nggwun proposal 1997/8

Mambila Nggwun - the construction and deployment

of multiple meanings in ritual.

Centre for Social Anthropology and Computing

The University of Kent at Canterbury

Research proposal to the ESRC (funded 1997)

Summary:

Recently there has been considerable debate about traditions and rituals ñ their
authentic or invented features and their implications for authority and power
(Hobsbawm and Ranger (eds), 1983; Jolly and Thomas 1992 and Boissevan
(ed.) 1992). In other research (Fischer 1994; Lyon and Fischer 1994) we are
developing and evaluating an approach which represents traditions as
specifications used (and manipulated) by people to identify and generate
practices and knowledge termed traditional, rather than a specific inventory of
past practices and knowledge; the features and components of specifications
will be altered or replaced over time, adapting to current requirements.

In the proposed research we use the study of ritual to extend this idea. As a
genre of tradition, ritual provides a useful focus. Like all traditional events, a
ritual is constructed by participants each time it is enacted. However, the
materials from which it can be formed are constrained and more subject to
scrutiny, intersubjective or otherwise, by both participants and outsiders.
Rituals often have explicit objectives, in addition to more subtle ones, and it is
important that these be seen to be fulfilled. Most rituals have core segments
which must be present, and other segments which are either optional or whose
form may vary considerably. Attitudes about which must be present and which
are optional vary between participants and observers, to the effect that there
are often more segments present in a specific ritual event than any one
participant judges necessary. As an object of study, however variable, a ritual
has a kind of predictability in its structure which makes it relatively easier to
study than some less structured events. This research will produce specific
models of how collective representations of a specific socio-cultural ritual
event are structured and distributed between participants and observers, and
how these are accessed and used by people to solve problems in the present. It
will thus complement and extend some recently proposed theories of ritual
(see Humphrey & Laidlaw 1994 as a recent response to Goodyís 1977
critique).

This project will produce a detailed ethnographic analysis of the Mambila
Nggwun ritual from the reference point of different individuals who are

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 2

Michael Fischer and David Zeitlyn

differently placed socially and spatially, and in terms of alternative composite
viewpoints consistant with this range of reference points. Using the data that
has been already collected in the course of other previous fieldwork among
the Mambila, including recent fieldwork funded by the proposers and the
Centre for Social Anthropology and Computing timed to coincide with the
performance of Nggwun that was held in December 1996.

To focus the discussions, to help overcome the problems of multi-lingual
interviewing (working with some informants, Kwanja particularly, will entail
using a mix of Mambila, French and Fulfulde), and to collect specific data on
observer reference and competence we will use segments of digital video and
still images (together with spatial and temporal collations of video and still
images) as elicitation tools. The response to these clips will themselves be
recorded which will have immediate benefit in the field as some of the
linguistic issues can then be clarified immediately.

One conventional problem with studying a ritual event is from whose
perspective to study it, especially if we want to capture the many different
perspectives that contribute to the event. This is a particular problem if the
ritual events are being recorded in the form of notes by the ethnographer, and
only later discussed with participants. As Morphy (1994) suggests, there are
circumstances where reliance on notes (for anthropologists) and memory (for
participants) are problematic, and can be enhanced by the use of non-verbal
recording. We are proposing to address this problem in two ways. Firstly, this
particular ritual continues for a period of more than 72 hours. However,
much of this is highly repetitive from any one perspective. In the recent
fieldwork we videoed segments of the event from as many points of view as
possible, filming under the advisement of indigenous consultants, and
attempted to elicit a range of comment from participants and observers during
the event itself. Secondly, for the proposed research we will select segments of
this video, (some prepared prior to the proposed fieldwork and some during
the fieldwork) under the advisement of indigenous consultants, to prepare
computer-based multimedia documents as an elicitation device for a range of
participants and observers, as well as a number of ëothersí. In particular, we
will investigate the range of comment and opinion these differently situated
individuals have on the different situations (in particular comparing their
reaction to differently situated observersí constructions of the ritual), to
identify their situated specification of what defines the ritual and what is
optional (and why), and to elicit their help in producing a ëdefinitiveí account
of the ritual based on the clips we make available (the materials for
construction) for each of the different consultants.

