
I This research was supported by NIMH pre- for Social Research
in the University of Zambia. A doctoral fellowship 5 FOI MHz8688-o5
and a field grant from Rutgers University Research Council
supplement. Field-work was conducted in Zambia financed preparation
of this manuscript. The author in the Zambezi-Kafue confluence area
from March particularly wishes to thank Thayer Scudder for his I967
to March 1969 during which time the author helpful comments on an
earlier draft. was granted a research affiliateship to the
Institute
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY 47 been more common in
undesirable refuge areas like much of the floor of the middle
Zambezi valley where low human population densities have been
associated with uncleared bushlands supporting big game and tsetse
fly. Because of this, kugarira has been the standard marriage form
among the Goba. This has led to an aberrant pattern, in Shona eyes,
of matrilocal extended families where ties through co-resident
females have come to assume major importance in structuring
residential units, descent groups, succession and inheritance, and
where both matrilineal and patrilineal principles of descent seem
to be operating at once.
In the region of the confluence of the Zambezi and Kafue rivers,
the site of the present study, there has been something of a chain
migration tradition whereby small groups of Goba have gradually
crossed the Zambezi to the north where they have eventually become
assimilated as Zambian Tonga, a well-known group of matri- lineal
Central Bantu. In a previous paper the changes in Goba social
organization brought about by kugarira have been described as an
environmental adaptation pro- viding a bridge or intermediate form
between the social organization of the patri- lineal Plateau Shona
and that of the matrilineal Zambian Tonga with whom the Goba
intergrade (Lancaster, I97I). The purpose of the present paper is
to describe Goba social organization in greater detail by examining
the brideservice system and the impact it has had on the patterns
of marriage, residence, and authority.
THE BRIDESERVICE CYCLE The Goba brideservice system consists of
a prolonged cycle of gifts, payments,
and changes in rights and obligations. It is considered a time
of testing for something important, the establishment of a marriage
and independent household and, as with most major status changes in
the life cycle, the people require a long time to draw the
experience out to be certain it is acceptable. This extended period
of testing can be terminated at the many points at which love,
courtship, elopement, sanctioned co- residence, and the filiation
of children finally result in a mature marriage between adults
acting in harmony with the local community. In the meantime the
girl's parents are in a position to derive considerable long-term
social and economic benefits if they can maintain control over
their daughter and her suitor or his successors.
In the case of a first marriage of a young man and a girl the
cycle usually begins when a suitor in his middle to late twenties
makes a formal opening gift, nhumbi, to a girl who is likely to be
6 to 8. For the past twenty years the nhumbi token has been a
shilling or two.2 If the girl is willing to let matters go further
she will pass the coin on to the female head of her family
compound, her maternal grandmother or mother, who consults with the
girl and passes on the news and the coin to the girl's father. It
is up to the father to decide whether the affair can continue and
to set the terms and receive subsequent services and payments. He
can reject the boy if he dislikes him or his family or if he wants
to encourage an elopement and the sizable damages that ensue. If he
accepts him the young man becomes an acknowledged suitor
(mukwasha).
The next step has traditionally been the handing over of tsambo
which usually passes from the young man's father to the girl's
father through the young couple who
2 At the time the research was begun the Zambian pound was
equivalent to the U.K. pound sterling and to U.S. $1.40.
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
48 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG act as
intermediaries between the consenting fathers-in-laws (tshishashi).
Like nhumbi, the payment of ritual hoes and the closing gift in the
cycle which will be discussed later, the handing over of tsambo is
a long-established step in the brideservice sequence to which other
elements have gradually been added. Tsambo today consists first of
a tiny ring of small white beads obtained from the towns. But by
the 1920s the avail- ability of cash in the hands of young
unmarried men who had taken up migrant labour for Europeans
encouraged most elders to demand a few coins along with tsambo. By
the I930S and I940s the beads were generally accompanied by
5s.-Ios. as wage work opportunities increased. By the I95os men
were paying as much as ?i or Ci. ios., which level has remained
something of a standard though by the late i96os the amount might
reach C6 depending upon the suitor's ability to pay and other
personal considerations. As is true of all substantial cash
payments marking stages in the cycle, the suitor must earn most of
the cash himself. For example in thirty-eight cases collected in
1967-9, thirty-five men had earned the tsambo cash supplement from
their own efforts. Only two had received help from their father and
one from a mother's brother.
The provision of tsambo beads symbolizes the suitor's father's
or sponsor's role and acquiescence in the proceedings just as his
willingness to receive tsambo sym- bolizes the girl's father's role
and acquiescence. Most suitors accept the beads from their fathers,
indicating his role as a general patron and sponsor and the
suitor's intention of returning to his compound if possible when
the uxorilocal service period has ended. In addition, if the suitor
accepts tsambo beads from his father a member of the father's
descent group is given an opportunity to finish any out- standing
service or payments and to inherit the wife and household estate if
the suitor should die. If the father or his successor has not
finished his own marriage cycle or is not interested and is a
relatively poor sponsor, tsambo beads may be ac- cepted from the
mother's descent group instead, enabling the suitor's maternal
kins- men to control any future positional succession and
inheritance stemming from the union. In former times the
descendants of female slaves were especially likely to follow the
latter course and in that event the source providing tsambo would
again be a factor in long-term residential alignments. In a sample
of 37 tsambo exchanges recorded in 1967-9, 24 men had received the
beads from their father or his successor, 3 from their mother, 6
from a mother's brother, and 4 had purchased the beads
themselves.
The two components of tsambo today, the beads and the cash,
represent dual spheres of exchange which have been developing
throughout this century. The beads continue to represent
traditional village controls backed by religious sanctions, the
need for sponsorship, and reciprocal kinship encumbrances. While
not neglecting the cash, it is this element of the tsambo exchange
that the village elders stress as most important. The cash, which
older village men acting as sponsors are unlikely to possess in
abundance, represents the suitor's independent activities in the
external cash economy. Men who have gained some affluence from this
source prefer to be unencumbered by village ties and may seek no
help with tsambo. They also tend to make large lump-sum payments
early in their marriage careers in order to evade uxorilocal
service and retain freedom of action. Such men are still few and
are usually shopkeepers or government employees.
