Who Made What?
Methodology for Identification
By Marc L. Felix - from the book "Mwana Hiti"
Above is a photograph of the piece being discussed in the text below.
It is from Page 92 of the book "Mwana Hiti"

ill. VII/
1"Pole depicting a woman carrying an initiate, wood, pokerwork, 154cm.
National Museum of African Art, Smithsonian Institution, Washington D.C."


In trying to identify a sculpture in terms of origin two approaches are possible: the stylistic and the typological. The ideal procedure combines both.
This dual method is easy enough to apply to a well-known and well-documented object or area, for the answers are usually to be found in the
existing literature. With a "mystery piece," however, identification is not so easy and the clues can be misleading.

As an exercise, let us consider a large staff-like sculpture surmounted by a female figure in the collection of the National Museum of African Art in
Washington, D.C. It used to belong to
Ernst Anspach, and has been described at one time or another as a mother-and-child staff from the
Ovimbundu of Angola or from the Bijogo (ill. VII/1). In fact, the object is not from Angola or from Guinea Bissau, nor is it a mother-and-child
figuration, nor is it a staff (in the sense of being an implement to assist walking).

The exercise would proceed as follows:
1. Identify the general type - a stick surmounted by a figure.

2. Identify the precise subtype — it has two suspension hooks.

3. Identify the subject or symbols depicted - it is topped by a standing female carrying another figure on its back.

4. Then, drawing on hands-on experience if possible, or on the iconography in the literature if this is reliable, make a guess as to what large group
of peoples use this type of object. In this case the evidence points to the matrilineal Bantu of eastern Tanzania.

5. Then look at the general style — in this case idealized realism, not extreme stylization, which suggests a Kwere, Kutu, or Luguru origin.

6. Consider the dimensions and proportions — the figure measures 53cm, and is one-third of the entire object. If the figure is proportionately this
large or larger, the object was more likely a pole than a staff. Its thickness also implies this, since these poles were temporarily put in a hole situated
in the center of a ritual enclosure, for initiations, exorcism, and healing.

7. Identify the material used — hard yellow wood in this instance (one arm is in a different wood and was probably a replacement).

Once we have assembled these elements, we can narrow our focus to a smaller zone - Kisarawe district. Then we look at specifics:

1. Consider the anatomical details, such as the shape of the eyes, mouth and ears, and also the coiffure and scarification, since these are details
that undergo revealing changes over time. In terms of the piece we are discussing, similar scarifications can be found on a sitting figure collected
around 1910 in Kwere country, in the Kisarawe district, that is now in the Berliner Mission in East Berlin.

2. Consider the appearance — shiny, not crusty, suggesting that the piece was collected early in its life and had not been used much.

3. Consider the general color of the wood — yellow - and the techniques employed - pyrogravure and pokerwork.

4. Analyze the depicted icon — an older woman carrying a younger one piggyback.

Following this procedure systematically, step by step, we can come up with a fairly precise answer: we can conclude that the object under
consideration is a post in the shape of a large staff mainly used in girls initiation ceremonies among the southern Kwere which live close to the
northern Zaramo, and it was probably made around 1910 and collected shortly thereafter This process also makes it clear why the object is not an
Ovimbundu mother-and-child figurative staff, as previously thought.

In other circumstances further clues will have to be considered:

1. the decorative design patterns - Muslim inspired;

2. the pigments used — iron oxide or camwood type of kaolin or ocher;

3. the various addenda, such as cloth strips kitenge or wax; the type of animal used for pelts am horns; the identity and geographic sources of
shells and coins; the type and color of attached beads (indicating the Atlantic or the Indian Ocean trade route as the source); etc.


