The Rock Art of Iringa Region, Southern Tanzania

A Descriptive and Comparative Study

Master's Thesis, 2013

113 Pages

Free online reading





Table of Contents

List of Tables

List of Figures

List of Plates

List of Abbreviations and Acronyms

1.1. Background to Rock Art Researches in Tanzania
1.2. Statement of the Problem
1.3. Research Objectives
1.3.1. Main Objective
1.3.2. Specific Objectives
1.4. Hypothesis
1.4.1. Null Hypothesis Test Implication
1.4.2. Alternative Hypothesis Test Implication
1.5. Significance of the Study
1.6. Limitation of the Study
1.7. Organization of the Study

2.1. Introduction
2.2. Location of the Study Area
2.3. Geology
2.4. Topography and Landscape
2.5. Vegetation
2.6. Modern Climate
2.7. Drainage Patterns
2.8. Settlements and Land-use Systems
2.9. Previous Archaeological Researches in Iringa Region

3.1. Introduction
3.2. Rock Art of Tanzania and Evolution of Interpretative Schemes

4.1. Introduction
4.2. Research Design
4.3. Sample, Sample Size and Sampling Procedures
4.4. Data Collection Methods
4.4.1. Ethnographic Inquiries and Interviews
4.4.2. Surveys
4.5. Recording, Documentation and Rock Art Sites Mapping
4.6. Excavation Procedures

5.1. Introduction
5.2. Surveys Results
5.2.1. Mlambalasi Hill Mlambalasi 1 Mlambalasi 2 Mlambalasi 3 Mlambalasi 4 Mlambalasi 5
5.2.2. Lutona Rock-shelter
5.2.3. Tavimienda Rock-shelter
5.2.4. Kihessakilolo Rock-shelter
5.3. Overall Assessment of the rock art State of Preservation
5.4. Ethnographic Inquiries Results
5.5. Excavations Results
5.5.1. Excavation Unit A6
5.5.2. Excavation Unit B5
5.5.3. Excavation Unit B6
5.5.4. Stratigraphy

6.1. Introduction
6.2. Lithic Analysis
6.3. Lithic Raw Materials
6.3.1. Typological Classification of Lithic Artifacts Whole Flakes Shaped Tools Scrapers Points Backed Pieces Cores
6.4. Bones
6.5. Land-snail Shells
6.6. Red ochre

7.1. Introduction
7.2. The Rock Paintings
7.3. Order of Superimposition
7.4. Traditions and Styles
7.4.1. Traditions Hunter-Forager Rock Art Tradition Bantu-Language Speaking Tradition
7.4.2. Styles Subject Matters Zoomorphic Figures Anthropomorphic Figures Schematic Human Figures in Crude White Frieze Reptiles, Birds and Insects Geometric, Abstract Figure and Script-like Design Set of Quadrangular Shapes Circular, Hemispherical, Oval-Shapes and Sun-like Designs . Triangular Shapes Dots Lines, Scaffoldings/Ladder/Millipede/Centipede like Patterns Script-like Signs

8.1. Introduction
8.2. Iringa Rock Art Traditions
8.2.1. The Hunter -Forager Bolded Red/Filled-in Red Tradition
8.2.2. The Hunter Forager Red -Geometric Tradition
8.2.3. The “Bantu Language-Speaker” Art Tradition
8.3. Present and Ancient Use of the Rock-shelters
8.4. Concluding Remarks
8.5. Recommendation for Future Studies



Table 1: Inventory of Recovered Materials from Excavation Test Pit A6 50

Table 2: Inventory of Recovered Materials from Excavation Test Pit B5 51

Table 3: Inventory of Recovered Materials from Excavation Test Pit B6 52

Table 4: Classification of Lithic Artifacts 57

Table 5: Identified/Diagnostic Fauna Remains from Kihessakilolo Rock-shelter 61


Figure 1: Map of Tanzania Showing Location of Iringa Region

Figure 2: Climate Data for Iringa

Figure 3: Map of Iringa Region Showing Rock Art Sites

Figure 4: A Cross-Section of Tavimienda Rock-shelter

Figure 5: Kihessakilolo Rock-shelter Excavation Units Layout

Figure 6: Northern Wall Profile for Excavation Unit B6

Figure 7: Lithic Raw Materials Distribution

Figure 8: Lithic Artifacts Illustration: a-Whole flake; b and c-Trimmed/Utilized flake.

Figure 9: Re-redrawing of Bantu Rock Art at Tavimienda Rock-shelter

Figure 10: Ladder like Designs and SGA Signs at Tavimienda Rock-shelter

Figure 11: Alphabet-like signs from Iringa Region compared to some Ancient Script..


Plate 1: Natural Vegetation of Iringa

Plate 2: Survey Team at Tavimienda Rock-shelter

Plate 3: Tracing the rock paintings at Kihessakilolo Rock-shelter

Plate 4: Mlambalasi 1 Rock-shelter

Plate 5: A Grinding Stone and a Pestle Rubber inside Mlambalasi 1 Rock Shelter

Plate 6: Mlambalasi 2 Red-geometric Rock Paintings

Plate 7: Late White Paintings at Lutona Rock-shelter

Plate 8: The location of Tavimienda Rock-shelter

Plate 9: Dusky White Paintings at Tavimienda Rock-shelter

Plate 10: Rock Paintings at Kihessakilolo Rock-shelter

Plate 11: Grafitti on Rock Paintings

Plate 12: Identifiable Fauna Remains from Kihessakilolo Rock-shelter

Plate 13: Ochre Pencils with Polished and Striated Surfaces at Kihessakilolo rock-shelter

Plate 14: Order of Superimposition

Plate 15: Naturalistic and Semi-naturalistic Paintings at Kihessakilolo Rock-shelter

Plate 16: A Panel of Head-dress at Kihessakilolo Rock Shelter

Plate 17: Lizard/Alligator and Insect Paintings

Plate 18: SGA signs at Tavimienda Rock-shelter

Plate 19: Complex Patterns Executed at Tavimienda rock-shelter

Plate 20: Sun/Sunflower-like and Oval/Circular Patterns at Tavimienda Rock-shelter

Plate 21: Triangular Dusky White Designs at Tavimienda Rock-shelter

Plate 22: Symbolic Representation on Nyaturu people ceremonial gourd in central Tanzania


illustration not visible in this excerpt


I am deeply indebted to my supervisor Dr. Pastory Magayane Bushozi for his guidance, assistance, support, and friendship during my entire period of graduate studies. He afforded me with an opportunity to conduct my MA study research in Iringa Region.

I would also like to thank Dr. Charles B. Saanane, the Coordinator of the Archaeology and Heritage Studies Programs and Dr. Emanuel T. Kessy who abridged well my academic sponsorship from World Bank Project CIA.6.2. Thanks so much for pushing me to expand my knowledge base by your guidance throughout the process of my postgraduate studies, and for airing your comments on early drafts of this dissertation.

Many thanks are also directed to Prof. Audax Z. P. Mabulla for his regular assistance in the identification and descriptions of the rock paintings. His constructive critiques and sharp eyes put on my work are wholeheartedly appreciated.

I will be forever indebted to my field research assistants Mzee Saidi Kilindo, Mzee Audax Korongo, Ms. Irene Musa, Mr. Baraka Mtewele and Mr. Aloyce Mpanduji. I cannot repay your priceless support you afforded me. I pray for the Almighty God so that he can repay you abundantly in all your endeavors.

I would like to extend my heartfelt gratitude to Dr. Elizabeth Kyazike for her critical and sharp eyes in editing this work and making the needful corrections. Mr. Richard Bigambo you helped me in arranging the photos in a more precise and correct order I owe a big debt to you.

Lastly, I should thank all the staff of Igeleke Primary School and those people of the study area who offered me with such a warm reception and cooperation during the entire period of fieldwork.


To my Parents

Mzee Petrus Mughwae Daniel Itambu, Mama Elizabeth Mathias Ikimbia Ghuo and Dada Maria-Nyalimu Itambu for your endless love, care, passion, encouragements, motivations, and supports; and, for that I will be forever appreciative!

(Mwajîfya inno afafi wanē, njarîyi káangî nnēnatîeüng’ongo ă ikhăkha!!)


The rock art of Iringa Region was firstly reported by Prof. Pamela Willoughby who led a research team that discovered important archaeological occurrences including Middle Stone Age (MSA), Later Stone Age (LSA) with animal and human remains, rock art and Iron Age sites. Based on their research objectives and other constraints no detailed study was conducted on the rock art. This research conducted in Iringa Region aimed to survey, document and record in detail the rock art sites. It discovered two new rock painting sites and also studied two sites that were discovered in 2006 by Willoughby and her team. The rock art of Iringa belongs to two rock art traditions: Hunter-forager and Bantu-speaking art traditions, the former is dominated by naturalistic animal and human figures executed in dark-red pigment while the latter consist of schematic animal and human figures as well as geometric designs executed in white colour. The comparative study show that the rock art of south, central and north central Tanzania share same traditional motives. It was revealed in this that only two rock-shelters with rock paintings are well preserved, the rest are in a poor state of preservation. The major threats affecting these priceless and none renewable resources include the anthropomorphic, natural agents. These rock paintings need to be preserved for future generation because of their cultural, scientific and economic values.


1.1. Background to the Rock Art Researches in Tanzania

Our current knowledge of rock art paintings in Tanzania mainly comes from central and north-central Tanzania as well as the Lake Victoria Basin where so far over 200 sites have been identified, constituting more than 550 painted rock-shelters (Fosbrooke, 1950; Fozzard, 1959; Leakey, 1983; Mabulla, 2005; Masao, 1976, 1979; Mturi, 1998).

Majority of the rock paintings from these main clusters are attributed to the hunterforagers and are dominated by large animals, hunting scenes and images of dancing humans. Most archeologists agree that this is the earliest rock art tradition in subSaharan Africa. According to Anati (1986) the rock art of Sub-Saharan Africa date back to about 40,000 BP. But some scholars do believe that the hunter-gatherers paintings date between 20,000 to 1,000 BP (Masao, 2007).

During the colonial period European travelers, missionaries and colonial administrators, reported the existence of rock art, archaeological sites and archaeological artifacts in different parts of the country. The existence of rock paintings was first recorded in 1908 by missionaries in Bwanjai, in Kagera Region (Mturi, 1998). In 1923, Bagshawe then a District Commissioner reported some rock-shelters with paintings in an area west of Lake Eyasi (Leakey, 1983; Mturi, 1998). In 1929, T. A. M. Nash published a short paper on paintings found in Kondoa District. In 1931, Mr. A. T. Culwick reported on the discovery of more sites with rock paintings in central Tanganyika. During that period, Henry Fosbrooke and his wife also recorded and documented numerous rock shelters with paintings in Kondoa District (Fosbrooke, 1950). Reports from European travelers, missionaries and colonial administrators group, however, were largely descriptive and focused only on location and content of the paintings and did not give meanings of the rock art (Bwasiri, 2007; Masao, 1979; Mturi, 1998). However, some reports made an effort to categorize the paintings according to styles, colors and established typological descriptions (Fosbrooke, 1950; Fozzard, 1966).

In the 1930s systematic documentation started in different parts of Tanzania marked by the work of Louis and Mary Leakey who briefly visited rock-shelters with paintings in Kisese and Cheke areas in Kondoa District and identified about thirteen styles of rock paintings (Mturi, 1998). Between 1934 and 1936, Ludwig Kohl-Larsen and his Wife Margit Kohl-Larsen conducted archaeological investigation along the Lake Eyasi Basin, whereby a number of rock art sites were discovered and documented (Kohl-Larsen, 1938). Later in 1951, Louis and Mary Leakey conducted detailed systematic surveys, and documented rock-shelters with paintings located on the eastern side of the Maasai escarpment. Through this Leakey identified other four styles leading to 17 stylistic descriptions (Leakey, 1983). However, results from these works remained unpublished until 1983, when were published in the Africa’s Vanishing Art: The Rock Paintings of Tanzania (Leakey, 1983).