Once the girl's father accepts tsambo the suitor should take up
residence in his
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY father-in-law's compound and
begin his service. If the father-in-law can control him this still
includes helping with the construction of sleeping huts, field
shelters, granaries, and animal pens, the clearing of fresh bush
fields and streamside gardens, weeding, cultivating, harvesting and
other heavy labour, and errands that may be required. If his
service is acceptable the suitor becomes a recognized member of the
village and while he may not openly cohabit with his girl she may
visit him and he can claim damages or the return of tsambo at the
end of the affair if she is seduced by another man. After a
variable period of some months the father-in-law may permit the
couple to live together openly while the suitor continues his
service in residence. At this point the latter may present the
girl's mother with a small gift of thanks for keeping his girl for
him (mucheka). After the first night of sanctioned cohabitation the
suitor has a traditional opportunity to reject the girl at the risk
of damages enforceable both by the village moot and the local
government court and which today may run as high as 30o. Unlike
tsambo these damages are not returnable. Her parents are thus
likely to be richly compensated if the courtship proceeds no
further and the threat of damages tends to encourage the suitor who
has reached this stage to continue on through the cycle. By this
time pregnancy is likely and a suitor who has displeased the girl's
parents and, supposedly, her guardian ancestral spirits is likely
to be held responsible if there is a long and painful childbirth,
miscarriage, weak child, or if the child should die. In these
circumstances the suitor is frequently pressed to accelerate his
progress, and payments, through the cycle. Some payments are likely
to be en- larged and additional levies may be imposed upon the
suitor in the form of subsidies for herbalist's and diviner's fees
and offerings to appease the appropriate spirits. If the suitor is
slow or recalcitrant he may lose access to his wife, and if she
should die at this point payments for terminating the marriage and
filiating the children are likely to be higher than would otherwise
be the case. These steps will be described later.
Uxorilocal brideservice can thus be a difficult period for a
man, particularly in its earlier stages when the work is hardest
and his status most peripheral. In many cases he finds himself in a
strange village where he has few close kinsmen or family friends.
He may soon develop close ties with his wife's younger male
siblings but must observe avoidance of the older in-laws and
continue with his service and payments while his unmarried friends
continue to enjoy the freedom and adventure of town life.
Recognizing the suitor's uneasiness in a difficult situation, most
in-laws have long capitalized on the attractions of the towns by
encouraging young husbands to serve them by means of migrant labour
and the remittance of payments and gifts rather than the
performance of manual labour in residence. This monetization of the
brideservice cycle is reflected in labour migration statistics
collected in two sample villages in 1967-9. The labour migration
rate is 65 per cent among married Goba males and this rather high
rate is due to a number of factors rather than the desire to escape
brideservice alone. For example, this high rate is in line with the
experience of other Zambian peoples occupying marginal lands in
reasonable proximity to major centres of employment (see Kay,
I967). Young Goba males begin to enter the migrant labour stream at
the age of I 5 or i 6, many years before entering the marriage
cycle. The labour migration rate for males aged 5 to 19 is a
substantial 57 per cent, with some 29 per cent almost permanently
absent from their rural homes (see Table I). While some of these
younger men talk of their trips to town in terms of saving for
E
49
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
50 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG
TABLE I
Labour Migration Rates from Two Villages
27 25
21
14 2 g 0P
AGE 15-19
19
0 z
*4% 74%
20-29 30-39
20
AGE 0-39 30-39
V////A HOME ? TIME OR LESS
!-- RARELY HOME A-Unmarried. B-Damages due. C-Tsambo accepted.
D-Pfuma accepted. E-Mutsimutso accepted.
'D
11
E
50+ 40-49
16
E
D
iB
15
MI,
E E
B D
40-49 50+
___
_I __~~~~~~~~~~~mm
_
__.___
E
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY their future wives, most young
men of this age leave home for the sake of adventure and to satisfy
other cash needs. The brideservice system begins to make itself
felt on labour migration statistics when we consider men in the 20
to 29 age-group. Over 80 per cent of these men are service
husbands, 93 per cent are regularly involved in the shuttle to the
towns, and 82 per cent are almost never seen in the villages during
these years. The situation is similar for men in the 30 to 39
age-group in which 88 per cent continue with their labour tours and
as many as 68 per cent are still almost totally absent from the
local scene.
The scarcity of fit men in their prime years has had its effects
upon younger village women. Though many parents would like to keep
their daughters at home to help in the fields as in former times,
most young women want a taste of modern town life too. Some girls
accompany their kin to look for rich husbands in town, and many
young wives like to follow the men, some of whom never return or
marry women in town. Rather than undergo the old formalities,
restrictions, and hard work traditionally associated with nhumbi,
tsambo, and years of local manual labour, both sexes have found it
in their interest to escape parental supervision through elopement
to town. While Goba women never seek wage work themselves, this has
resulted in a 'labour migration' rate of some 36 per cent among
married women in general though the movement is much more marked in
the I5 to 19 age category of girls and young wives in which 74 per
cent make regular journeys and 63 per cent are almost totally
absent from their village homes. As men progress with their
payments and make their marriages more secure, women burdened with
growing families tend to spend more time at home and in the 20 to
29 age-group the female migration drops below that of the men to 64
per cent and is negligible thereafter.
Claims for elopement damages constitute a potential civil case
in the government administered local courts and must be cleared up
satisfactorily before a suitor is allowed to proceed with the
remaining marriage stages. Based upon the suitor's estimated
savings from town and his earning power from future labour tours,
damage payments have come to take a regular place in the marriage
cycle between nhumbi and tsambo. Elopement damages usually consist
of three parts. Some time not long after eloping the abductor may
make a small payment to let the girl's parents know officially who
has taken her and where she has gone (vunzirakuno). Usually little
more than a voluntary token, this may amount to ?I or ?2 and is a
nicety designed to soften later demands. After returning from town
he may then pay for having taken her 'outside the kraal' (mariye
kudzoresa mukadzi). This has also been considered something of a
token. It is paid upon returning her to her parents and should be
settled promptly for if she should sicken or die while in her
abductor's hands the suitor is subject to the pressure of spiritual
sanctions and higher final damages. In recent times this payment
has ranged from ?2 to as much as ?I 5. This is followed by the
largest single payment in the cycle, the damages for the elopement
itself (mariye murandu or mhoswa). The suitor is told how much is
expected at the time he returns the girl and is generally allowed
to pay in instalments over a period of time inasmuch as the local
courts will uphold this claim and the girl and her children are now
safely in her parents' hands while he continues his labour
tours.