We believe that using this computer-based process in the field to record,

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 3

Michael Fischer and David Zeitlyn

identify and discuss the ritual is perhaps the only manner in which we can
begin to approximate the interactional effect of the many participatory
viewpoints on an ethnographic event, and to communicate our results
analytically.

We will be drawing on technology largely developed from previous and
current research, but will be using this technology in a more ambitious
manner than we have attempted before.
Specifically, we will:

examine the process by which a ëtraditional ritual performanceí is

understood, identified and/or invoked to serve in different social and
cultural contexts by different participants of the rites,

produce intermedia documents with indigenous experts describing some of

these processes, used as both a source of data and a field tool for
ethnographic research,

develop detailed models of how a specific ritual domain is understood by

participants and witnesses from different ethnic groups, and

test these models against indigenous judgements.

Background

In recent years there has been considerable debate about traditions, their
authentic or invented features and their implications for authority and power
(Hobsbawm and Ranger, 1983; Jolly and Thomas 1992). In the context of
studies in the Pacific Thomas suggests that questions concerning tradition
should be reformulated to ask ëAgainst what are traditions invented?í
(Thomas, 1992), thus arguing for an approach which recognizes complex
historical relations, between outsiders and insiders, in the articulation of
tradition. Issues related to tradition have been readdressed by anthropologists
working in Europe and in Africa. They note that public celebrations and local
rituals are currently expanding or being ërevitalisedí (Boissevain, 1992).
Other researchers have noted that practices indigenously presented as
traditional are apparently either invented altogether (Larcom 1982; Borofski
1987, Wright 1992), re-invented (Keesing and Tonkinson 1982), are based on
ëforeigní influences (Thomas 1991), or are based on the reports of
ethnographers (Jolly and Thomas 1992).

Based on other research (Fischer 1994b; Fischer and Lyon 1994; Lyon and
Fischer 1994; Fischer forthcoming) we are developing and evaluating an
approach which represents traditions as specifications used (and manipulated)
by people to identify and generate practices and knowledge termed traditional,

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 4

Michael Fischer and David Zeitlyn

rather than a specific inventory of past practices and knowledge; the features
and components of specifications will be altered or replaced over time,
adapting to current requirements. This research will produce specification
models of how collective representations of a socio-cultural event are
structured and distributed, how these are accessed and used by people to solve
problems in the present.

Specifications are rules that identify when goals or outcomes have been
achieved. Specifications are usually either partial or provisional; that is a
specification does not attempt to be a comprehensive description, but rather a
partial description. A specification is not itself generative of phenomena - it is
descriptive - but we explicitly assume a specification is embedded in a
generative process. A specification is a tool for describing what is obligatory,
permitted, and impossible, in general and within specific contexts.

The advantage of this way of looking at social phenomena is that it can account
for the very important interactive components of accomplishing a goal - what
is sometimes referred to as negotiation, but also including mistakes, accidents
and unaccountable circumstances, as well as incorporating differing objectives
by different participants from unique perspectives. Specifications can also be
used as standards against which ëreal lifeí results are measured. Both of these
senses correspond to the more usual concept of norms in anthropology, but
unlike norms specifications explicitly assume that a separate set of processes
will be used to achieve the specification (and will be guided by the
specification), and do not assume that norms are defined in exactly the same
manner by all agents.

In the proposed research specifications will be used as a tool for representing,
relating and comparing the varying positions of the participants in the
Nggwun ritual in our attempt to develop an analytic model which
incorporates these different perspectives concurrently.