During the post-colonial period (1965-1980), intensified archaeological and rock art studies continued being directed towards the same areas previously reported by colonial agents (Bendera, 2011; Bwasiri, 2007;Kwekason and Chami, 2003; Chaplin, 1974; Leakey, 1983; Mabulla, 2005; Mahudi, 2008; Masao, 1979, 2005, 2007; Mturi and Bushozi, 2002). In 1976, more extensive surveys on rock paintings in Mpwapwa, Kondoa, Manyoni and Iramba Districts were conducted by Fidelis Masao (Masao, 1979). Masao (1979) documented the rock art sites at KwaMwango-Isanzu and Kirumi Isumbirira in Iramba District as well as Kandaga A9 and Majilili 2B sites in Kondoa District.

Recently, local researchers have expanded rock art researches outside the three main clusters Tanzania in Bukoba Rural and Urban Districts (Mturi and Bushozi, 2002), Muleba District (Chami and Kwekason, 2004), the Lake Eyasi Area (Mabulla and Bushozi in preparation) and in Mara Region (Mabulla, 2005). Recent archaeological investigations by Bendera (2011) in Manyara and Arusha regions, northern Tanzania have documented more rock art sites with majority paintings attributed to hunter-forager community. The paintings are mainly depicted in red color and characterized by naturalistic animals, stylized humans, and geometric motifs (Bendera, 2011). Similar works on rock-shelters although not focused on rock art, but very useful and relevant in the understanding of past human behavior have been recently conducted in Babati and Singida districts by Elinaza Mjema (Mjema, 2008). Mjema succeeded to establish the culture history of Endadu rock-shelter ranging from the LSA to historical period. At Endadu the LSA was radiocarbon dated between 21,000 and 8,100 BP (Mjema 2008). It is possible that the LSA people who occupied Endadu rock-shelter were also responsible for the rock paintings recorded in this area.

1.2. Statement of the Problem

Our current knowledge of rock art in Tanzania mainly comes from central and north central Tanzania as well as the Lake Victoria Basin (Bwasiri, 2007; Kwekason and Chami, 2003; Chaplin, 1974; Leakey, 1983; Mabulla, 2005; Masao, 1979, 2007; Mturi and Bushozi 2002; Mahudi 2008; Bendera 2011). From these main clusters, a number of rock art sites were initially reported and recorded by colonial administrators such as missionaries, travelers and traders (Bagshawe, 1923; Chaplin, 1974). During the post- colonial period (1965-1980), intensified archaeological studies continued being directed to areas previously reported by colonialists and/or colonial agents (Masao, 2005). The tendency of designing research projects based on reports of pioneers made other places of Tanzania including Iringa to remain less-known archaeologically. Until recently, Iringa was only known for the late Acheulian occurrences at Isimila. Nonetheless, recent researches in Iringa by Canadians and Tanzanian scientists have discovered a range of archaeological sites including painted rocks (Biittner, 2011 and Bushozi, 2011; 2012 and Willoughby, 2009; 2012).

Even though rock art sites were reported by this research team, no detailed study has been directed towards the rock art sites (Bushozi, 2011). This realization led to the conception of this study. Rock art is an important cultural heritage asset that can give more insights into prehistoric people’s mind, culture and daily life. Yet, rock art is very vulnerable to anthropogenic, physical and biological agents of deterioration. Therefore, locating, recording, documenting and assessing their cultural significance and evaluating potential threats to their long-term survival become the research problem of this study. Yet, Paintings may not have been separated from other prehistoric cultural activities such as tool use and manufacture, food procurement strategies and occupation of rock shelters. Therefore, excavating some painted rock shelters become an important aspect of this study. Thus, rock paintings and archaeological materials were tied together to develop the cultural history of Iringa.

1.3. Research Objectives

1.3.1. Main Objective

The main objective of the study was to record, document, map out, and assess the potential threats of the rock art sites of Iringa Region, southern Tanzania.

1.3.2. Specific Objectives

i. To identify, document and describe in detail the sites with rock paintings;
ii. To evaluate the condition of the rock paintings and the source and potential threats to their integrity and survival;
iii. To assess the cultural significance of rock art in Iringa;
iv. To establish the culture-history of the rock art sites through archaeological excavations;

1.4. Hypothesis

This study had one null hypothesis, one alternative hypothesis and two test implications.

1.4.1. Null Hypothesis

The rock paintings of Iringa Region are dissimilar/different in comparison with those of central and north-central Tanzania. Test Implication

I expected to find/observe differences in rock art affinities, subject matter, composition, styles, color and age indicating that they were produced by groups of people that were not connected with those of central and north central Tanzania.

1.4.2. Alternative Hypothesis

The rock paintings of Iringa Region are similar and share affinities, subject matter, composition, styles, color and age with those located in central and north central Tanzania. Test Implication

I expected to find/observe similarities rock art affinities, subject matters, compositions, styles, colour and age indicating that they were produced by groups of people that had a relationship with those of central and north central Tanzania.

1.5. Significance of the Study

This is the first systematic archaeological study on rock art carried out in the southern highlands of Tanzania. The study has managed to shed lights to the current understanding of rock arts sites in Iringa Region and that the rock paintings are not limited to sites documented in central and north-central and Lake Victoria basin in Tanzania.. The research further showed that the rock art of Iringa Region is similar to that of central Tanzania in many aspects including the fact that they are made up of paintings executed predominantly in two pigments, i.e., shades of red and white like paintings of Singida Districts (Masao, 2007). They also share many similarities that tempt to conclude they belong to the same tradition and probably painted by artists with similar culture. Geographically, there are no major ecological and geomorphologic barriers between central and southern Tanzania. The new knowledge radiating from this study will possibly attract more researchers and tourists who will bring more funds for the sites promotion, protection and preservation. Information generated from this study will be helpful for cultural heritage policy formulation and providing a platform for making decisions.

1.6. Limitations of the Study

This study encountered some constraints especially during the ground reconnaissance surveys which varied from logistics, time and topographic nature of Iringa Region, whereby the latter hindered reliable transport. The four weeks for the research were inadequate to enable the researcher to accomplish the task due to the mountainous nature of Iringa Region. To circumvent the problem, two more local informants joined the research team and curb the financial and time constraints only three districts of Iringa Urban, Iringa Rural and Kilolo districts were covered. The vastness of the sample size and of the region and remoteness of the areas with paintings was another stumbling block. For instance, some of the areas covered by Precambrian rocks were vast and inaccessible either on foot or by motor vehicles yet the exposed boulders were very difficult to reach or climb due to the rough terrain.

In addition, there was serious resistance from the local people during surveys, interviews and ethnographic inquiries. This was because the local people had divergent speculations on the intentions of the research. Some thought that the purpose was treasure hunting which made them reluctant to offer full cooperation. After sensitization they assisted in locating some sites. However, due to the lack of knowledge, in some instances, they showed us rocks with modern paintings (graffiti), or big boulders with dirty-lined strips that might have resulted from oxidation, weathering, hyrax excrements and other biological factors. Despite all these constraints, the study have succeeded to record, document and map out some potential rock art sites at Kihessakilolo, Mlambalasi, Lutona and Tavimienda rock-shelters.

1.7. Organization of the Study

This dissertation consists of eight chapters: Chapter one gives introductory remarks on the background information about the rock art studies conducted in Tanzania. The first chapter defines the research objectives, hypotheses, significance and the limitation of the study. Chapter two presents the general overview on the study are with special emphasis on the environment, geology, and previous archaeological. Information related to geology, soils, landscape, vegetation, topography and drainage patterns, settlements and land-use systems across the region are widely explained. This chapter also discusses the prehistoric human activities and settlements patterns. Chapter three deals with literature review on evolution of theories and schemes used in studying rock art. Chapter four outlines the research methods employed in data collection. Chapter five presents field works results while Chapter six presents excavation results. Chapter seven is presenting analysis and the classification of rock paintings. Chapter eight consists of discussion, concluding remarks and the recommendation for further studies.


2.1. Introduction

This chapter gives the general overview of the research area, with much emphasis on the geography, geology, and previous research investigations. Information related to geology, soils, landscape, vegetation, topography and drainage patterns, settlements and land-use systems across the region are widely explained. This chapter also discusses the prehistoric human activities and settlements patterns.

2.2. Location of the Study Area

The study was conducted in three districts of Iringa Region namely Iringa Urban, Kilolo and Iringa Rural. Geographically, Iringa is located in the southern highlands of Tanzania lying between 7º 46'12" S and 35 º 41' 24’’ E and covering an area of about 58,936 square kilometers (Figure 1). The region is surrounded by three major mountainous ranges of Kipengere and Livingstone mountains to the south and the Udzungwa Mountain to the northeast.

illustration not visible in this excerpt

Figure 1: Map of Tanzania Showing Location of Iringa Region (Source: Google

2.3. Geology

Geologically, the bedrock of the Iringa Region is composed of Precambrian migmatites, granite and Konse group outcrops formed during the late Archean and late Neoproterozoic eras (Harpum, 1970; Harris, 1981; Howell et al., 1962; Kröner, et al., 2001). The Precambrian granite rocks represent reworked remnants of the Tanzania Craton Archean that form most of the rock-shelters and overhangs across the landscape of Iringa Region (Harpum, 1970). It is these rock-shelters and overhangs served as readymade prehistoric human houses and offered refuge for MSA and LSA people.

Much of the geological and geomorphologic data about Iringa Region came from the ancient lake and exposed gully at Isimila (Cole and Kleindienst, 1974; Hansen and Keller, 1971; Howell, et al. 1962; Bushozi, 2011). The sedimentary deposits at Isimila indicate that the site was created in a deltaic environment (Hansen and Keller, 1971; Bushozi, 2011, 2012). The archeological deposits are found in the exposed overlaying series of progression semi-arid gullies that are characterized by alluvial sands, silt-clay, mudstone and claycrete (Howell, et al. 1962; Bushozi 2011, 2012).

In previous studies, the geo-cultural chronology of this place was divided into two geological strata that are: Lisalamagasi and Lukingi members (Cole and Kleindienst, 1974; Hansen and Keller, 1971; Bushozi, 2011). The Lukingi Member sediments are characterized by clays, sand clays and sandstones. The sediments underlying the Lukingi Member represent the earliest deposits and are composed of paleosol, colluvial and claycrete deposits that overly the granite bedrock (Cole and Kleindienst, 1974; Hansen and Keller, 1971; Biittner, 2011; Bushozi, 2011). The depositional processes revealed at Isimila suggest that for the most of the Pleistocene the regional was highly precipitated than it is at the present.

2.4. Topography and Landscape

The topography of Iringa is generally dominated by highlands and escarpments ranging from 1200 m to 2700 m above sea level in the southern part (Regional Commissioner Office Iringa, 2007). The lowland areas range from 900 m to 1200 m above sea level and are mainly found in the northern part of Iringa (Bushozi, 2011). As noted above, the most prominent features of the landscape of Iringa Region are the numerous kopjes or castle kopjes. A kopje is a “steep-sided pile of massive crystalline boulders” that was formed by the collapse of the bornhardts or inselbergs (Buckle, 2007; Biittner, 2011). Inselbergs are high steep-sided dome or table-shaped hills formed by pediplanation in arid and semi-arid areas or by deep weathering and stripping in forest and savannah regions (Biittner, 2011). Cycles of subsurface weathering create weathered rock debris which is stripped away. This weathering and stripping process leaves behind fresh un- weathered rock called an inselberg (more specifically bornhardts). While deep weathering continues, surface erosion begins to attack joint systems and cracks are formed within the un-weathered rock (Biittner, 2011). If the joint system and weathering are extensive enough, then the bornhardt may collapse into kopjes; therefore, kopjes are thought to evolve both directly by deep weathering and indirectly by the collapse of bornhardts (Buckle, 2007). Kopjes are of high archaeological potential as they form natural rock shelters inhabited by prehistoric people.