As of 1967-9 most elders had paid no mhoswa in their own
marriages, having given nhumbi, tsambo, and served for periods
ranging up to 2o-5 years though the service
5z
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
52 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG
period might be shortened in the second marriage of women and in
cases where a persuasive husband held important kinship positions
requiring his return home. Thirty or more years ago damages seem to
have been limited to 5s. or a few pounds if the elopement was brief
and service was also provided. Some twenty years ago damages
generally ranged from CI 2 to ?I 5. Mhoswa now ranges from ?6 to
?45 and in normal circumstances the average villagers may expect to
pay about o20 and the trend seems to be rising. Damages remain
lower in cases involving men with impor- tant kinship positions,
good bargaining ability, plural wives, and women who are
undesirable because of age, barrenness, or a poor reputation, while
damages tend to be higher in cases involving important women, local
shopkeepers, and other wealthy men unwilling to perform lengthy
manual service in residence. Damages are never returnable. Some
young men today manage to finish their damages within a year or two
of eloping but many take much longer and eight to ten years is not
uncommon, depending upon the perseverance of the father-in-law, the
size of his claim, the suitor's health and willingness to work for
Europeans, and the state of the wage labour market. Delays during
this portion of the cycle contribute to the dissolution of many
marriages so that serial monogamy together with polygyny keeps some
men working for damages well into their forties (see Table I).
During his period of labour migration and until elopement
damages are com- pleted, a son-in-law is generally not allowed to
live regularly with his mate in her village until tsambo has been
accepted. Until that time he may arrange to visit the girl from
time to time when home from town or when she travels to see him but
cannot enjoy normal domestic relations. Many men at this stage
abandon their village wives, emigrate permanently to the towns, or
take another wife elsewhere so that through- out this often lengthy
period a man is pressured to finish his damages or risk losing his
wife and children. Only when he settles his elopement damages case
and tsambo can he move into the village as in a traditional
marriage, build a house near his father-in- law, and continue the
modern version of the service cycle. In some situations such as
illness or inability to find work, he may temporarily be allowed to
occupy an old or vacant shelter before that time but he is not
allowed to build a regular village house for that is a signal that
tsambo or its equivalent has been accepted and that damages have
been completed.3
Once the suitor has been allowed to erect a house in their
village he is expected to indicate his compliance with the wishes
of the girl's family in marital and domestic affairs by supplying a
ritual hoe (badza), plus a few shillings, 'to please her ancestral
guardian spirit' (sazita mudzimu).4 Her father and mother then
announce the marriage to the spirit community, dedicate the hoe,
and ask their family spirits to protect the
3 In former times Goba headmen did not count eloped couples and
absent labour migrants as village a suitor among their residential
followers until he members even before tsambo has been delivered.
had paid tsambo. This severs his residential connec- This practice
has been followed in enumerating tion to his mother's compound and
fixes his position residence in Table II. as a junior member of his
wife's compound where 4 Some women have two or more family spirits
his behaviour is supervised by the fathers-in-law to be recognized.
In childhood everyone is given the who control the circulation of
tsambo and consented name of an ancestor whose spirit (sazita
mudzimu) to the marriage. But because modem elopement acts as a
guardian for all its namesakes. In addition damages are usually
satisfied over a period of years older women sometimes inherit the
spirit of a de- interspersed by many visits to the wife's village,
ceased sibling or ancestor of the same sex (mudzimu headmen now
tend to overlook this and count re nhaka).
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY couple and their house.
Henceforth the couple will depend upon her parents to com- municate
with her family spirits on their behalf and this is still an
important sanction, for it is believed that the spirits will attack
her and the children, causing illness or death and casting blame on
the husband, if things go wrong with the service or payments. And
as he is a junior member of their extended family household at this
stage in the marriage, it is thought that her family spirits can
also attack her husband. To be sure, the husband has a protective
spirit from his own family which can be invoked on his behalf by a
senior member of his own consanguineal descent group. But until his
status improves and he becomes head of his own independent
household, his guardian spirit is believed to be a relatively weak
one attached to him since the days of his childhood and therefore
inadequate to protect him from the adult de- mands of his in-laws.
Moreover, his own family spirits are likely to side with those of
the in-laws if he fails to perform his marital duties in a
trouble-free manner. The spiritual sanctions of the wife's family
are therefore particularly strong at this stage. For example, if
the wife should die the husband enters a period of ritual pollution
during which his activities are restricted until his in-laws
co-operate by removing his wife's spirit from his own (kuchenesa).
A special payment may be required for this service, especially if
an unpopular son-in-law has been implicated in the death. The
marriage hoes that symbolize a groom's general dependence upon his
wife's people are known as hoes of the realm (mapadza re nyika) and
they should be as ancient as possible. Pre-Colonial iron hoes
smelted locally or obtained in trade are still pre- ferred. Owing
to their scarcity they may now be obtained from any quarter
available and are often purchased for about L2.
In former times the hoes were supplied by the husband's sponsor,
with much the same significance as tsambo. Together with a reed mat
or two, some small stock, and an indefinite period of uxorilocal
service, the mapadza or hoes completed the sequence of transactions
necessary to establish a marriage. However, as wealth in cash
increased and came to be valued for its own sake, the mats and
small stock gradually gave way to a cash equivalent known as pfuma
or lobola in simulation of the bridewealth, especially cattle, used
in marriage contracts among highland Shona and Ndebele groups.
Traditional older men disinclined to seek wage work naturally
preferred to observe the mapadza customs in their own marriages but
some men over 60 were found to have paid an average of ?6 to ?7 for
pfuma. Most men from 40 to 6o had paid from ?L I to ?I2 and cash
pfuma is now expected in all marriages. The amount now seems to
have stabilized at ?9 to ?Io for the average villager though
shopkeepers may be asked to pay as much as 5 5.
Some fathers-in-law like to minimize the cash claims made in
their daughters' marriages in hopes of cementing their bond with
her husband and keeping the couple at home permanently. But others
are out to maximize their cash income from the marriage cycle and
they usually accomplish this by inflating their claims for
elopement damages rather than pfuma. Elopement damages are set
higher thanpfuma for a number of reasons. High damages can be set
with relative impunity and they lengthen the suitor's period of
dependency. During this time he may be pressured for gifts while
the girl is still not irrevocably committed and the suitor may be
re- jected in court with no abatement of the claim. But after
paying damages the husband is likely to be less tractable in the
face of a sizable demand forpfuma, especially if
53
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
54 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG damages were
high. Once pfuma has been paid the husband is at last head of his
own household. His personal spiritual protection is then stronger
in relation to the influence of his in-laws and he can address the
spirit community on behalf of himself and the children. And upon
payment ofpfuma the children are ritually severed from their
mother, formally filiated with father's descent group in addition
to mother's, and the mother's kin may thereafter be replaced by the
patrikin as the children's intercessors with the spirits. Should
the wife die after the payment ofpfuma, her kin are obliged to
purify her husband and terminate the marriage without any further
com- pensation. Should the husband die his wife and household are
inherited by a member of his own descent group signalled by those
who provided his tsambo beads. The payment of pfuma thus bespeaks
the younger family's potential spiritual and resi- dential
independence should the father-in-law's behaviour seem too harsh.