A case study: the Mambila Nggwun ritual

The Mambila Nggwun ritual has not been studied as extensively as other
Mambila rites such as sugàgà (see Zeitlyn 1994). In this project we will
produce a detailed ethnographic analysis of the ritual, using the data that has
been already collected in the course of other previous fieldwork among the
Mambila, and using it as the basis for further research timed to coincide with
the performance of Nggwun which is due to be held in December 1996.
Zeitlyn was present during most of the ritual in 1994 (the ritual occurs on a
two year cycle) and was invited by the Chief to return with a video camera.
In addition he was given express permission by the ëchiefs of medicineí (the

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 5

Michael Fischer and David Zeitlyn

men who run the ritual), to witness the most private parts of the rite of which
he has never even had a description to in the (then) nine years of his field
experience with Mambila in Somié village. Being granted access, particularly
with this combination of video recording plus innermost organisation, raises
the possibility of investigating the different meanings of the rite for widely
varying populations who are all participants in the ritual. Furthermore, the
use of digital video in the field provides a method of cross-linguistic elicitation
where the video images can provide consistent foci for discussion in
neighbouring Mambila villages, and in villages of neighbouring ethnic groups
who have similar rituals.

In essence the Nggwun rite repeats the installation rites of the chief every two
years¹. The outlying hamlets who owe allegiance to the chief enact their
subservience in the war dance performed around the chiefís palace. In this
those from the village centre fight those from the outlying hamlets, they are
beaten back but the warriors from the hamlets acquiesce when confronted by
the chief himself. In the private ritual the Chief is treated with special
medicines which protect him from evil actions (conceptualised as witchcraft) -
those of others and from performing any himself. Most of this is performed
privately in the innermost house of the Palace by the mgbe loe (the chiefs of
the medicine), who control the overall performance. In 1994 Zeitlyn was
given permission to witness these private rites, which previously he had never
even received a description. He was not, however, in a position to fully
exploit the opportunity since he was then engaged in the final stages of a socio-
linguistic research project, and was concentrating on checking translations.
His hasty notes and partial documentation form the basis for the current
proposal. On this basis we made a similar proposal to the ESRC in 1996,
which was rejected in the final stages due to concerns about our available time
and confusion over our analytic methods. Zeitlyn went to Cameroons to do the
necessary video work, funded by the proposers and CSAC. We have a clear
idea of what to expect which greatly assists the planning of the research and
undertaking of the filming which is part of this project.

Once the main ritual treatment has been performed in private the chief, the
mgbe loe and his senior sisters and wives process from the royal palace to the
square. They circle this the dance while the crowd bows. The chief and his
senior sister then sit on royal stools and drink three gourds of beer to the
accompaniment of cheering. Before leaving the palace, the chief repeats his
oath of office. He is within the innermost enclosure, but selected seniors are
allowed to remain outside that fence to witness his words. A final aspect to

1Some recent work on the analysis of royal ritual may be found in Cannadine and Price 1987.

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 6

Michael Fischer and David Zeitlyn

note is that some senior participants - a mgbe loe and a sistersí son of the
chief, go round the village blessing every house to protect them form evil.
In sum, this is a complex ritual which has different components with many
participants. There is no single ëfinalí or ëcorrectí account of Nggwun from
even a single Mambila village, rather, different participants will have very
different understandings of the event. This project seeks to explore some of
the ways in which different understandings are constructed and transmitted.
Two aspects of this will be studied. Different categories of participants in the
village of Somié will be interviewed about what the ritual means to them, and
how they conceptualise their participation in the ritual. A similar set of
questions will also be asked of people from neighbouring villages - of
Mambila, Kwanja and Tikar identity - many of whom attend the Nggwun
rites (although seeing only the public events in the square outside the palace).
Both Kwanja and Tikar have similar rites to Nggwun, but the meanings of
these rites may be expected to differ granted the very different social position
of chiefship among these groups. Analysis of the Kwanja material will be
undertaken collaboratively with Quentin Gausset, an anthropologist from the
Free University of Brussels, who has recently (1993-4) begun to undertake
research with the Kwanja, and after discussion with David Price who remains
the only anthropologist to conduct fieldwork with a Tikar group (see e.g.
Price 1985).

In part, Nggwun is a civic as well as a royal rite (see articles in Cannadine &
Price 1987), and as such is a critical focus of local (village) as well as of
ethnic (Mambila) identity. The study of Nggwun thus raises the question of
how ritual is used as an element in the construction and maintenance of a
variety identities, both local and ethnic. Mambila identity has not been a
major regional issue by comparison to that of the Tikar (see Chilver and
Kaberry 1971, Price 1979 and Fowler & Zeitlyn forthcoming) which makes
the comparison between Mambila and Tikar understandings of the rite
potentially revealing of the ways in which identity is constructed and
maintained.