2.5. Vegetation

The vegetation cover strongly reflects the topography, rainfall and human activities (Plate 1). It is heterogeneous composed by montane forests, shrubs and woodlands (Minja, 1991). The first group entails the montane forests that are characterized by tree species such as Euphorbia sp., Acokanthera laevigata, Cassipourea malosana, Bersama abyssinica, turrea holstii, Halleria lucida, Macaranga kilimanscharica,Ekerbigia capensis, Podocarpus sp.,Rapanea, Schefflera, umbellifera etc (Minja, 1991; Kimaro, 2008). The lowland is covered by shrubs such as; Clausena anisata, Coffea mufindensis as well as sparse trees in particular, Alangium chinense, Albizia gummifera, and Chrysophyllum gorungosanum (Minja, 1991).

illustration not visible in this excerpt

Plate 1: Natural Vegetation of Iringa

2.6. Modern Climate

The climatic conditions of Iringa Region are mainly influenced by its relief features such as mountains and mountain ranges (Figure 2). These topographic features and vegetation cover influence climate greatly, resulting in micro and macro variations across the region. Generally, the climatic conditions of Iringa Region can be subdivided into three distinctive zones: the highlands, midlands and lowlands. The highlands zone lies at an altitude of 1,600 - 2,700 m above sea level. This area includes the eastern fringes of Iringa Rural and Mufindi Districts, the central and eastern part of Njombe, Ludewa and Makete districts. Temperatures are normally below 15° C with rainfall ranging between 1,000 and 1, 600 mm per annum, falling in a single season from November through May (Figure 2). The dry and cold season lasts from June to September occurs after the rain season. The midlands zone lies at an altitude of 1,200 to 1,600 above sea level. This zone constitutes the central part of Iringa Region covering Mufindi, parts of Njombe, Ludewa and Makete Districts. Temperatures range from 15 °C to 20° C, with average rainfall of about 600 to 1,000 mm per annum. The lowlands zone range between altitude of 900 and 1, 200 m above sea level. This zone includes the low lying northern part of the Iringa Rural District (Kimaro, 2008). Temperatures vary between 20 º C to 25º C with rainfall ranging between 500 and 600 mm per annum (Regional Commissioner Office, 2006).

illustration not visible in this excerpt

Figure 2: Climate Data for Iringa (Source: Tanzania Weather Base, 2011)

2.7. Drainage Patterns

Iringa Region forms part of the Indian Ocean drainage zone, its drainage pattern is dendrite which can be properly described as pinnate, with enlarged amphi-theatres at valley heads (Porter, 1991). The Great and Little Ruaha River dominates the drainage system across Iringa Region and joins the Rufiji River that flow eastwards to Indian Ocean. The flow of the Ruaha River is permanent, one remarkable incident happened in 1997 when the river became seasonal due to severe drought, but the situation ceased in 2000 (Kimaro, 2008). Currently, there is marked water flow during the rainy season. The southern part of Iringa drains into Lake Nyasa (Regional Commissioner Office, 2006).

2.8. Settlements and Land-use Systems

The landscape provides reliable settlements for human population in the region whereby the dominant inhabitants are the Hehe accounting to about 43%, of the total population.

They are mainly located in Iringa and Mufindi Districts. The Bena (37%) are the most numerous ethnic group in Njombe while the Kinga (11%) are dominant in Makete District. Ludewa is shared by smaller ethnic groups such as the Pangwa, Kisi and Manda (The Regional Commissioner Office, 2006). These ethnic groups are scattered throughout the landscape of Iringa. Mud-thatched houses dominate in rural areas while modern houses are limited in towns and urban centers.

Agriculture is the largest economic sector, followed by livestock keeping and fishing. Maize is the major staple food crop. Other food crops of great importance include: Irish potatoes, sweet potatoes and beans. Cash crops include tobacco, a crop mainly grown in Iringa Rural District. Sunflower, tea, pyrethrum and coffee grow in various places across the region but in varying proportions. Horticultural crops include onions, tomatoes, fruits and cabbages. The agricultural sector contributes over 75 % of the regional economy and employs about 90% of the working population in the region. Iringa is one of the “Big Four” regions with surplus food production of mainly maize that is exported to other regions in Tanzania. In this region maize is also regarded as a cash crop whose production surpasses consumption. Maize production accounts for about 53 % of the major food crops, followed by Irish potatoes that constitute 27%.

2.9. Previous Archaeological Researches in Iringa Region

As noted before, the archaeology of Iringa Region has been known primarily based on the findings from Isimila Early Stone Age site (Cole and Kleindienst, 1974; Hansen and Keller, 1971; Howell et al. 1962; Bushozi, 2011). Archaeological researches have demonstrated that Iringa Region has been occupied since the Acheulian period around 270,000 years ago to the present (Biittner, et al. 2007; Cole and Kleindienst, 1974). However, the age of Acheulian assemblage at Isimila need to be reconfirmed because the dating method used to date the site initially is outdated. Other Acheulian sites such as Olorgesailie in Kenya have been re-dated using new dating techniques and now the dates have changed from about 0.4 million years ago to 0.9 million years ago (Evernden and Curtis, 1965).

Recent archaeological investigations in Iringa Region mainly directed by Professor Pamela Willoughby have revealed a substantial number of archaeological sites ranging from Stone Age to historic periods (Bushozi, 2011; Biittner, 2011; Willoughby, 2012). At the present a large body of knowledge related to the ancient subsistence economy, ecological adaptation, and prehistoric mobility systems has been amassed. Biittner (2011) has examined the mobility systems and stone raw material sourcing patterns during the MSA period, and concluded that decision making and socioeconomic organization during the MSA were much influenced by the availability and accessibility of ecological resources. She commented that:

“In particular, the types of raw materials that are present in stone tool assemblages, and the sources from which they are acquired, provide information relating to decision making processes, planning, organization of technology, and group mobility. The characterization of Stone Age lithic artifact assemblages from two rock shelter sites in southern Tanzania; Magubike and Mlambalasi, allowed for the evaluation of inter- and intra-assemblage variability and raw material characterization…” (Biittner, 2011:3).

From her study chert, quartzite and basalt lithic raw materials were more preferred for stone tool making during the MSA, probably because were easily sourced and plentiful. Another recent study has been conducted by Pastory Bushozi and focused on the MSA projectile points and the way they have been used over time and space. According to Bushozi (2011) spear and arrow projectiles coexisted in southern and northern Tanzania during the MSA and LSA periods.

In 2002, Dr. Paul Msemwa, Director of the National Museum, conducted archaeological fieldwork aimed at reconstruction of the cultural chronology of Iringa Region (Msemwa, 2002). However, strong emphasis was placed on Iron Age sites owing to a theoretical emphasis to understand the factors which may have led the southern populations of Tanzania to interact with coastal peoples much later in time compared to the other peoples (Msemwa, 2002). Another important fieldwork in Iringa Region was carried out by Edwinus Lyaya (2008) and focused on the bio-metallurgy of the Bena ironworking community in Njombe District. Bio-metallurgy is the study of the biological species, both plants and animals used in the process of iron production (Lyaya, 2008). The main objective of his study aimed at identifying the impact of iron production to the environmental degradation. Results from this study demonstrated that Bena ironworkers were species-selective and that deforestation recorded in the southern highlands did not resulted from iron production as it has been emphasized in the previously (Lyaya, 2008). Even though new archaeological investigations have made a significant contribution towards the better understanding the archeology of Iringa Region themes related to rock art has been not one of the research agenda (Figure 3).

illustration not visible in this excerpt

Figure 3: Map of Iringa Region Showing Rock Art Sites (Adapted from Bushozi 2011)


3.1. Introduction

This chapter reviews some techniques and evolutionary schemes employed by archaeologists to study rock in Tanzania. It was revealed from this study that various evolution schemes were employed by various researchers across the country to analyze and describe the rock art. However, it appears that previous researchers failed to develop a common analytical scheme.

3.2. Rock Art of Tanzania and Evolution of Interpretative Schemes

The study of rock art styles, motifs, execution techniques and subject matter in Tanzania can be traced starting from the work of Louis Leakey (1936, 1950) who came up with 17 styles of rock paintings based on a few sites investigated in Kondoa, central Tanzania. However, Leakey’s model was later criticized because it was very restrictive and site specific and did not cover many sites across the country. It was believed that the rock art in other regions may represent a large amount of stylistic attributes more than it was previously estimated (Anati, 1986; Masao, 1979).

Fozzard (1959) studied six rock painting sites in the western part of Kondoa District and suggested a sequence of five styles. These are:

1. Human figures where the body filled in with paint and the head is drawn in outline with the center or face left blank with a combination of solid and open line styles;
2. Open line style where animals are drawn in thin outline only. Although the animals are well shaped, the drawings are a little crude without much detail;
3. Spot styles where shapes of animals are constructed by grouping large sports of orange vermillion paint. There is no outline in this style;
4. The finest detailed line whereby animals are drawn in thin delicate outline with details such as manes, tails and body lines, and are lifelike and full of expressions; and;
5. Solid where both humans and animals were drawn in a thick coat of paint.

Under Fozzard (1959) all styles except the “spot style” are drawn in claret red ochreous paint (see also Masao, 1979). Fozzard’s model was found more applicable, but it had a limitation of its restrictiveness when applied in a wide region.

Many other attempts to study the rock art of Tanzania were employed by Shorter (1967) who recorded and documented rock art sites in Ukimbu. In 1971, Knut Odner undertook recording and documentation of rock shelters with paintings in Iramba District (Odner 1971) and came up with a new scheme that emphasized on three styles;

1. Realistic or near realistic animals and human figures in solid red and other symbols such as geometric and amorphous figures;
2. Schematic animals or crudely depicted human beings with white as a dominant colour, while other colours are red and yellow; and
3. Symbols of handprints, suns and comb-like representations in white and black, sometimes red in color (Odner, 1971).

Odner’s scheme focused on chronological aspects, colors and subject matter. However, the scheme ignored to outline drawings, which presumably should be included in naturalistic and semi-naturalistic works (Willcox, 1984). Other scholars who recorded and documented Tanzania’s rock art include the works of Ten Raa in Usandawe area and Masao in Kondoa, Singida Rural and Iramba Districts (Masao, 1976, 1979). Masao (1979) made a synthesis based on Odner’s and Fozzard’s models that produced four chronological phases for the rock art of central Tanzania. These include:

1. Red conventionalized schematic human and naturalistic figures filled in with animals whereby pigments varied from vermillion to scarlet red;
2. Naturalistic animals executed first in open line outline profile then filled in with various motifs such that a few semi-naturalistic human figures sometimes depicted with a loin cloth also occur with red being a predominant colour but a few examples of brown and white are also found;
3 White semi-realistic silhouettes. This is a very prolific style representing both human figures and animals, predominantly naturalistic but some elements of stylization are present. Paintings are executed in various shades of white and filled in with either a thin wash or a thick paste of the same colour; and
4. Abstract and geometric figures being representations of different geometric motifs of lines, circles, square dots and non-geometric signs and symbols. They are more commonly done in white but orange, brown, red and black are also found. The more common geometrics found are simple lines, crosses, checkers, ladders and circles. Abstract figures include dragon-like pictures, symbols and dots (Masao, 1979).