The government courts are relatively conservative about the size
ofpfuma, which must be returned if the marriage is dissolved. By
the time pfuma becomes an issue the wife, children, and
mother-in-law are likely to side with the husband in tempering the
father-in-law's demands if a large claim threatens to destroy the
unity of the matri- local extended family.
While mapadza firmly establishes the marriage and pfuma
increases a husband's rights and independence, the remaining tokens
have been more in the nature of closing prestations designed to
mark the end of the cycle. Muchato was introduced in the early days
of the British Colonial Administration. As touring District
Officers passed through their large districts in the rugged valley
and attempted to collect their annual tax many men evaded them by
alleging residence in some remote village, perhaps in another
district, where their wives lived. In addition, service husbands
complained about their endless 'slavery'. The administrators
attempted to stabilize the taxpaying population and deal with tax
evasion and alleged slavery by issuing a receipt witnessed by the
husband, wife, and father-in-law in cases where the marriage cycle
had been completed. This gave the union something of the flavour of
a registered marriage and came to be known as muchato. This
frequently required the father-in-law's co-operation in making a
lengthy trip to the nearest administrative post and in return he
required some clothing gifts from town, perhaps a few shillings, a
goat, or some beer to leave a good taste in his mouth. In many
cases this step might be avoided if the son-in-law decided to cast
his lot permanently with the affines and in such amicable
situations it has not been unknown for a sister to be provided to
replace a deceased wife. Even after muchato has been provided in
her marriage the youngest daughter is expected to remain with her
ageing parents.
After muchato the husband might offer to pay 5s. to Ios. as
'money for taking the wife away' (mutsimutso or nhakura), should he
decide to join the original tsambo pro- vider, or some other
sponsor who seemed significant by that time, or if he simply wished
to increase his bargaining power among his wife's people.
Especially before the development of a significant cash pfuma the
father-in-law might use spiritual sanctions to delay acceptance of
mutsimutso for some time. An offer to pay mutsimutso was most
likely to be pressed in cases where the husband had a chance to
succeed to an important kinship position in another village.
Failing that, this final payment might be foregone though if a
father-in-law was weak a husband might simply take his wife and
children away with no further ceremony.
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY Both muchato and mutsimutso are
probably things of the past. Before the develop-
ment of a sizable cash pfuma the son-in-law incorporated as a
junior member of his wife's residential compound by the ceremony of
the hoe might eventually be adopted into his wife's family, much as
a 'slave' might have been, and his period of uxorilocal service
might continue indefinitely unless the father-in-law could be
induced to accept muchato and mutsimutso. The final steps in the
cycle were important. Without them the children of many men felt
themselves closer to the descent group of the mother's father,
mother's brother, or other maternal kinsman who looked after their
mother's interests and they tended to take their descent names and
praise names5 from this source of patronage rather than from their
father. But now that a sizable bride- wealth equivalent has been
established this practice has been modified. The matrikin still
command prime loyalties but once the father has paid pfuma the
children take descent and praise names from their father's kin
group and the final closing tokens are likely to be ignored.
RESIDENCE
Data concerning the residence of persons at successive stages in
their lives are presented in Table II. Of course the majority of
children live with their parents (8I per cent). The remaining
residential ties of children illustrate their distribution upon
separation or death of the parents or the need of an elder for
company. Whether a divorced or deceased father has completed his
marriage cycle or not, most un- married children remain with the
mother or their matrikin.
The Goba male who marries for the first time in his mid to late
twenties can expect to leave his father-in-law by age 38 if he
wishes. Although this age varies widely among individuals, 74 per
cent of men under the age of 40 still find their choice of
residence determined primarily by the rule of uxorilocal marriage.
And as Table I indicates, most men under 40 are still in the early
stages of their marriage cycles and deeply involved in labour
migration. The significance of these figures may be made clearer if
we remember that the life expectancy for a male live-birth in
Zambia is probably of the order of 32 to 3 3 years, rising to about
5 z years for those who survive to age o. Comparable figures for
women are 35 and 53 (Coale and Demeny, I966). Sufficient data are
available to establish that at least 5 per cent of these uxorilocal
males under 40 in Table II had married into unrelated families
frequently some dis- tance from their homes. Because the service
period can be long and difficult for a husband many men prefer to
marry available women within their home communities or, failing
that, related women in other villages where they can expect greater
kind- ness and easier terms from kinsmen and family friends.
The marriage pattern is illustrated in Table III. An analysis of
217 marriages of men of all ages in which the pattern of kinship
and residence were well known to the investigator revealed that 5 3
per cent were kin marriages. And in Table II at least 29 per cent
of husbands under 40 whose residence is primarily determined by the
rule of uxorilocal marriage are known to have improved their
situation by marrying related women or neighbours in 'home'
villages. Since uxorilocal residence im- mediately after marriage
is still universal and generally prolonged for the average
s I am following Bourdillon (1972) in referring to the Shona
mutupo as a praise name rather than a clan or clan name.
55
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
56 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG
TABLE II
The Goba residence pattern Men under age 40
Principal Number Percentage residence tie of cases marrying
kin'
self S D B 7 57 z MB 4 I00 MF I M II 91 M&F I9 20% 95 F Io
40 uxor. 146 = 74% 29
Total 198 42
Women under age 35
Principal Number residence tie of cases
D B 5 Z I MB 4 MM I M 328 M&F I04 86% F I8J H I5 = i5%
Total I80
Men over age 40
Principal Number Percentage residence tie of cases marrying
kin'
self 5 S 2 D 2 B1 5 )14% B 8 )4% 38 MB 6 MF M 18 i M & F 42
51% 48 F 21 43 uxor. 41 = 26% 24
Total I6o 28
Women over age 3 5
Principal Number Percentage residence tie of cases marrying
kin'
D 4' 27 B I5 I9% Z 7 MB 4/ MM M 27 41 M&F 30 46% 57 F 14 2I
H 54 = 35% 9
Total 155 26
Unmarried children
Principal Number residence tie of cases
MM & MF I7 B 5 Z I MB 21 I MZ 2 M 6I) M & F 557= 81% F
i6 FZ 2 FM5
Total 687
Note I: 'kin' refers to categories II, II, and IV in Table
III.