Method

We will:

examine the process by which a ëtraditional ritual performanceí is

understood, identified and/or invoked to serve in different social and
cultural contexts by different participants of the rites,

produce multimedia documents with indigenous experts documenting some

of these processes, used as both a source of data and a field tool for
ethnographic research,

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 7

Michael Fischer and David Zeitlyn

modify and refine existing tools, and develop new tools, for formally

coding and referencing multi-media, including texts, images and time-
based video and sound. These are based on work relating to the APFT
Content Coding System (ACCS), which itself is based on earlier ESRC-
sponsored research by the proposers.

develop detailed models of how a specific ritual domain is understood by

participants and witnesses from different ethnic groups, and

test these models against indigenous judgements, particularly with

reference to competence.

The field methods to be used in this project include:

1) Participant observation
2) Semi-structured interviews unaided and using still images, digital

video clips and composite-multimedia as referents to focus discussion

3) Consultant-aided analysis of field recordings and recordings digitised
prior to the field
4) Multi-perspective (multi-camera) recording of central and peripheral

components of the Nggwun ritual

5) Consultant-aided selection of materials for 2)
6) Archival research in US Museums and UK libraries

The post-field methods to be used in this project include:

1) Indexation and digitisation of field material
2) Crosslinking (layering) of these to relevant interview material and
fieldnotes
3) Expanding the framework to include other field data
4) Investigating and formalising the multi-perspectival aspects of the
material in 2)
5) Developing an analytic model and specification of Nggwun
6) Testing the model by re-interviewing using material selected to test

the model of 5)

7) Consolidate results
8) Prepare datasets for archiving

Fischer and Zeitlyn will travel to Cameroon in July 1997 to undertake field
work in Somié village following the December 1996 performance of
Nggwun which Zeitlyn has videoed with the assistence of indigenous
camerafolk, which documented different aspects and viewpoints of the
Nggwun ritual, interviewing key participants before, during and after the
ritual.

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 8

Michael Fischer and David Zeitlyn

Some of the video footage is currently being digitised at UKC (funded by
CSAC) at higher quality than will be possible in the field. In order to focus
the discussions, and to help overcome the problems of multi-lingual
interviewing (working with some informants, Kwanja particularly, will entail
using a mix of Mambila, French and Fulfulde) we will use still images and
clips of digital video as elicitation tools to investigate differing conceptions
and interpretations of the ritual. The response to these clips will themselves
be recorded which will have immediate benefit in the field as some of the
linguistic issues can then be clarified immediately.

The work of producing digital video in the field will be undertaken by David
Zeitlyn and Michael Fischer, exploiting the unparalleled experience of Fischer
in using digital video in field conditions which will leave Zeitlyn relatively
free to continue ongoing ethnographic issues concerning Nggwun.

The University of Kent University Research Fund and CSAC have given some
pilot funding to digitise some existing video of public activities and
presentations of the ritual in advance of the fieldtrip. This will be used
immediately on arrival to consolidate workable procedures for conducting
interviews using digitised video clips as elicitation tools, and adapting the
methods which Fischer has used successfully in the Cook Islands and Pakistan
to the African context.

Analysis

The research will develop the following data:

Still images of the Nggwun ritual. Many of these will include annotations

from multiple respondants, and will be linked into specific interviews
where these were discussed.

Video sequences of the Nggwun ritual. Many of these (and subsequences)

will include annotations from multiple respondants, and will be linked into
specific interviews where these were discussed.

Interviews with people about the Nggwun ritual. These will be linked to

specific materials from 1 and 2 where these were discussed. The audio
record will be included with the interview where it is transcribed. Some
interviews will be maintained as digital audio documents with text
abstracts (formal and informal) rather than full transcriptions.

Fieldnotes with reference to the Nggwun ritual. These will be linked to

specific materials from 1-3 where these were discussed.

Rough maps of the area relating to the ritual, with point locations captured

by GPS. This is important with reference to kinship and social reference

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 9

Michael Fischer and David Zeitlyn

to the ritual.