Anati (1983) identified six chrono-socio-economic styles. His scheme was a by-product of 102 documented rock art sites of Kondoa District and it emphasized the typological and chronological descriptions. Anati’s scheme reflected the main economic activities of the society based on the depicted figures. Anati proposed six styles for the rock art of Tanzania starting from the later to the earliest ones:

1. Late White Bantu (Farming and Mixed Economy) in which figures are dominated by white colours. Depicted animals are commonly domesticated cattle, goats and sheep. In rare cases, there are snakes, lizards and crocodiles. The style belongs to the Iron Age period dating from early centuries of the first millennium AD, though some of the paintings might be as recent as 300 to 200 years ago;
2. Pastoral culture dominated by oxen-zebu or humped animals. They are painted in black, grey and green-gray colour. They are believed to date between 3000 BC and early centuries of the first millennium AD;
3. Stone Bowl Culture: the culture is dominated by earth and mud colours characterized by large animal figures in which elephant is mostly depicted. Other animals include a few giraffes and wild carprines;
4. Art of the Late Hunters: the dominant colours are the various shades of red. Other colours include yellow, orange, brown and violet. Painted figures are mostly in monochrome though there is a sporadic representation of bichrome and polychrome figures. Depicted figures include humans carrying bows and arrows, in hunting scenes. Particular attention was given to body decoration, hairstyles, clothing and headdresses. The eland is the most dominant animal depicted. Other depicted animals include the giraffe and antelope. It dates after 7000 BP and continued through the last millennium BC;
5. Art of Early Collectors: the main colours are various shades of red and brown. The subject matters include anthropomorphic figures engaged in dances, social activities and religious activities dated between 12,000 and 7,000 BP;
6. Early Hunter Style: this is dominated by a dark reddish brown colour. The subject matters are animals and symbols while human figures are rarely depicted and sometimes absent. It probably began as early as 40,000-50,000 BP, but flourished during the period of 30,000-12,000 BP;

Later on in the 1980s, Mary Leakey (1983) made a more detailed assessments of Kisese 1 sites concluded that the rock art of central Tanzania has 9 stylistic sequences. Mary Leakey’s stylistic scheme was arranged in a sequence from the earliest to the latest as indicated herein:

1. Black figures in thick outline, rather poorly drawn;
2. Brown-faded brownish-red finger-tip paintings such as the two groups of stylized human figures and unidentified spotted animals at Kisese 1;
3. Naturalistic paintings in fine outline executed in free-flowing lines with great appreciation of form;
4. Animals in red outline filled with yellow wash, most often used when depicting friezes;
5. Naturalistic animals in full red;
6. Naturalistic streaky style;
7. Stiff wooden animals in rather thick, brick-red-outline, often without tails;
8. Crude daubs in white, black and red; and
9. Representations of humped cattle including symbols in both black as well as white pigments (Leakey, 1983).

In the late 1980s, Imogine Lim surveyed and documented the rock art in Usandawe land using a site-oriented approach. She managed to study the relationships between paintings, site and the surrounding landscapes (Lim, 1992). Later on, similar related research projects were conducted along the Lake Victoria Basin, especially the western and southern of the Lake shore. The region was found endowed with paintings attributed to the Bantu or early Iron Age community (Mturi, 2001; Kwekason and Chami, 2003).

In Lake Victoria Basin, stylized cattle are the most dominant motif, especially in Bukoba Rural, Bukoba Urban and Muleba districts. There are also executions of figures that represent lacustrine adaptations, such as fish, canoes and/or boats (Mturi, 2001). Another rock art research project termed as Mara Region Archaeological Reconnaissance Project (MRARP) was conducted by Professor Audax Z. P Mabulla in 2001. Mabulla’s research project documented about nine rock art sites on the basis of colour type, subject matter and artistic style, Mabulla (2005) grouped the rock art of Mara Region into two traditions:

1. Red Geometric Tradition: is characterized by red geometric designs that include “uterus-like” designs, ”oval” designs, circles, concentric circles, circles with externally radiating lines, circles surrounded by ordered lines of dots, semi- circles, dots, divided circles, ladders, lines, sets of parallel lines and “H-like” designs. Other figures include a few wild animals, some weapons (bow and arrows), crude human (“stick”, “alligator-like” and “rectangular”) designs, distended human figures and snake-like designs. The Red Geometric Tradition is believed to extend into Zambia;
2. Red Animal Tradition: is characterized by thick-line paintings in red and yellowish red with animals, humans, spears, an arrow and vertical parallel lines occurred only at Chanyamsinja rock-shelter in Mara Region. All painted animals of this tradition are portrayed in a stylized form.

Despite the fact that there are numerous evolutionary schemes used by archaeologists in studying the rock art of Tanzania, it appears that previous researchers failed to develop a common analytical scheme. Most of the schemes used colour and subject matter as the basis for classification (Willcox, 1984). However, Chami (2006) believe that the general way to classify paintings and engravings is not just to rely on colours or types of engravings, but rather on subject matter like naturalistic and semi-naturalistic figures. Such classification also includes schematic, geometric or amorphous (SGA) signs. Closely related classification has also been used by Masao (1979), Willcox, (1984) as well as Coulson and Campbell (2001), whereas the rock art is categorized into four complex stylistic units, namely, stylized presentation, naturalistic presentation, white semi-actualistic silhouettes and abstract/geometric figures. But still there is no standard classification scheme to be used as a landmark for the evolutionary scheme for the rock art in Tanzania. This study employed Masao (1979) that also incorporate Odner (1971) analytical scheme, which I think it is the most applicable scheme for the study of rock art, I encountered during my fieldwork (see chapter 5).


4.1. Introduction

This chapter describes the various methods that were employed for data collection. It describes the research design, sampling techniques and sample size as well as data collection methods which included archaeological surveys, ethnographic inquiries and test excavations.

4.2. Research Design

A research design is defined by Kothari (2004) as an arrangement of conditions for data collection and analysis. In this study, both rock art and other archaeological materials data were collected, analyzed and described in details to establish the cultural history of the Iringa Region. The study was conducted for four weeks during which surveys, test excavations, interview and documentation of rock art sites were accomplished.

4.3. Sample, Sample Size and Sampling Procedures

Sampling procedure was based on the natural landscape whereby all accessible Precambrian rocks and outcrops were purposely sampled. Therefore, rock boulders, overhangs, rock-shelters and caves were examined in detail so as to identify, record and document the rock art sites. The paintings were described and classified according to their subject matters, stylistic characteristics, composition, action, aspect, and techniques of execution. During the fieldwork I did not encounter any engravings.

4.4. Data Collection Methods

Data collection methods are research techniques employed in producing data (Kothari, 2004). For this study, the data were collected through interviews, surveys and test excavations. Interviews were aimed at assessing community’s perception about rock art in their areas. Initially, my plan was to interview people aged to about 60 years and above, but eventually I realized that even some older people were not aware about the rock paintings. Instead I decided to interview even people aged below 45 years. In addition, I purposely sampled key informants who can provide useful information. These include Antiquities Officers, Cultural-tourism Officers, Village Executive Officers (VEO) and people living close to the rock art sites. For instance, I carried out a detailed conversation with teachers at Igeleke Primary School, which is located few meters southeast of Kihessakilolo rock-shelter. Through such conversation teachers were able to give some information about the current uses of the rock-shelter for ritual practice and other religious aspects.

Surveys were aimed at recording, documenting and analyzing their subject matter with regard to their thematic compositions as well as pigments used in the paintings. Test excavations were aimed to recover archaeological materials and figure out the relationship between archaeological materials and painted art. By the means of survey and interviews, I was able to document about 10 rock art sites, and three rock shelters with well preserved rock paintings that are discussed in detail in the following chapters.

4.4.1. Ethnographic Inquiries and Interviews

Ethnographic techniques is one of research methods commonly employed in studies aimed at linking the past and present. This approach allows the researcher to study the lifestyle and belief systems of modern societies in reconstruction of their past life ways. It is argued that indigenous people like the Bushmen, Sandawe, and Hadza can offer the best opportunity for better interpretations of the rock paintings in their places (Bendera, 2011; Coulson and Campbell, 2001; Mahudi, 2008). It was realized from this study that the knowledge of local people is fundamental for discovery of rock-shelters with paintings. Local people understand better their surrounding landscapes and constantly inventory available resources of the landscape for daily uses (Mabulla, 2005). In Iringa Region, local people understand well places with rock-shelters and their knowledge about rock art was relatively fair. Given the time, labour and financial constraints, I managed to locate two new rock-shelters with paintings based on the information provided by the local people.

Interviews were aimed to get clues about the presence of rock art sites and their current uses, to understand local people’s perception and interpretation of the rock art, and to shed more lights on the meaning, authorship, position and location of rock paintings. I intended to understand new aspects attached to the rock paintings and the day to day uses of the painted shelters.

4.4.2. Surveys Results

Surveys were employed so as to locate new sites with rock art. Both systematic and random surveys were employed depending on the geomorphologic features and vegetation cover. The alternation of survey strategies depended on geomorphologic factors such as terrain, vegetation, relief and the location of rock boulders. Field surveys were carried out for two weeks, from February 15th to 28th, 2012. The research crew comprised about 10 people directed by myself (Plate 2). The survey was hampered by some difficulties such as passing through thorny bushes, trekking long distances on foot and climbing hills and escarpments. Worse enough it was during the rainy season and we got stuck several times in mud and sometimes we failed to climb on rocks as it was very slippery. During survey, shelters with paintings were recorded, mapped out, evaluated and described in details (Plate 3).

illustration not visible in this excerpt

Plate 2: Survey Team at Tavimienda Rock-shelter

4.5. Recording, Documentation and Rock Art Sites Mapping

In accomplishing the surveying task, the following were done: recording of site names, geographical location, photographing, studying the nature of the rock and examining the current use of the sites. I also tried to count the number of painted figures as well as types of the paintings (naturalistic or semi-naturalistic) and studying their subject matter (Plate 3). The geographical location of a rock art site was documented using a hand held Global Positioning System (GPS) device. A Sony digital camera was used in taking photographs.

There are different ways by which rock paintings are recorded and photographed some simple and others fairly sophisticated and their adoption are determined by resources and time factor. Even though, there is no standard way of recording rock art, many scholars agree tracing and photographs as best recording means (Plate 3). All information related to the rock-shelter, qualitative, quantitative descriptions of rock paintings, artifacts, features and other field observations were recorded in documentation forms and field notebooks. Qualitative descriptions of rock art included explanation of different variables such as, stylistic characteristics, composition, execution technique, colour, subject matter and the state of preservation. Quantitative description included physical attributes such as length, width and height of the rock-shelter, number of painted figures, size of the painted area, slope of the painted surface, and exposure to sunlight. Furthermore, paintings were evaluated and analyzed in detail to determine the state of preservation and map out the potential threats affecting the rock art. This approach was part of a conservation and management strategy aimed at saving these priceless cultural heritage assets. Deacon (2006:35) argued that “documentation is the process of describing in a written, permanent form, all or some of the sites attributes. It includes the gathering and integration of all relevant written and graphic information about the site. The completion of the site recording form, with location, ownership, description of site, photos and references to relevant written records, is the most common and effective approach to the site documentation. The documentation of a site may be part of a survey report, heritage study, or research report”.

I also mapped out a site plans and cross-sections at Kihessakilolo and Ikula rock-shelters so as to give an idea of the size of occupation area. This is very important in recording and understanding changes in the site uses over time and space. The site plans helped to select suitable area for test excavations and were intended to guide future research directions i.e., avoiding irregularities when trying to re-excavate the site (Figure 5).

illustration not visible in this excerpt

Plate 3: Tracing the rock paintings at Kihessakilolo Rock-shelter

4.6. Excavation Procedures

In the course of surveying and assessment of rock art sites scatters of artefacts on surface were observed. Kihessakilolo rock-shelter site contains a large scatter of artefacts on surface and its northern part of shelter is not disturbed was deliberately selected for an archaeological excavation. The southern corner of the shelter has been disturbed by treasure hunters and other human activities. Quadrants A6, B5 and B6 were selected for excavation because (Figure 5). The first trench was established at B6 quadrant that later was further extended by a 1 m2 in northern and western direction (Figure 5). Excavation was done following arbitrary levels of 10 cm apart and it went down to the sterile layer about 100 m below the surface. A wire mesh of 5 mm was used to sieve small archaeological materials uncovered from the trench.