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY
TABLE III
The Goba marriage pattern (Sample of 217 marriages of men of all
ages)
Number of marriages Percentage
I. Marriages to strangers among whom no kin ties were recognized
at the time of inquiry
II. Marriages to village kin associated with the village who
were either unrelated people living in the village or assumed to be
distant kin whose ties were no longer traced precisely
mI. Marriages to distant kin generally in or near the village
where a clear and con- sistent tie could not be established but
where kinship is undoubted
IV. Marriages to specifiable kin in or near the village where
precise ties were recog- nized as follows:
Married his real or classificatory:
z Z ZD
MBD= D MZD = Z
MZDD = ZD MZSD = D
MMBD = D MMBDD = DD MMBSD = SD
MMZDD = Z MMZDSD = D
MMZSD = D MFZSD = D
FBD = Z FBSSD = SD
FZD = ZD FZDD = ZD FZSD = Z FFZD = ZD
FFZDD = Z FMZSD = Z
FMZSDD = ZD FMZDD = Z
FMZDDD ZD FMBDD = DD
FMMZDDD = Z FFMBSSD = D
Before marriage he called her: hanzadzi muzukuru mwana hanzadzi
muzukuru mwana mwana muzukuru muzukuru hanzadzi mwana mwana mwana
hanzadzi muzukuru muzukuru muzukuru hanzadzi muzukuru hanzadzi
hanzadzi muzukuru hanzadzi muzukuru muzukuru hanzadzi mwana
102 47
Io%
21
I2%
26, \
8 3
I3 8 2 I I I 2 I I I I I I
9 2
3 I I I I I I I I
685
217
k3I%
53
I00
57
\ =
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
58 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG villager, the
matrilocal extended family is the most stable unit in social life
and uterine ties are both important and plentiful. As a result a
significant number of men aiming to practise local endogamy marry
their matrikin. Not counting marriages to 'sister' and 'sister's
daughter', 55 per cent (32 of 57) of the marriage to specifiable
kinswomen in Table III are to matrikin, for example. Of course the
prevalence of in-marriage makes its possible to redefine many
matrilateral links as patrilateral ones and vice versa. Marriage to
a kinswoman in the home community is the preferred pattern as it
obviates the need to leave it during an extended period of
brideservice; failing that it is easier for a man to bring his wife
back to his childhood surroundings of parents, siblings, and
matrikin at an early age if his wife is a relative and feels that
she too has made a good marriage among her own people. With the
exception of a full sister or mother, a man may marry any other
kinswoman. While no preferences are expressed it is still
particularly prestigeful in village circles to marry one's half or
classificatory sister. As Table II suggests, if a man is to escape
uxorilocal residence among relative strangers before the age of 40,
local endogamy and the practice of marrying kinswomen have been the
likeliest solutions.
Men over 40 still exhibit a significant tendency towards
uxorilocality as the prime mover in their choice of residence (26
per cent in Table II). Like younger men, many of these uxorilocal
husbands have improved their situation by marrying kinswomen (24
per cent) and in any event many older men decide to remain
permanently with their affines, especially if they have no
attractive collection of matrikin to return to. But compared to
younger men, most men over 40 have finished their service cycle and
labour migration tours (see Table I). Many more men over 40 have
had time to rejoin the compounds of one or both parents (5 per cent
in Table II) or to reunite with a sibling (14 per cent), while a
few go to other villages as relative strangers or join their
offspring in old age.
Most women under 35 are still residentially tied to their
parents (86 per cent) or other close uterine kin. Women begin to be
taken away by their husbands at an average age of 3 3 but even in
older age the majority of women live with their parents (46 per
cent) or close uterine kin (19 per cent). This of course is the
preferred life for a woman and marriage to a kinsman or neighbour
increases the chances that her spouse will not want to take her
away to live with strangers in her old age.
AUTHORITY
When they are still young, dependent children are under the
immediate control of their mother and father who handle most daily
household problems and, especially for daughters, this dependence
upon parents continues up to the time thatpfuma has been accepted
in their own marriages. Before that time the children are more
likely to turn to their father with their problems than at any
other time in their lives because a father who has finished his own
marriage cycle is always head of his own household. Daughters are
not detached from the father until their pfuma has been received.
Meanwhile until he has deliveredpfuma in his own marriage a son
continues to depend upon his father and father-in-law.
Afterpfuma has been paid over for a daughter she and her
children are still members of the father's kin group (mugowa) but
now the strength of her attachment to her mother's kin group
emerges more clearly. After all in most cases her children like
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY 59 herself, her sisters, and her
mother's sisters have spent their formative years and in- deed the
bulk if not all of their lives among their mothers' matrilateral
kin. Compared with virilocal and patrilineal Plateau Shona who
marry with cattle, Goba local units are based upon a nucleus of
mothers, sisters, and daughters and their descendants of both
sexes. It is largely to their children, the vazukuru, that wealth
and social position are handed down when male household heads die
and family estates are given over to a successor. The Goba point
out that many fathers are mere strangers, present in their wives'
villages only for marriage while the dominant, generally
co-resident matrikin 'eat nothing' from a daughter's marriage
payments though she has lived and may continue to live among them.
Once pfuma has been settled in her marriage a daughter's frame of
reference widens as her status grows from that of dependent junior
member of her father's household to that of full adult member of
her matrilateral descent group (mugowa) which is likely to control
the entire village. She is also forced to think of her children's
position as they are likely to find their loyalties divided between
her own mugowa, which supports her, and that of her hus- band who
generally wants to take his wife and children away if he is a
stranger in the village. Especially after pfuma has been delivered
for her, a woman is likely to heed the authority of her mother's
father, if he is still alive, her mother's brother, mother, or
senior brother with the father's authority running second
particularly if he has no close ties with the mother's descent
group.