Data actively used in the research will be content-coded using an extension of
the APFT Content Coding System (ACCS) (ISBN xxxxx -
http://lucy.ukc.ac.uk/Accs/) developed by CSAC for another project. ACCS is
a reduced set of coding terms (180) arranged into syntactic categories with
explicit semantic referents, and includes categories for temporal and spatial
references, agent/patient relationships and ethnographer and pragmatic intent.
ACCS is in part based on earlier ESRC research by Fischer on formal
languages in ethnographic research. Use of ACCS facilitates context-sensitive
retreval of texts, images and video sequences in response to formal queries.

In practical terms this means that multimedia documents can be partially
constructed by predictive linkage between elements with reference
touser/researcher stated goals/definitions, rather than fixed ëhardí links. In the
field this aspect will be used to test definitions and coding using indigenous
competence judgements, ie their reports of whether we have predicted the
proper links. This data will be useful both as a check on our understanding of
the relations between the data, and to identify variation between observers in
the judgements, which we will then attempt to relate to other variables, such as
political reference, social status and activity, spatial relation to the village,
their role in the ritual, age, gender and linguistic variables.

Abstracts - coding - referencing

Technical requirements

The researchers will be using two video cameras in addition to cameras (film
and digital) and tape recorders. One video camera will be fixed and the other
two will be handheld by the two researchers who will have different rights of
access and hence be in different locations. This will enable multiple views to
be recorded so that the video documentation will better capture the
multiplexity of the ritual performance itself. It is estimated that we will video
some 24-30 hours of the 72 hours that the ritual runs. In addition we will
video some of the interviews in order to capture the way that our
collaborators interact with the digitised video. In the field small clips will be
digitised using the PCMCIA plug-in digitiser, producing low quality c. 10
f.p.s. quicktime movies - these will be adequate for interviewing and will
display adequately on the powerbook screen - although even at that speed we
will be pushing the capabilities of the screen - anything other than active LED
with high refresh rates will not display video adequately so would necessitate
the further expense of an additional portable monitor. The clips will occupy
some 5-12 Mb per minute (depending on quality and sampling rate), but this is

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 10

Michael Fischer and David Zeitlyn

handball with the specified disc capacity.
The clips will be used to form the foci of interviews to be held locally and in
other villages with groups (Kwanja and Tikar) who have similar rituals to
Nggwun. The software that Fischer has developed in other funded research
will allow the way in which informants examine and interact with the video to
be recorded, making the interaction itself available for analysis.
The two researchers have extensive field experience and technical competence.
The facilities of CSAC (once enhanced by the new digitising board) will be
sufficient for the technical requirements in the post-field phase (granted the
MO storage discs included within the consumables budget).
Outcomes
The final products of this project will comprise a set of annotated and
documented digital videos in the form of ëintermediaí documents which supply
data and the context for its understanding (interview material, still
photographs, specifications, analytic commentaries, coding schemas and
working papers), and papers to be published in mainstream anthropology
periodicals such as the Journal of the Royal Anthropological Institute and
American Anthropologist. Such papers will report on the specific ethnography
and advance the understanding of ritual in general. Further papers will
describe our advances in using visual time-based material both in and out of
the field. The research will also contribute to an existing long-term
programme of research into the formal representation of time-based material.
Working papers and reports will be published electronically at the UKC
WWW site which is one of the most important access points for anthropology
on the Internet. It has a large constituency of academic and non-academic
users and is indexed by all the major indexing services. The WWW site will
provide supporting data for material published in conventional means (e.g. as
has already been done in at least three cases: Ellen 1993, Fischer 1994b, and
Zeitlyn 1993 which have links to material at our WWW site).
The budget includes provision for travel to two of the major collections of
Mambila artefacts in USA (Metropolitan Museum, New York and Portland
Art Museum, Portland, Oregon to integrate our findings with the documents
(critically the photographs archived in New York and the film archived in
Portland) made by Paul Gebauer an early missionary who was in Cameroon in
the late 1930s and 1940s. Following this we will produce a CD version of the
final report in which research analysis will be combined with digital video as
an ëintermediaí document. This will be sent to major centres of the use of
video analysis (such as Xerox Parc) the major museums holding Mambila
collections and elsewhere such as centres of visual anthropology in UK,
Europe and USA.