5.1. Introduction

This chapter presents the results of fieldwork. These include results from archaeological surveys, ethnographic inquiries and test excavations.

5.2. Surveys Results

Numerous kopjes (koppies) or castle kopjes are found all over the Iringa Region are the most common feature in areas with rock paintings. Archaeological surveys recovered a total of two new rock shelters with paintings. I also revisited a total of six rock-shelters that were previously reported in 2006 (Figure 3) by Katie Biittner and her colleagues but have not been studied in detail (Biittner, et. al., 2007). This is due to the fact that the study of rock art was not among of their primary objectives. Their reports inspired a deliberate and systematic study in this area through recording, mapping and documentation of the rock art. The new rock art sites include Lutona and Tavimienda rock-shelters while the previous reported ones included Kihessakilolo and Mlambalasi Hill.

5.2.1. Mlambalasi Hill (7° 43' 7" S,35° 42' 6" E)

Mlambalasi hill is located 50 km west of Iringa town. The area consists of five rock- shelters most with fainted paintings and is endowed with the archaeological materials. Professor Pamela Willoughby and her team excavated one of these rock-shelters in 2006 and 2010 respectively (Plate 4). This place is best known as the burial place of Chief Mkwawa (1855-1898), the leader of the Hehe people who strongly resisted against German colonization. Rather than surrendering, Chief Mkwawa killed his servant before committing suicide (Biittner, 2011). His body was buried at Mlambalasi and a monument erected to commemorate the 100th anniversary is located adjacent to the rock- shelter. His head was cut off, and sent to the Bremen Anthropological Museum where it remained until 1954 when it was finally returned to his family. His skull is currently displayed at the memorial museum at Kalenga, along with other personal belongings and items representing the cultural and economic activities of Hehe people (Biittner, 2011). As noted before, Mlambalasi is made up by a large number of isolated boulders in a big granite outcrop. Rock-shelters with paintings are located mainly in the northeastern direction and they are close to each other. Due to the presence of many rock-shelters in the area, I purposely divided them into four sites based on their archeological significance namely Mlambalasi 1 to 5. Mlambalasi 1

Mlambalasi 1 is the largest rock-shelter and it has two chambers (Plate 4). The first chamber faces north-east and the second chamber to the western direction. Archaeological remains including lithic artifacts, land snail shells, bones, iron slag, tuyeres and pottery have been reported to scatter inside and outside of the rock-shelter (Willoughby, 2012). Other materials found inside the rock shelter include hearth, grinding stone and a pestle rubber (Plate 5). Mlambalasi 1, was initially excavated in 2006 and later in 2010, whereby archaeological remains ranging from the LSA to historic period were discovered. The LSA deposit has been radiocarbon dated between 11,000 and 15,000 BP (Biittner, 2011; Bushozi 2012).

illustration not visible in this excerpt

Plate 4: Mlambalasi 1 Rock-shelter

In terms of paintings, there were two geometric designs painted in red and few late white dots. Most of the paintings are depicted in red pigments but they are highly faded (Plate 6). It seems that in past both chambers were painted. The painted area in the first chamber is about 8 m long, 5 m wide and 7 m high. In the second chamber the painted area measures to about 2.5 m long, 3.5 m wide and 1 m high. Generally, it is difficult to identify the nature and form of painted figures because of rock weathering and paintings are not wealthy enough for cultural tourism. However, this place is potential for tourism due to its archaeological, historic and cultural significance associated with Chief Mkwawa.

illustration not visible in this excerpt

Plate 5: A Grinding Stone and a Pestle Rubber inside Mlambalasi 1 Rock Shelter Mlambalasi 2

Mlambalasi 2 occur about 20 m south-west of Mlambalasi 1. In this shelter there is a predominance of fainted red paintings. The identifiable figures include red-lines and strips of paintings that it very hard to interpret and a large symmetrical figure painted in a light red colour measuring to about 25cm long (Plate 6). Faint paintings are spread in western and eastern walls. The floors of contain dense scatters of LSA artifacts and pottery.

illustration not visible in this excerpt

Plate 6: Mlambalasi 2 Red-geometric Rock Paintings Mlambalasi 3

This site occurs in the south-eastern side of Mlambalasi 2 about 50m from Mkwawa grave. It is a granite boulder with three shelters. The first shelter has a scatter of pottery on floor, but it has no paintings. The second shelter has many fainted red paintings. The third shelter has red paintings on the roof, but they are very faint. Few grinding stone, animal bones and potteries were found inside and outside the rock shelter. Even though most of the paintings are hard to indentify because of rock weathering, it is likely that geometric figures predominates. Mlambalasi 4

This is a three sided rock-shelter facing northwards, just about 100 m away from Mlambalasi 3. Only two walls have red paintings. Most of the paintings are in red colour, but they are very faint and not easy to determine. The second wall faces northward and it has a gentle slope, also the paintings are faded and unidentifiable. This rock shelter is rich in terms of archaeological materials including stone artifacts and pottery. Like other shelters of this place, rock paintings at Mlamablasi 4 are not in a good state of preservation.

5.2.2. Lutona Rock-shelter (7° 46' 7" S and 35 °26' 2" E)

Lutona rock-shelter is a large circular rock-shelter found in Kitwiru area about 12 km west of Magubike Village. It is surrounded by numerous kopjes, forests, small hills and granite boulders of Precambrian era. The rock-shelter faces the Kitwiru ephemeral river- valley which perhaps served as a permanent source of water for prehistoric people. The shelter has a lot of archaeological materials scattered on the floor such as potsherds, slag, bones and lithic artifacts. Surface area of the shelter measures about 116 m2. The site is highly vandalized by iron-smelters and treasure hunters. There is a big hole in shelter probably dug by treasure hunters. The Rock-shelter contains few Late-White geometric paintings. At least 5 figures executed on the roof were documented. Majority of the paintings are faded due to the anthropogenic actions (Plate 7). The site is not so much attractive for cultural tourism and public display because of vandalism.

illustration not visible in this excerpt

Plate 7: Late White Paintings at Lutona Rock-shelter

5.2.3 Tavimienda Rock-shelter (7° 36' 13" S and 36° 30' 26" E)

This site is located in Kilolo District, about 198.4 km from Iringa town. It is found on the peak of Udzungwa Mountains, about 1280 m above sea level, and almost 9.4 km south-east of Great Ruaha River (Plate 8). Tavimienda rock-shelter is the most extensively painted rock-shelter accounting to about 100 figures. Over 50% of the painted figures are in excellent state of preservation while the rest are faded and unidentifiable. Given fact that most of the paintings are superimposed, it was difficult to count all the paintings (Plate 9). Paintings are depicted in white pigment and are executed on the roof. The shelter measures about 17 m in length and 47 m wide. The total surface area is about 108.1 m2 (Figure 5). Due to the mountainous nature of the site, the rock paintings are in a good state of preservation (Plate 8). Because of time, financial and weather constrains, I was unable to survey the entire area as a result only one rock shelter with paintings was documented in Kilolo District (Figure 8).

The floor of the shelter is littered with fragments of bones and a few potty sheds. The painting subject matter includes animals, humans, letter like signs and amorphous or abstract figures (Plate 9). Other painted figures include circles and concentric circles, and externally radiating lines. Depicted animals include reptilians such as alligator, lizard, and crocodile. Humans are stylized and painted in different forms, some of them kneeling down with stretched hands and legs. Paintings are in dusky white. The site is very suitable for public display and cultural tourism because a good number of paintings are well preserved. Nonetheless, most of the concentric circles and schematic humans are overlaid by animals, white dots and geometric signs suggesting that the site was seasonally inhabited or the paintings represent two different cultural traditions.

illustration not visible in this excerpt

Plate 8: The location of Tavimienda Rock-shelter

illustration not visible in this excerpt

Plate 9: Dusky White Paintings at Tavimienda Rock-shelter

illustration not visible in this excerpt

Figure 4: A Cross-Section of Tavimienda Rock-shelter

5.2.4. Kihessakilolo Rock-shelter (7° 43' 32" S and 35° 30' 22" E)

This rock-shelter is located at about 1556 m above sea level, northwest of Igeleke Primary School. The shelter is surrounded by granite-kopjes and euphorbia candelabrum (candelabra) trees. The floor of the shelter has a basement rock with abundance scatter of LSA artifacts, slag and potty sheds. The painted area is about 7.31 m2. The painted wall has more than 30 figures in which naturalistic human and animal figures predominate. Animals represented include three giraffes, one superimposed on the other, naturalistic eland, wildebeest and three elephants, both depicted in a naturalistic style (Plate 10). Figures like candelabra trees (Euphorbia candelabrum?), parallel lines, and dots in black and white colour also widely represented. White dots extended up to 4 m high. Majority of the paintings are in various shades of red and are in a good state of preservation. Generally, Kihessakilolo is the most accessible and suitable site for public display and cultural tourism.

illustration not visible in this excerpt

Plate 10: Rock Paintings at Kihessakilolo Rock-shelter

5.3. Overall Assessment of the rock art State of Preservation

Rock art studies in Tanzania much attention has been directed towards understanding of their meanings, styles and subject matters, with less emphasis on the integrity and survival of these priceless treasures (Mabulla, 2005). It was realized from this study that most of the rock art sites in the southern highlands are in danger and are highly affected by physical weathering, biological, and anthropogenic actions. The major physical agents threatening paintings in this place are rock weathering, exfoliation, and oxidization. The major biological threats include vegetation growth on rocks as well as birds and hyrax dropping. Threats related to the anthropogenic actions include graffiti (Plate 11) and smoke that cause the formation soot on paintings. The situation is worse at Mlambalasi rock-shelter where most of the paintings are faded due to the rock weathering and biological actions.

As noted in previous sections, most of the rock-shelters are crowded by archaeological and historical remains suggesting that they are still in use. The continual utilization of rock-shelter is a cruel phenomenon in human history and it has been documented in a number of places across the country (Mabula 2005; Chami, 2008). In this study, repeated utilization of rock-shelters was recorded at Lutona, Mlambalasi, and Tavimienda whereby anthropogenic actions was among the leading threats facing rock paintings. However, at Kihessakilolo and Tavimienda, moderate numbers of paintings were found in a fair state of preservation as majority of paintings were still visible, decipherable and identifiable. The major problem observed at Kihessakilolo rock-shelter is attempts made by the local government to fence the shelter. It is likely the local government mission was to keep local people outside the site catchment. This is contrary to UNESCO’s regulations as local people are the immediate custodians and stewards of the sites. In Tanzania, most of government institutions responsible for cultural heritage management believe that local people have nothing to play for the sustainable heritage management. They have forgotten their role of educating the local community about the scientific aesthetic and economic values of heritage resources. For instance, the lack of sensitization has made most of local people in Iringa Region to believe that rock paintings are signs left by German colonialists to locate places with buried treasures. Therefore, treasure hunters are looking for German’s possessions in rock shelters with paintings. Such like deems need serious education mitigations otherwise most of the paintings will disappear because of vandalism from treasure hunters. Once the former UN-Secretary Kofi Annan argued that:

“Africa’s rock art is the common heritage of all Africans and all people. It is the common heritage of humanity. As populations increase and vandalism and theft of Africa’s rock art are on the rise, this irreplaceable resource is highly threatened. It is time for Africa’s leaders to take a new and more active role. We must save this cultural heritage before it is too late” (Former UN Secretary-General Kofi Annan, 2005).