A corresponding pattern holds true for a son. After he delivers
pfuma his father-in- law's hold is relaxed and a son's children's
loyalties are then mainly divided between their father and his
matrilateral descent group and that of their mother. In their own
youth his children are likely to consider their mother's mother's
brother, mother's father, or mother's senior brother as their chief
authority figure while the father occupies a marginal position and,
being younger, is generally away seeking wage work. As they
progress further in the life cycle the children are ever more
likely to recognize and identify with a senior male matrikinsman as
head of their descent group unless the father belongs to a strong
mugowa of his own and has been able to take his family home while
he is still young or has been able to marry and live within it in
the first place.
Even before he marries and leaves the family compound, a son
begins to assume a certain brotherly concern for his sisters'
welfare. As he grows older he begins to take the father's place
when necessary, in normal Shona fashion, and this informal
patrilineal succession may be of more than token significance if
there are many fertile sisters to look after and if the father is a
'stranger' from a distant section of the chieftaincy so that his
genealogical superiors (father's father and father's mother's
brother) and other potential successors (father's brother, his
sisters' sons and other sons) are unable to look after minor
problems on a regular basis. All sons represent and eventually
replace the father in this manner, at least on a situational basis
in con- junction with other successors, including the father's
brother or father's sister's son who eventually inherits father's
household estate. Personal dominance, relative age, and accidents
of early death usually determine which son shall have senior
fraternal responsibility for his sisters and their families and if
a son is ever to break away from his in-laws and amass an
independent residential following larger than his own extended
family its nucleus is likely to be a married sister. As he grows
older his main source of support will probably continue to consist
of a co-resident sister
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
6o BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG along with her
co-resident daughters and his own. This residential core, or
minimal mugowa, is about all most independent men have been able to
amass in the relatively short time left after the marriage cycle is
ended. With such a following a son may become head of his own
village section and be a welcome ally of competing headmen.
Much more influence can be attained by a son who succeeds to a
position of authority among his mother's kin, especially if they
control an important headman- ship; most of his real and
classificatory sisters, women of his mother's generation and other
matrikin are likely to be gathered together by the combined forces
of uxorilocal marriage, matrilocal residence, and the largely
matrilineal processes of succession and inheritance. Succession to
the position of a mother's brother can be an important step in such
circumstances inasmuch as it tightens a man's hold over his
immediate circle of sisters and daughters while adding influence
over women of his mother's generation, their wider circle of
descendants, and other of the mother's brother's former
dependants.
Effective succession to the critical position of senior mother's
brother in a major descent group is keenly contested in a system
where co-resident blocks of sisters are linked together by
matrilateral ties. Nevertheless the Goba still prefer patrilineal
succession in matters affecting a descent group as a whole. The
sister's son whose father was a 'village kinsman' or close member
of the mother's descent group and, preferably, the sister's son who
is also an actual or classificatory son of the mother's brother has
a competitive advantage over the sister's son whose father was
merely an uxorilocal 'stranger'. The candidate who succeeds a key
mother's brother as a son as well as a sister's son may achieve
effective positional succession as a leader of the descent group to
the exclusion of the sister's son who succeeded the mother's
brother as a compound head at the time his household estate was
settled. Succession by men who follow their parents in marrying
close relatives, who can therefore remain in or near their base of
support in a home village, and who can trace claims to positions
within the kindred through both their mothers and fathers,
represents the preferred mode of succession in major matters. In
view of their patrilineal traditions and largely matrilineal
practices this has been the only way a man can succeed to major
positions of authority among the Goba. In the case of an important
succession, such as a major headmanship or chiefship, where the
descent group is unusually large, influence over the various blocks
of sisters likely to be involved usually has to be shared with
other more closely related brothers and mother's brothers and as
time passes fresh in- marriages tend to bring peripheral sorority
groups back into closer contact with the central descent group
factions. Finer rank distinctions between competing mother's
brothers within the descent group are based upon positional
succession and manceuvr- ings for genealogical seniority, but in
general men related to the 'original' chief or headman through an
uninterrupted line of men rank ahead of those have to utilize one
or more uterine links. Table IV illustrates the pattern of
succession to positions of authority.
DISCUSSION The Goba brideservice system exerts a substantial
influence on the individual's life
cycle, directly encouraging a high rate of labour migration and
influencing experience throughout much of the individual's life
span, as we have seen. In a region where life
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY 61 expectancies are short, a
considerable number of men in the 40 to 49 age-group are still
involved with the brideservice cycle in their own marriages. The
brideservice system also has a substantial impact on the residence
pattern and through it the descent pattern. While the father's kin
are still referred to as comprising a rudzi, the word for patriline
among neighbouring cattle-owning Shona, uxorilocal residence
TABLE IV
The locus of authority
Descent group leaders Adult Informants whose marriage (Nhundu
heads) informants cycle is incomplete
IMMMBSS = MB I 2MMF- MMB- MB 2 I 4 MMB- MMBS = MB 2 3 MMB- MMBZS
= MB I MFMBS - MB I MF- MFBS- MB I I MF MB 2 I I 2 MB 27 40 27 22
MB- MBS = MB 2 2 I M 3 MZ= M I MB-B= MB 2 4 MB-> self (male
speaker) 5
3MB- self- ZS = MB I 3MB- self-- S =MB I
Totals 46 53 3 32
F I 7 7 10 F- FZS= F 2 5 FMB -F I FF I FZ I F-B=F I
Totals 4 13 8 I I
= means 'terminological equivalent'. 2 = means 'succeeded by'. 3
Anticipatory succession as indicated by two old men.
and succession by sisters' children have broadened the Goba
rudzi into a patrilateral descent group. The minimal effective
descent group that crystallizes out of the pre- vailing residence
pattern is the nhundu whose primary reference is to the 'dependent
family members' or vazukuru. The nhundu is normally the unit
composed of a brother who looks after the interests of his
co-resident sisters and their co-resident married daughters.
Ambitious men seek to gain influence over as many such units as
possible via the force of personality, the application of religious
sanctions, and manceuvrings for seniority on the kinship grid. A
larger collection of such followers, not all of whom need be
co-resident, is called the mugowa, chipani, or tsaka. In practice
the
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
62 BRIDESERVICE, RESIDENCE, AND AUTHORITY AMONG mugowa is
actually a bilateral kindred although the Goba conceptualize it in
dual terms as a house or set of uterine kin who are looked after by
a line of in-marrying father's sons.