Because the different peoples that make up the area we are studying did not diverge from their common origin until two hundred years ago, we still
find strong stylistic similarities and analogous typologies. This makes identifying the precise ethnic community of origin that much more challenging.
Until very recently it was generally agreed that East Africa was an area almost without figurative sculpture. For the African art loving public, little
information has been available. G. W. Hartwig wrote some articles on the subject and H.-J. Koloss produced an essay for the catalogue of the
Ostafrika exhibition that visited German museums in 1974, but only Ladislav Holy has devoted a whole, fully illustrated book to central and
southeastern African sculptures. Now and then a picture with an uninformative caption appeared in an African Art book, but most books avoided the
subject altogether or devoted only a few lines to East Africa. Some well-known authors even claimed that Tanzania was devoid of sculpture.
A wide variety of East African pieces have recently been surprising the world market. Some have come out of Mozambique, Kenya, Zambia, and
Malawi, but most are from Tanzania. Suddenly, scholars have been faced with a number of "new" pieces of definite aesthetic and cultural
importance. These pieces pose a double challenge as their style is often unfamiliar and their typology hard to classify. Some anthropologists and
historians of African art instinctively knew that they were faced with East African works of art; some recognized the Tanzanian origin, and others
narrowed the style down to "Zaramoid." Others identified these sculptures, as coming from distant countries, if not continents. This is not
unexpected since many of these hitherto unknown objects have their stylistic or typological origin in faraway places (see "The Search for the
Source," p. 144 ff.).

Because I have been fortunate enough to see and handle a large number of these as yet unidentified or unlocated pieces, both in the villages in
which they were used and in museums and private collections, I feel confident enough to attempt the identification and geographic location of these
sculptures (see "Stylistic Diagnoses," p. 112ff.).

As a starting point we took a number of sculptures of identical typology, mwana hiti figutines, which is why this is the title of this book. Most
probably, all of them were made about the same time, and we can positively identify the village in which some were found, which usually means they
were made there. Having located a number of pieces and situated them on the map, we can then extrapolate by applying Olbrechts's triangulation
method. By this method person with a trained eye starts from known styles it points A and C so as to guess what style B must look like. Using this
technique, we were able to identify some Zaramo, Kwere, Luguru, Kami, Kutu, Nguh Kaguru, and Doe mwana hiti figurines. Once we ha physically
placed these pieces on the floor in geographical relation to each other, some patterns emerge. Stylistic traits and evolutionary paths became
evident Next we looked at objects of different types, sue as staffs, fly-whisks, and gourd stoppers, that showed the same stylistic traits or details as
the localize mwana hiti figurines. We then established a tentative typology for each ethnic unit based on actual pieces o photographs in our

Once we had established this ethnic unit's actual typology we made a list of the pieces that would also have to exist for an ideal typology to be
completed. The frame of reference was a list of all types of objects known to have been made by the various neighboring peoples in the larger area
under study.

We then had photographs made of counterparts of these missing types that were used in neighboring groups. We showed the photographs of
these counterparts (backed up by verbal descriptions and drawings) to selected elders (non-Muslim and non-practicing Christian) in the villages of
the ethnic unit under investigation (ill. VII/3) and asked them if:

a) they had ever seen pieces of a similar type;

b) they still possessed similar pieces in their village; and

c) if the answer to a) or b) was yes, what was the name of the object, its function, and so forth.

These questions were asked both in Swahili and the local language or dialect, and the interviews were taped along with their simultaneous
translation. Index cards were filled in on the spot. We repeated this process in a number of old traditional villages among the Kwere, Doe, Kutu,
Zaramo, Luguru, and Zigua. We were thus able first to draw up an ideal typology and then to confirm our tentative ideas about stylistic identities
and the physical traits assignable to them. We then showed our findings to the elders to see if they agreed. We later cross-checked the style with
the typology to try to understand who made what, why, and in which style or fashion, and this allowed us to make a tentative stylistic attribution (see
"Stylistic Zones of the Tanzanian Hinterland," p. 105ff., and "Stylistic Diagnoses," p. 112ff.). We now know whether an object comes from the north,
the south, the east, or the west of the general area, but giving precise ethnic identification can sometimes be difficult because of the various
overlaps and hybridizations, and some question marks will be inevitable."

Marc L. Felix - Mwana Hiti - 1990