This argument from the former UN Secretary General is a call for common and joint efforts to fight against vandalism and looting of rock art. He advised the African leaders to carry out quick measures for preservation, management and conservation of these priceless heritage assets.

illustration not visible in this excerpt

Plate 11: Grafitti on Rock Paintings

5.4. Ethnographic Inquiries Results

In regard to this rock art study, ethnographic inquiries revealed that some rock art sites are still attached to traditional beliefs such as rituals and offerings. At Tavimienda rock- shelter, one informant Mzee Zubeir Nzigilwa (95 years old) proclaimed to use the rock art shelter for especially for rain-making rituals, blessings, forgiveness and appeasing their gods and ancestors. Another informant Mr. Mitrick Kihongosi (36 years) from Magubike Village confirmed that Lutona rock-shelter is currently used by iron smelters because of superstitious belief attached on it. Mr. Kihongosi claimed that sometimes local people uses some of the shelters for overnight sleep particularly, during the rainy seasons.

5.5. Excavations Results

Three excavation units were established at Kihessakilolo rock-shelter. These were A6, B5 and B6 (Figure 4). The sub-datum point was established to the NE corner of unit B6. Excavation units (B5, B6 and A6) were established in undisturbed areas. All excavated units were taken down to the bedrock about 80 cm below the surface.

5.5.1. Excavation Unit A6

This excavation unit was established 6 m east of the shelter’s wall and 2 m from the northern end of the shelter (Figure 5). This unit was excavated to 80 cm below surface. The recovered materials are presented in Table 3. These include 173 lithic artifacts, 24 bones, four red-ochre, five shells fragments, one charcoal and one petrified wood (Table 1).

illustration not visible in this excerpt

Figure 5: Kihessakilolo Rock-shelter Excavation Units Layout

Table 1: Inventory of Recovered Materials from Excavation Test Pit A6

illustration not visible in this excerpt

5.5.2. Excavation Unit B5

Excavation unit B5 yielded a total of 180 lithic artifacts, 32 bone fragments, 10 pieces of red ochre, and nine pieces of shell fragments (Table 2). Excavation ended at about 80 cm deep where we encountered a bed rock.

Table 2: Inventory of Recovered Materials from Excavation Test Pit B5

illustration not visible in this excerpt

5.5.3. Excavation Unit B6

Like other units, excavation procedure was based on arbitrary spit of 10 cm apart. It went down to 80 cm below surface, where we encountered the bedrock. It yielded many artifacts compared to other excavated units. It yielded a total of 299 lithic artifacts, 32 bone fragments, 10 pieces of red ochre, one piece of a charcoal, and one piece of petrified woods and nine pieces of shell fragments (Table 3).

Table 3: Inventory of Recovered Materials from Excavation Test Pit B6

illustration not visible in this excerpt

5.5.4. Stratigraphy

The wall profile was drawn from northern wall of Unit B6 because the stratigraphic layers were apparent. The stratigraphic were characterized by fine clay- reddish brown soils in the upper level, gravel and reddish grey in the middle level and light grey in the lower level (Figure 5).

illustration not visible in this excerpt

Figure: 6 Northern Wall Profile for Excavation Unit B6


6.1. Introduction

This chapter presents results of text excavations carried out at Kehessakilolo rock- shelter. About three units were test excavated. Collected materials belong to the LSA culture, characterized by microlithic stone tools, bone fragments, petrified woods, land snail shell fragments and ochre. It is likely that ochre performed a number of functions including rock painting.

6.2. Lithic Analysis

The analysis of lithic artifacts was carried out at Archaeology Unit laboratory, University of Dar Es Salaam. A total of 593 lithic artifacts were analyzed using Mehman’s (1989) and Mabulla’s (2005 unpublished) analytical schemes. I decided to use these schemes because they are widely used for lithic analysis in Tanzania. The initial analysis indicated that shaped tools account to about 70 pieces, cores 138, flakes 99 and angular fragments about 283 pieces. Since one of the goals of this study was to establish the culture history of the rock art sites, angular fragments were excluded from the statistical analysis. Therefore, only 320 lithic artifacts were analyzed and results are presented in detail (Table 4).

6.3. Lithic Raw Materials

Quartz is the most dominant lithic raw material accounting for about 84% of the total. Basalt accounts to about 8%, quartzite 6% and chert 2% of the total. These raw materials were constantly distributed in all three excavated units. Today, local vein-quartz and quartzite outcrops, are locally found in the Precambrian granite outcrops that characterize most of the landscape in Iringa Region (Bushozi, 2011). Chert is not locally found probably, they were obtained from early archaeological sites such as Isimila (Bittner, 2011; Bushozi, 2011). The abundance of quartz and few quartzite artifacts suggests a pattern of localized raw material procurement in Iringa Region during the LSA.

illustration not visible in this excerpt

Figure 7: Lithic Raw Materials Distribution

6.4. Typological Classification of Lithic Artifacts

6.4.1. Whole Flakes

All un-retouched pieces preserving some part of the talon were classified as whole flakes. One notable feature in this lithic category was the presence of faceted platform for the most of analyzed flakes suggesting for the application of platform core reduction technique. Whole flakes comprised about 99 (20%) of the analyzed artifacts. These include 71 flakes, 20 blades, 5 utilized fakes and 3 levallois flakes (Table 4). All flakes with a length that is two times their width were classified as blades (Mehlman, 1989). Blades have mean length of 49 mm long, 20 mm wide and 6.4 thick. About 5 flakes were found with irregular edge modifications suggesting that they performed various activities in their life history (Figure 8). Majority of flakes have restricted platforms suggesting that toolmakers were able to control force and direction of striking (Bushozi, 2011).

Table 4: Classification of Lithic Artifacts

illustration not visible in this excerpt

Figure 8: Lithic Artifacts Illustration: a-Whole flake; b and c-Trimmed/Utilized flake

6.4.2. Shaped Tools

Three types of shaped tools were identified in this study. These include: scrapers, points and backed pieces. Scrapers

Scrapers account for about 46 (65.7%) of the modified tools (Table 4). The scrapers physical condition indicates that majority of them 45 (97.8%) are fresh while the remaining 2.2% are moderately abraded. Scrapers are dominated by non-cortical surfaces 39 (84.4%) suggesting that tertiary flakes were the most preferred for making scrapers. It may also indicate that initial tool manufacturing was taking place away from the shelter. However, the presence of scrapers with cortex (12.6%) suggests that some initial core reduction took place at the site. Possibly the site was used for multifunctional purposes i.e., tool manufacturing and residential activities. Points

There are 13 points, forming 18.6% of the analyzed tools (Table 4). Three types of points were identified. These include three Levallois points (4.3%), 8 unifacial points (11.4%) and 2 bifacial points accounting about 2.9%. The mean length of points is 29.5 mm, mean breadth is 15.0 mm and mean thickness is about 6.7 mm. Majority of the analyzed points have thinned platforms suggesting that they were hafted into wooden handles to form a composite tools (Bushozi, 2011;). Backed Pieces

There were 11 backed pieces, forming 15.7 % of the analyzed tools. The mean length of backed tools is 22.5 mm long; breadth is 11.8 mm, and thickness of about 6.9 mm. One backed piece is made of basalt, three are made of chert and the rest are made of quartz. Backed pieces were further analyzed to sub-types. These include six (8.6%) curve- backed pieces, four (5.7%) straight backed pieces and one (1.4%) crescent. The majority of them indicate that were either backed on the proximal or distal ends to facilitate hafting (Plate 17). Cores

According to Mehlman (1989), a core is a distinctive artifact that results from the practice of lithic reduction. Thus, a core is a scared nucleus resulting from detachment of one or more flakes from a coble. The core is usually marked with flake scars. The study analyzed a total of 138 cores forming about 27.9% of all analyzed lithic artifacts. Cores were further divided into sub-categories that include: 30 bipolar cores (21.7%), 15 amorphous cores (10.9%), 10 part-peripheral cores (7.2%), 18 radial cores 13.0%), 10 disc cores (7.2%), 16 divers single platform core (11.6%), 15 levallois cores (10.9%), 4 opposed double platform cores (2.9%), 10 pyramidal single platform cores (7.2%), 3 adjacent double platform cores (2.1%), 2 single platform core scrapers (1.4%), 2 platform peripheral core scraper (1.4%) and 2 opposed double platform core scrapers (1.4%) (Table 4). This trend indicates that different varieties of knapping techniques were employed by toolmaker at Kihessakilolo rock-shelter.

6.5. Bones

The analysis of faunal remains was done by Ms. Agness Gidna (PhD student from the University of Complutense, Madrid, Spain). A total of 79 bone fragments were recovered from excavations, but only diagnostic specimens are discussed and presented in this section. These include two astragalus and two distal metapodials of bovid, one long shaft of large mammal, one rib of a large mammal and one tooth of a canid (Table 5). Other identifiable bones include two pieces of ribs, a femur and humerus of rodents. Rodents are animals that inhabit rocky hills and kopjes. All the identified bones show the evidence of stone cut-marks suggesting that they were processed using stone tools, but some of the fractures resulted from natural processes (Plate 12a and b). The presence of highly fragmented bone remains, plus those with cut-marks could be interpreted as a reflection of food residues.

illustration not visible in this excerpt

Plate 12: Identifiable Fauna Remains from Kihessakilolo Rock-shelter

6.6. Land-Snail shells

A total of 19 pieces of land-snail shell fragments were recovered during test excavations. The presence of land snail shells in archaeological deposits may imply that prehistoric people exploited molluscs as part of their dietary composition (Bushozi, 2011). This is supported by evidence from Mumba and Magubike rock-shelters, where giant tropical land snails (Burtoa nilotica and Achatina fulica) were found associated with MSA and LSA artifacts (Mehlman, 1989; Bushozi, 2003; 2012). Ethnographic studies indicate that land snails are widely exploited by modern hunters of northern Tanzania (Bushozi, 2003). Recently, more substantial evidence for systematic exploitation of shellfish and molluscs has been reported from the MSA context at Pinnacle Point near Mossel Bay in South Africa (Bushozi, 2011). In addition, two petrified woods were found in units B5 and B6 suggesting that they may have served some domestic activities. Wooden artifacts are not common in archeological deposits probably because of post-depositional challenges.

Table 5: Identified/Diagnostic Fauna Remains from Kihessakilolo Rock-shelter

illustration not visible in this excerpt

6.7. Red ochre

This study recovered a total of 16 pieces of red ochre, most of them being ochre pencils (Plate 12). Red ochre was retrieved from about 20-40 cm below the surface. The use of red ochre has been linked to the executions of rock paintings (Leakey, 1983). Intensifications of the use of red ochre during the LSA period have been accepted by the majority of rock art scientists to indicate were commonly used for paintings in rock walls (Musa, 2011). Human burials, rock art, red ochre, ostrich eggshell beads are among the symbolic revealing objects that date back to MSA and LSA periods (Leakey and Leakey 1936; 1983; Mabulla 1996). The ochre pencils from Kihessakilolo are characterized by pointed tips suggesting they were used for painting (Plate 13).

illustration not visible in this excerpt

Plate 13: Ochre Pencils with Polished and Striated Surfaces at Kihessakilolo rock-shelter


7.1. Introduction

This chapter compares and contrasts the rock art of southern and north-central Tanzania in terms of shape, stylistic motifs and subject matter. Attribute analysis includes the nature and size of depicted figures or signs, styles, colours, execution techniques, and body position.