Richards (1950: 208) has demonstrated how variations in family
structure in Central Africa 'depend largely on the nature of the
marriage contract and the extent to which the husband is able to
gain control over his wife' and children. She found the rule of
residence at marriage to be the most important index of the
husband's status. In keeping with this, kugarira or prolonged
uxorilocal brideservice has been an influential institution among
the Goba in both personal and structural terms. It has also been a
key characteristic traditionally used to distinguish the poor
lowland Goba from more powerful Shona-speaking groups on the
Southern Rhodesian highlands. In pre-Colonial times the manual
labour of service husbands and de- pendent sisters' children
(vanyai) in the family compounds was augmented by their activities
as hunters, raiders, and traders and this service or tribute was an
important element in the hierarchical political structure of the
Shona confederacies in the Zambezi Valley and in the prazo system
developed by the Portuguese. Today, in an entirely different
political context, the kugarira system continues to characterize
the Goba but because it has been monetized through labour migration
remittances from a money economy the modern Goba brideservice
system now generates a cash flow rivalling that of some
bridewealth-paying groups.
Regardless of the wider political context, as long as a society
remains kin-oriented it would seem that some culturally approved
form of commensurate value must be exchanged in order to detach a
wife from her family of orientation. The Goba have learned to value
money and today the Goba male can expect to pay a total ranging
from ?26 to as much as ?130 for his first wife, though the average
village husband pays something of the order of ?40 and his
father-in-law may have to wait some twelve years to receive all of
it. Unlike the Goba, other Zambezi Valley groups such as the Tavara
and Valley Korekore of Southern Rhodesia and the Valley Tonga of
Zambia have used the growing wealth obtainable from migrant labour
to convert their former uxorilocal brideservice systems6 into
patrilocal bridewealth systems. Like the Goba the Tavara used to
marry uxorilocally in the absence of cattle, marry close relatives,
and inherit the praise names of their matrikin but all this is
being replaced as the Tavara, like the Valley Korekore, seek to
copy the more general and prestigious Shona practice of substantial
bridewealth followed by patrilocal residence. No figures are
available for Tavara bridewealth but the Valley Korekore paid some
?C35 as of I963-4 (Bourdillon I972; Garbett, 1967). The Tavara and
Valley Korekore use the practices of neighbouring highland Shona
groups as standards in evaluating their own performances whereas
the Zambian Goba have used the obvious matri- lineal aspects of
their social behaviour to imitate their own prestigeful neighbours,
the matrilineal Plateau Tonga of Zambia with whom they intergrade.
But in recent years both the Plateau and Valley Tonga of Zambia
have themselves taken up bride- wealth and developed a patrilocal
residential bias. No recent figures are available for the Plateau
Tonga but the Middle River Valley Tonga had established
bridewealth
6 That the Valley Tonga of Zambia formerly Tour Reports, Gwembe,
dating from colonial times. practised uxorilocal brideservice on a
large scale is These documents are on file in the Zambian National
reported in the unpublished District Notebook and Archives,
Lusaka.
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY and damages averaging some ?50 by
I963 (Colson, 1971). It seems clear that this Tonga precedent
together with government efforts to eradicate tsetse, introduce
cash cropping, and eliminate local poverty will eventually lead to
a more obvious asser- tion of patrilineal ideals among the Zambian
Goba, especially once border problems limiting freedom of travel
between Zambia and Southern Rhodesia have been resolved.7
REFERENCES BOURDILLON, M. F. C. 1972. 'The Manipulation of Myth
in a Tavara Chiefdom', Africa xlii. 2: 112-21. BULLOCK, C. I928.
The Mashona, Cape Town: Juta. COALE, A. J., and DsmENt, P. 1966.
Regional Model Life Tables and Stable Populations. New Jersey:
Princeton
University Press.
7 This reassertion of patrilineal practices among the Goba will
probably occur through slight changes in the normal composition of
the residential nhundu, the critical local unit in Goba social
organization. From the viewpoint of its male leader the residential
nhundu is most commonly composed of dependent sisters' children
(vazukuru), as described earlier, though his own set of descendants
and those of his brothers are also recognized as vazukuru who com-
prise a nhundu. In localities where cattle survive and wealth
accumulates from growth of the herd, income from ploughing
services, and the higher crop yields obtainable from ploughed
fields, a father can help his son terminate the marriage cycle
earlier, young men with growing families tend to live with their
fathers, and the nhundu of descendants stemming from a father can
become an important factor in village organization. This is also
noticeable in villages where a dominant father is able to control
the cash savings of migrant sons or sons getting involved in local
cash cropping. As acceptable bride-wealth and patri- local
residence increase the patrilateral nhundu assumes greater
importance in the local-level competition for residential
followers. For example, among the pre- dominantly patrilocal Middle
River Valley Tonga the patrilateral lutundu (Tonga for nhundu) is a
key institution (Lancaster, x966). A large group of Middle River
Valley Tonga were resettled among the Zambian Goba in I957-8 in
connection with the Kariba Hydro-Electric Power Project. Presumably
they will influence Goba hoping to assimilate as Tonga. Of course
the uxorilocal Goba have long recognized the usefulness of
in-marriage combined with control of the patrilateral nhundu as a
device for constituting a meaningful patriline.
In this connection it may be of interest to note that an
historical understanding of the dynamics underlying Goba social
organization may shed some light on the unusual 'double descent' of
the Ila originally studied by Smith and Dale (I968 [I920]) and
later reanalysed by Richards (1950). To re- capitulate briefly, a
father-son nhundu (lutundu in Tonga) becomes localized and
important among the Goba as the wherewithal for acceptable bride
pay- ments increases. We then find patrilocal residence
upon marriage or fairly soon thereafter. This results in a
patrilocal extended family, or nhundu, which is in effect a
localized patriline superimposed upon a matrilineal framework
supplied by the matrilateral kindred, or mugowa, most of whose
members are usually dispersed. Because of local variations in
wealth, personality, and other factors, a localized nhundu or
'minimal' mugowa composed of a woman and her co-resident married
daughters and grand- daughters may also be found, so that both
patrilineal and matrilineal groups can emerge from a kinship nexus
that is essentially bilateral. In the circum- stances a Goba male
can succeed to kinship positions both through his father and
mother's brother. This would appear to fit Richards's description
of Ila social organization rather than that of the Goba if we
substitute the Ila term lunungu for nhundu or lutundu to refer to
the Ila cattle inheritance unit and the Ila term mukoa for mugowa.