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 11

Michael Fischer and David Zeitlyn

References

Boissevain, Jeremy. ed. 1992. Revitalizing European Rituals. London:
Routledge.
Borofski, Robert. 1987. Making History. Pukapukan and anthropological
constructions of knowledge. Cambridge University Press.
Cannadine, D. & S. Price (eds) 1987. Rituals of royalty: power and
ceremonial in traditional societies. Cambridge: C.U.P.
Chilver, E.M. & P.M. Kaberry. 1971. The Tikar problem: a non-problem.
Journal of African Languages 10(2), 13-14.
Ellen, R. 1993. The cultural relations of classification. Cambridge: CUP.
Fischer, M. 1994. Modelling Complexity and Change: Social Knowledge and
Social Process. in When History Accelerates: essays on the study of rapid
social change, ed. C. Hann. Athlone.
Fischer, M.D. 1994b. Applications in computing for social anthropologists
(ASA Research methods in social anthroplogy). London: Routledge.
Fischer, M.D. and W.L. Lyon 1994. Tradition and Symbol in the Cook
Islands, CSAC Ethnographics Gallery, (on the World Wide Web)
http://lucy.ukc.ac.uk/tradition.html
Fowler, I. & D. Zeitlyn. forthcoming. Introductory Essay: the Grassfields and
the Tikar. In African Crossroads (eds) I. Fowler & D. Zeitlyn. Oxford:
Berghahn.
Gausset, Q. 1994 Fieldnotes.
Goody, J.R. 1977. Against ìritualî: Loosely Structured Thoughts on a Loosely
defined Topic. In Secular Ritual (eds) S.F. Moore & B.G. Myerhoff.
Amsterdam: Van Gorcum, Assen.
Hobsbawm, Eric, and Terence Ranger, eds. 1983. The Invention of Tradition.
Cambridge University Press
Humphrey, C. & J. Laidlaw. 1994. The archetypal actions of ritual: an essay
on ritual as action illustrated by the Jain rite of worship (Oxford studies in
social and cultural anthropology). Oxford: Clarendon Press.
Jolly, Margaret, and Nicholas Thomas, eds. 1992. The Politics of Tradition in
the Pacific. Oceania 62(4)
Keesing, Roger, and Robert Tonkinson. 1982. Reinventing Traditional
Culture: The Politics of Kastom in Island Melanesia. Mankind 13(4).
Larcom, Joan. 1982. The Invention of Convention. Mankind 13(4):330-7.
Morphy, H. 1994. The Interpretation Of Ritual - Reflections From Film On
Anthropological Practice. Man 29(1), 117-146.
Peel, J. 1984. Making History: The Past in the Ijesha Present. Man 19:111-32.
Price, D. 1979. Who are the Tikar now? Paideuma 25, 89-98.
Price, D. 1985. The Palace and its Institutions in the Chiefdom of Ngambe.
Paideuma 31, 85-103.
Thomas, N. 1991. Entangled Objects. Exchanged, Material Culture, and the

Mambila Nggwun - the construction and deployment of multiple meanings in ritual 12

Michael Fischer and David Zeitlyn

Colonial in the Pacific. Harvard University Press. Thomas, N. 1992. The
Inversion of Tradition. American Ethnologist 19(2):213-32.
Wright, S. 1992. "Heritage" and critical history in the reinvention of mining
festivals in North-east England. In Revitalising European Rituals (ed.) J.
Boissevain. London: Routledge.
"Reconstructing Kinship or the pragmatics of kin talk". 1993, Man (n.s.)

28(2): 199-224. NB An electronic version of part of this article that
includes digitised sound recordings may be found using World Wide
Web at http://rsl.ox.ac.uk/isca/mambila/mambila.html or using gopher
rsl.ox.ac.uk within the anthropology corner.

Zeitlyn, D. 1994. Sua in Somié. Mambila Traditional Religion (Collectanea
Instituti Anthropos 41). Sankt Augustin: Academia Verlag.

Contents

Return to the Virtual Institute of Mambila Studies