7.2. The Rock Paintings

The painting subject matters were both representational and non-representational. The former is represented by naturalistic and semi-naturalistic animals, semi-naturalistic humans, and semi-realistic inanimate figures. The latter depicts schematic, geometric and amorphous (SGA) signs. In this study, only monochrome paintings in red, white, black and yellow were encountered. Neither bichrome nor polychrome styles were used for painting in Iringa. The monochrome dusky-red (stale and fresh) is the leading colour at Kihessakilolo, followed by the yellowish, white, and black colours (Plate 10). Six figures were documented at Lutona rock-shelter whereby four figures were depicted in white pigment and two figures were in yellowish pigment. At Tavimienda rock-shelter there were over 100 paintings. At least 80% of the paintings were in dusky-white pigment, the rest 20% were depicted in white and black colours. At Mlambalasi, majority of the paintings were faded and unidentifiable. Few recognizable paintings were depicted in red colour. Variation in colours observed at Lutona, Tavimienda and Mlambalasi rock-shelters may have resulted from number of factors including weathering processes, ingredients used to make painting materials, preference of the painters and age differences (Namono, 2010; Roselandrainy, 2011). For instance weathering processes can alter the colour from red to stale-red or white to yellow. It is likely that change of colour from whitish to yellowish revealed at Lutona rock-shelter might have resulted from weathering processes. Sometimes deterioration in colour depends on the nature of ingredients used to make the painting materials (Namono, 2010; Rasolandrainy, 2011). For instance paintings depicted by using materials made from rocks containing oxidized iron such as hematite and ochre are likely to change from reddish to yellowish brown or faded red (Rosondrainy, 2011). Apart from alternation of colour from weathering and deterioration processes it is obvious that the rock paintings of Iringa Region represent two different traditions. The red paintings, which are widely represented at Kihessakilolo belongs to the hunter-foragers communities, while the white paintings observed at Lutona and Tavimienda rock- shelters represents later cultures, most likely early farming communities. These cultural traditions are also represented in Kondoa and Singida, north-central Tanzania suggesting that the two regions were socially and culturally connected.

7.3. Order of Superimposition

At Kihessakilolo rock-shelter, three orders of superposition were identified (Plate 14). The first pigments are red and are overlain by black paintings. The black paintings are superimposed by white paintings. This superimposition pattern shows three cultural sequences in the same rock-shelter. Red paintings represent the oldest occupation sequence, probably the hunter-gathering community, early farmers are represented by black painting, and the white colour stand for late occupants. At Lutona and Tavimienda, two orders of superimposition were observed. The first pigment is white overlaid by yellow pigments. This may imply that Lutona and Tavimienda were inhabited by farming communities

illustration not visible in this excerpt

Plate 14: Order of Superimposition (a) Kihessakilolo and (b)Tavimienda Rock Shelters

7.4. Traditions

Two cultural traditions are apparent in the rock paintings of Tanzania. These are:

7.4.1. Hunter-Forager Rock Art Tradition

According to Mabulla (2005), Mabulla and Gidna (2013) the hunter-forager paintings predominate in central, north-central Tanzania. The clarity of painting edges suggests that the hunter-gatherer paintings were depicted with fine brushes. Mabulla (2005) subdivide this art tradition into two broad categories: the “Hunter-Forager Figurative Fine Line” (HFFFL) and the “Hunter-Forager Red Geometric” (HFRG). The HFFFL art mostly depicts wild animals in naturalistic and semi-naturalistic forms and humans in stylized form. Other subject matters include depictions of birds, vegetation, handprints, and anthropomorphic, their anthropic and shamanistic figures. Human depictions show some subsistence and socioeconomic activities such as hunting, singing, dancing and dynamic movements (Mabulla and Gidna 2013). It is common to find a large number of these subject matters painted at a single site. Mostly the HFFFL paintings are depicted in red, dark red, reddish brown, and reddish-orange. Very few were painted in yellow color and in two or more colors. In northern and north-central Tanzania these artistic expressions are widely found in Kondoa, Singida, Babati and Lake Eyasi Basin. In Iringa such like paintings are found at Kihessakilolo and Mlambalasi rock-shelters.

The HFRG tradition is made up by geometric designs such as sets of parallel lines, horizontal lines and ladder-like designs and concentric rings depicted in red colour (Mabulla, 2005). Other figures include scaffolding-like, millipede-like, snake-like line and indeterminate shapes as well as few wild animals and stylized humans. In most case, the HFRG paintings overlie the HFFFL paintings suggesting that HFFFL paintings are relatively older than HFRG paintings (Mabulla, 2005).

7.4.2. Bantu-Language Speaking Tradition

The Bantu or Iron working tradition appear crude and was mostly painted using fingers. It is dominated by geometric designs including dots, lines, circles, squares and smears (Mabulla and Gidna, 2013). Other represented figures include schematic depictions such as spread-eagle or distended human figures, stick-like human figures and stylized animals (Mabulla and Gidna 2013). However, there is a sub-tradition under this category, namely script/letter-like designs (Chami, 2008). These are artistic expression usually depicted in white, yellowish and black colours. In Iringa such like painting were documented at Lutona and Tavimienda rock-shelters (Plate 8).

7.5. Styles

These are collective characteristics of the art of a given culture that include motifs, techniques and forms (Masao, 1982). Styles can be used to categorize paintings into different groups. The main styles found in Iringa Region include naturalistic or semi- naturalistic style and schematic, geometric or amorphous styles (Masao, 1982). Naturalistic and semi-naturalistic styles include animals, humans, and objects depicted in various forms. Stylized, schematic or geometric styles includes six categories, namely quadrangular shapes, circular/hemispherical/oval shapes, triangular shapes, set of ordered dots, lines/strokes, amorphous designs and script-like/alphabet-like signs (Rasolandrainy, 2011; Chami 2008). The script-like or alphabet-like signs have been correlated to the development of alphabetic writings systems. The naturalistic and semi- naturalistic styles were documented at Kihessakilolo, whereas the stylized, schematic and geometric styles were mainly documented at Lutona and Tavimienda rock-shelters.

7.5.1. Subject Matters

The recorded rock painting sites in Iringa Region yielded six major subject matters which are easily identifiable. These categories of subject matter include zoomorphic (reptiles and insects), anthropomorphic figures, equipments or paraphernalia, geometric and script-like designs and indeterminate/non-representational paintings (abstract). Zoomorphic Figures

This category includes depictions of naturalistic and semi-naturalistic animal figures. These are large, medium and small sized animals of different species. Painted animals include elephants, giraffes, elands, wildebeests, antelope, zebra, and other unidentified animal figures that mainly appear at Kihessakilolo rock-shelter (Plate 15).

Depictions of giraffes are dominant over other animals which naturalistically were executed in such a way that it was easy to identify them (Plate 16). The dominance of giraffe species may suggest that they played a certain role in the concerned community. In modern context, giraffes are sometimes associated with either religious or ritual activities. For instance, the Nyaturu people of central Tanzania, giraffes are normally associated with beauty expression or calmness. Among the San of the Kalahari, Botswana giraffe plays an important role in the ritual and folklore practice (Bisele, 1993;

Bendera, 2011). Other widely painted animals include elephants, elands and wildebeests depicted in red colour. Such paintings appear only at Kihessakilolo, but they are well elaborated and easier to identify. However, ears and tusks are missing in all cases, while the tails are well elaborated. For instance, all executed elephants were naturalistically drawn in solid/ bold/dark red colours. Masao (1979) referred this category as hunter forager art and they are widely found in central Tanzania, especially at Chungai 3 (Leakey, 1983).

illustration not visible in this excerpt

Plate 15: Naturalistic and Semi-naturalistic Paintings at Kihessakilolo Rock-shelter Anthropomorphic Figures (Stylized and Schematic Human Figures)

Stylized human figures were documented at Tavimienda and Kihessakilolo rock- shelters. All gender, males and females are executed, but males predominate. However, this assertion requires further researches. All recorded human figures do not have primary sexual expression; instead they are gender expression and can be revealed based on the styles of body articulations. For example, women bodies were represented in reverse articulation of legs and exaggerated part of the spine, without breasts.

Sometimes female figures are characterized by the head-dresses expression (Plate 16). The male figures were characterized by slender, stretched, and upright bodies (Plate 15). Similar styles of body expression have been reported from Limpopo in South Africa, where male figures were identified by slender and upright body articulation contrary to females, who are represented by reverse articulation of legs, exaggerated lordosis of the spine and frequent breast attached to the thorax (Bendera, 2011).

illustration not visible in this excerpt

Plate 16: A Panel of Head-dress at Kihessakilolo Rock Shelter Schematic Human Figures in Crude White Frieze

Unlike the previously stylized red human figures with longer arms and legs than a trunk, schematic human figures have square/box like trunk in open lines. They are also characterized with stylized legs and arms as well as schematized tails (Figure 9c, e and g below). Others are monster or dragon-like designs (Figure 9).

illustration not visible in this excerpt

Figure 9: Re-redrawing of Bantu Rock Art at Tavimienda Rock-shelter Reptiles, Birds and Insects

This group of painting includes crocodiles, alligators, lizards and insect like figures. These artistic representations appear in light, dark red and white colours. There are represented in hunter-gatherers and Irion Age traditions. Reptiles depicted in red are associated with hunter-gatherers community and were documented at Kihessakilolo rock-shelter. Reptiles depicted in white colour are associated with early farming communities and they were widely documented at Tavimienda. Reptiles are exaggerated measuring to about 30 cm long. They are executed horizontally in the roof of the shelters (Plate 17).

illustration not visible in this excerpt

Plate 17: Lizard/Alligator and Insect Paintings: right is from Tavimienda and left Mlambalasi rock-shelters Geometric, Abstract Figure and Script-like Design

Geometric and script like design includes sets of parallel lines, ladder like figures, strokes arranged in a row, horizontal snake-like lines, zig-zag lines, scorpion like designs and other figures with unidentified shapes. In all cases, figures of this group appeared in association with other figures like humans, animals or both. They are painted with various colours especially, light and dark red. Such like paintings were mainly documented at Kihessakilolo and Mlambalasi rock-shelters.

Other category of geometric design figures is the Schematic Geometric Amorphous (SGA) figures. The SGA figures are represented by various squares, rectangles, circular, hemispherical, quadrangular and other amorphous designs. Such artistic designs are widely distributed at Tavimienda rock-shelter and consisted over 10 % of the total paintings (Plate 17). Some of the figures convinced me to support an idea raised by Chami in 2006 that were possibly sorts of ancient writing systems (Chami, 2006; Rosolandrainy, 2011). The SGA figures are widely distributed in sub-Saharan Africa, and have been reported in number of places including Muleba in northwest Tanzania, Nubia in Sudan, Madagascar and Egypt (Rasolondrainy, 2011),

illustration not visible in this excerpt

Plate 18: SGA signs at Tavimienda Rock-shelter Set of Quadrangular Shapes

Quadrangular shape categories included square-and-cross, square-and-X, square-and- bolded dot(s), spirals, coils, square-and-line(s), square-and-grids, rayed square-and-line, squares-and-circles, rectangle, rectangle-and-vertical/horizontal line(s), rectangle-and- grids, rectangle-and-grids-and-dots, and elongated rectangle-and-lines. Others are human-like skeleton designs (Plate 19).

illustration not visible in this excerpt

Plate 19: Quadrangular shapes at Tavimienda rock-shelter Circular, Hemispherical, Oval-Shapes and Sun-like Designs

This category represents different circular, hemispherical and elliptically frieze shapes including outlined circles, filled-in circles, circle-and-cross, concentric circles, circle- and-lines, circle-and-dots, rays circle, rays filled-in circles, encircled rayed filled-in circles, rayed concentric circles (Rasolondrainy, 2011). Others are encircled rayed concentric circles, rayed circle-and-X, open semi-circle-and-dots, semi-circle, oval, oval-and-dots, tailed ovals, concentric ovals, concentric D-like and O-like signs or sunflower like patterns (Plate 19).