Field research is needed to verify this hypothesis, of course. And
it must be remembered that in common with many Central African
peoples the Ila have been amor- phously organized politically so
that rather than representing a neatly bounded and homogeneous
ethnic unit the Ia label has probably been applied to peoples
occupying somewhat different environ- ments who may therefore be
expected to display varying social patterns. However, the
hypothesis may be favoured by the fact that there has been
historical contact between the Ila and the Goba. The Ia appear to
have migrated west along the Zambezi to reach their home in the
Kafue basin and in the nineteenth-century, if not earlier, the Goba
along the lower Kafue River near its junction with the Zambzsi
intergraded with the Ila. Smith and Dale included the population of
the Kafue-Zambezi con- fluence among their Ila-speaking peoples.
One major difference between the Ila and the Goba, of course, is
that the Goba have been known for their posses- sion of few if any
cattle, so that only a thin father-son patriline of in-marrying
mothers' brothers exists to look after a descent group core
composed of co- resident descendants of female agnates. The Ila, in
contrast, are known for their large cattle herds and localized
cattle-holding patrilines.
63
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
-
THE GOBA OF THE ZAMBEZI VALLEY COLSON, E. I971. The Social
Consequences of Resettlement. Kariba Studies IV. Manchester:
Manchester Uni-
versity Press. GARBETT, G. K. I966. 'Religious Aspects of
Political Succession among the Valley Korekore (N. Shona)',
In E. Stokes and R. Brown (eds.), The Zambesian Past.
Manchester: Manchester University Press. 967. 'Prestige, Status,
and Power in a Modem Valley Korekore Chiefdom, Rhodesia', Africa,
xxxvii.
3: 307-26. HOLLEMAN, J. F. I969. Shona Customary Law.
Manchester: Manchester University Press (reprint of I952
original). KAY, G. 1967. A Social Geography of Zambia. London:
University of London Press. LANCASTER, C. S. I966. 'Reciprocity,
Redistribution and the Male Life Cycle: variations in Middle
River
Tonga social organization', African Social Research, 2: I39-57.
- I97I. 'The Economics of Social Organization in an Ethnic Border
Zone: the Goba (Northern Shona)
of the Zambezi Valley', Ethnology, IO: 445-65. n.d. 'Ethnic
Identity, History, and "Tribe" in the Middle Zambezi Valley'.
RICHARDS, A. I. I950. 'Some Types of Family Structure amongst
the Central Bantu'. In A. R. Radcliffe- Brown and D. Forde (eds.)
African Systems of Kinship and Marriage, London: Oxford University
Press.
SCHAPERA, I. I929. 'Matrilocal Marriage in Southern Rhodesia',
Man 29: II3-17. SMITH, E. W., and DALE, A. M. 1968. The
Ila-Speaking Peoples of Northern Rhodesia (reprint of 1920
original).
New York: University Books.
Resume PRESTATIONS EN NATURE, RESIDENCE ET AUTORITEI PARMI LES
GOBA
(N. SHONA) DE LA VATLL,TA DU ZAMBIZE LES Goba de langue Shona de
la vallee du Zambeze qui habitent pres de l'affluent de la riviere
Kafue en Zambie occupent une region infestee par la mouche tse-tse
qui detruit le cheptel. Contrairement aux autres groupes Shona des
hautes terres de la Rhodesie du Sud qui epousent traditionnellement
des femmes dotees de cheptel et qui sont patrilineaires et
patrilocaux, les Goba, eux, pratiquent une forme de mariage
comportant des prestations en nature effectuees dans la residence
de la future femme, vivent en vastes families matrilocales et
observent a la fois des regles de descendance matrilineaires et
patrilineaires.
La migration de la main-d'ceuvre comportant une retribution en
especes ont quelque peu, ces dernieres annees, mis en echec le
travail manuel effectue par le fiance dans les champs de la famille
de sa fiancee, mais les Goba, qui sont peu nombreux en Zambie, ont
conserve quelques traits essentiels de leur systeme de prestations
en nature afin de faciliter l'assimila- tion avec leurs voisins
Tonga matrilineaires qui constituent le groupe ethnique dominant en
Zambie du Sud.
64
This content downloaded from 163.200.81.2 on Mon, 9 Dec 2013
04:50:47 AMAll use subject to JSTOR Terms and Conditions
Article Contentsp. 46p. 47p. 48p. 49p. 50p. 51p. 52p. 53p. 54p.
55p. 56p. 57p. 58p. 59p. 60p. 61p. 62p. 63p. 64
Issue Table of ContentsAfrica: Journal of the International
African Institute, Vol. 44, No. 1 (Jan., 1974), pp. 1-4+1-116Front
Matter [pp. 1-4]Daryll Forde: Further Tributes [pp.
1-10]Woman-Marriage, with Special Reference to the Loedu. Its
Significance for the Definition of Marriage [pp. 11-37]An
Analytical Note on the Land and Spirits of the Sewa Mende [pp.
38-45]Brideservice, Residence, and Authority among the Goba (N.
Shona) of the Zambezi Valley [pp. 46-64]Youths as Elders and
Infants as Ancestors: The Complementarity of Alternate Generations,
Both Living and Dead, in Tiriki, Kenya, and Irigwe, Nigeria [pp.
65-70]Arabic Literacy among the Yalunka of Sierra Leone [pp.
71-81]Meeting of the Executive Council, London, 26-7 June 1973 [pp.
82-83]Notes and News [pp. 84-87]Ruth Jones, Librarian of the
International African Institute, 1944-73 [p. 88]Letters to "Africa"
[pp. 89-91]Reviews of BooksReview: untitled [pp. 92-93]Review:
untitled [pp. 93-94]Review: untitled [pp. 94-95]Review: untitled
[p. 95]Review: untitled [pp. 95-96]Review: untitled [pp.
96-97]Review: untitled [pp. 97-98]Review: untitled [pp.
98-99]Review: untitled [pp. 99-100]Review: untitled [pp.
100-101]Review: untitled [pp. 101-102]Review: untitled [p.
102]Review: untitled [p. 103]Review: untitled [pp. 103-104]Review:
untitled [pp. 104-106]Review: untitled [pp. 106-107]Review:
untitled [pp. 107-108]Review: untitled [p. 108]Review: untitled
[pp. 108-109]Review: untitled [pp. 109-110]Review: untitled [p.
110]Review: untitled [pp. 110-111]Review: untitled [pp.
111-112]Review: untitled [p. 112]Review: untitled [pp.
112-113]Review: untitled [pp. 113-114]
Books Received [pp. 115-116]Back Matter