illustration not visible in this excerpt

Plate 20: Sun/Sunflower-like and Oval/Circular Patterns at Tavimienda Rock-shelter Triangular Shapes

This category represented by triangle, double facing triangle, rayed triangle designs (Plate 20). There widely represented at Tavimienda and Lutona rock shelters.

illustration not visible in this excerpt

Plate 20: Triangular Dusky White Designs at Tavimienda Rock-shelter Cycle of Dots

Designs and patterns in this group consist of single white bolded dots and a range of different cluster of dots (Plate 21). Most of the dots encountered were in white pigments whereas others were even bigger than the size of the executed animals and humans.

illustration not visible in this excerpt

Plate 21: Thick White Dot superimposed on Earlier Dirty White Paintings Lines, Scaffoldings/Ladder like and Millipede/Centipede like Patterns

Patterns of this category include strokes, vertical line, parallel horizontal lines, parallel oblique lines, curved line, and meandering line. Some of them stand like-railway-line or part of the vertebral column of an animal (Figure 10).

illustration not visible in this excerpt

Figure 10: Ladder like Designs and SGA Signs at Tavimienda Rock-shelter Script-like Signs

Some of the SGA signs look like alphabets or numerical signs like alphabetic, syllabic or ideographic representations of ancient writing or numerical systems (Chami, 2008; Rasolandrainy, 2011). This category includes vertical lines, parallel vertical lines, parallel horizontal lines, cross, X-signs, X-barred on top, V-sign, Chinese-hat sign or upside down V, arrow sign, square open at the base, inverted U, up-side down E, H signs, B signs, line and dots sign, Y-like sign, numerical signs like 9, cobra-like signs, cow’s head-like signs, pottery/gourd vessel-like sign (Figure 11). Though some of the paintings such as circular and quadrangular shapes have been classified in other categories, they also uphold some element of this category.

illustration not visible in this excerpt

Figure 11: Alphabet-like signs from Iringa Region compared to some Ancient Script Alphabets from other parts of the world (Re-drawn after Evans 1897: 384, 386; modified from Rasolondrainy 2011 and Phonetic value of the letter is not included.

These script/letter-like signs, patterns or symbols were also compared with symbolic signs painted on ceremonial gourd of the Nyaturu people of central Tanzania (Plate 22). According to Mzee Petrus Mughwae (70 years) from Mwisi Village in Singida, all signs engraved or painted figures have symbolic meanings. For instance, cattle figures symbolize of wealth, parallel and curved lines symbolize peace, and bridal signs symbolize fertility. These symbolic signs are also represented in the rock paintings of Iringa (Figure 12). The correlation between the ethnographic objects and the rock paintings could be taken as the linkage between the past and present.

illustration not visible in this excerpt

Figure 12; and Plate 22: Symbolic Representation on Ceremonial Gourds among the Nyaturu of Singida, central Tanzania


8.1. Introduction

This chapter discusses the research results and compares them with other findings from central and north-central Tanzania to establish their relationship or variation in terms of subject matter and stylistic motif. It also underscored crucial information related to the state of preservation, sustainable heritage management and cultural tourism. Lastly, the chapter gives some concluding remarks and recommendations for future researches.

8.2. Iringa Rock Art Traditions

It would be presumptuous to think that the rock art sites described in this dissertation are the only one with rock art in Iringa Region. It is possible that the rock art sites are scattered in the entire Precambrian rocks outcrops that characterize most of the southern highlands. The southern highland rock outcrop extends from Morogoro to Ruvuma, southern Tanzania. Indeed, this area has received less archaeological attention compared to other places of central and northern Tanzania. Results from this study indicate that the rock paintings of Iringa Region share some stylistic motifs and subject matters with the rock art sites of central and north-central Tanzania suggesting that occupants of these regions were culturally connected.

Traditionally, the rock paintings of Iringa Region can be divided into two cultural sequences namely the hunter-foragers and early farming traditions. Hunter-foragers paintings were documented at Mlambalasi and Kihessakilolo and the Bantu or early farming community rock art sites were documented at Lutona and Tavimienda. These cultural traditions can be distinguished based on their colour, subject matter and stylistic motifs. For instance, the hunter-gathers paintings at Kihessakilolo and Mlambalasi rock- shelters are depicted in red, dark red and reddish brown colours. They are characterized by stylized humans some of them engaged in different activities such as hunting, singing or dancing. Other depicted figures include wild animals, scenes such as concentric rings and sets of parallel lines. Concentric rings and parallel lines have been referred by Mabulla (2005) as figurative fine line or red geometric traditions and are associated with late hunter-foragers community. Kihessakilolo rock-shelter provides a good example of superimposition whereby naturalistic animals and humans are overlaid by red geometric and figurative line traditions suggesting that the shelters were seasonally occupied by group of hunter-foragers. Detailed descriptions of the overlying hunter-gatherer traditions: the hunter-foragers bolded red or filled in red and red geometric traditions are described in the following sections.

8.2.1. The Hunter -Forager Bolded Red/Filled-in Red Tradition

This tradition is dominated by animal and human figures executed in dark-red color. Animals are naturalistically drawn while human figures appear stylized. Observed animal figures include large animals like elephants, giraffes and eland. Other animals are antelopes, zebras and wildebeests. Hunting scenes were clearly observed. A particular attention was paid to human body especially long hair styles and headdresses or plumes.

Some human figures were very long up to 1.45 meters long. However, sometimes human figures were executed with animal parts such as animal heads. Their posture is either bending at the waist, bending forward, standing, while others were lying down. As it was noted earlier most of the rock paintings at Mlambalasi rock-shelter are unidentifiable, but the red bolded figures were widely documented at Kihessakilolo and they are overlain by later traditions. Similar cultural tradition has been reported at Kolo, Cheke and Pahi sites in Kondoa, Tanzania (Leakey, 1983; Masao, 1979).

8.2.2. The Hunter Forager Red -Geometric Tradition

Observable figures under this tradition are geometric designs executed in various red colours. Dominant figures include lines, sets of parallel lines, horizontal lines, and millipede/centipede like as well scaffolding/ladder like designs. Other figures are strokes arranged in a row, snake-like line and other figures with undefined shapes both are in monochrome red. Paintings of this tradition are common at Mlambalasi rock-shelter even though they are faded, at Kihessakilolo they are relatively few. Somewhere else in Tanzania red-geometric paintings have been reported at Kisese 2 in Kondoa and Mara Region (Leakey, 1983; Mabulla 2005).

Generally, the hunter gatherers rock art represented at Kihessakilolo rock-shelter seemed to have the subject matter of animals and humans more than any other type. Giraffes dominate painted animal species suggesting that they had significant cultural expectations. This trend of praising giraffes has been documented in different sites of central and north-central Tanzania (Masao, 2007). The dominance of giraffes in both regions (central and southern) Tanzania indicate that were either inhabited with similar people or artists had closely related cultural traditions.

Taxonomically, most of the painted animals are bovids and they are the most taxa represented in the archaeological assemblage at Kihessakilolo, suggesting were probably most hunted animal species. For this reason, painted animals had some subsistence or socio-economic implications. In both central and north-central Tanzania bovid species such as elands, antelopes and wildebeest are the most painted animals suggesting were symbolically ranked higher and had higher socio-economic value. Leakey (1983:23) argued that, selected animals for paintings were either representing animals that supplied most meat or they had deeper symbolic value. It was also revealed that most of the hunter-foragers paintings were done by red colour, the aspect which is also common in central Tanzania. The preference and consistency in colour for painting suggest a trend of cultural continuity over time and space in Tanzania.

8.2.3. The “Bantu Language-Speaker” Art Tradition

The early farming or Bantu painting were documented at Tavimienda and Lutona rock- shelters. They are depicted in dark-white, white-darkish and yellowish pigments. They are crude probably painted using fingers. Paintings are dominated by human figures, reptiles, geometric designs including series of dots, lines and circles as well as script-like designs. Somewhere else such like paintings have been referred as Bantu-language speakers’ art or Late White Tradition and occur in many places in central and northcentral, Tanzania as well as the Lake Victoria Basin (Annati, 1984; Bendera; 2011; Kessy; 2005; Mabulla; 2005; Mahudi; 2008; Masao, 1982; 2005). However, there is unique style of symbols, signs and undecipherable script-like figures present at Tavimienda rock-shelter, which are uncommon in central Tanzania.

Alphabetic-like symbols or SGA designs have been associated with the evolution of ancient numeric and alphabetic designs (Chami, 2006, 2008). In Iringa such like paintings were documented at Tavimienda rock-shelter. Somewhere else in Tanzania such like paintings have been reported from Nyakateme rock-shelter in Muleba District, northwestern Tanzania (Kwekason and Chami, 2003). Chami (2008), argue that the SGA paintings in Muleba District share same cultural traditions with people from Lake Turkana. According to Chami (2008), archaeological remains from Nyakatema rock- shelter represent early farming communities dated to about 400 BC. Other places where the geometric rock art designs have been recorded include Kilwa, whereas ceramic decorations suggest for the early contacts and interactions between East African Coast and the rest of the world in particular, the Middle East, Mediterranean and India. Such interactions could be used to explain the diffusion and exchange of ideas, skills and technology between sub-Saharan Africa and the rest of the world (Chami Pers. comm 2012). It is possible that such interactions influenced early innovation of symbolic communication such as the SGA signs.

8.3. Present and Ancient Use of the Rock-shelters

The surface scatters in most of the rock shelters of Iringa including Mlambalasi, Tavimienda, Kihessakilolo and Lutona rock-shelters were found to date from the Stone Age to historic period. The most recent cultural materials include the grinding stone with a pestle rubber, hearths, potteries, and bones indicating these rock-shelters are still intermittently used by modern inhabitants for ritual and other religious practices and sometimes they are still used for overnight sleep during rainy seasons (Mabulla, 1996). For example the presence of the grading stone, pestle rubber and hearths at Mlambalasi rock-shelter indicate that sometime people use shelters to process some foodstuffs.

Archaeological remains indicate that Kihessakilolo rock-shelter was inhabited from the LSA to historical period. Even though there in no Iron Age artifacts were revealed from the test excavations cultural materials representing the Iron Age such as pottery, slags, beads and rock art were also documented at Lutona, Tavimienda and Mlambalasi rock- shelters suggesting that Iringa Region landscape supported the human life from the Stone Age period to present. For the most cases the LSA people depended much on the local environmental resources and some of the resources in particular, ochre were used for paintings. It was also revealed from this study that the current use of rock-shelters is among of the leading threats facing the rock art sites in Iringa Region. These include quarrying of rocks, fire setting, graffiti, and treasure hunting within site catchments.

8.4. Concluding Remarks

All rock-shelters of Iringa Region indicate evidence of recent use despite the fact that some of them are found in remote areas and others located in the hill-tops with rough terrain and difficult to reach such as Tavimienda. This implies that rock-shelters still are socially and culturally attached to human survival and are still in use for ritual purposes. Apart from rituals and religious practices, some of the site in particular, Kihessakilolo it can be developed for cultural tourism as it can be accessed easily. However, the government should change the conservation strategies from protectionism such as fencing, as it was revealed at Kihessakilolo to community oriented approach whereby the local people could be involved in cultural heritage management.

Even though indisputable similarities have been documented between the rock art of central, north-central and southern Tanzania, much survey and documentation is still needed in southern Tanzania. For instance, some of the rock paintings such as bird images (ostriches) documented by Masao (1976) and Mahudi (2008) in central and north central Tanzania are lacking throughout my presentation. This can be explained in terms of the limitation of my study as much of the hunter-gatherers discussions are based on the findings from a single site Kihessakilolo rock-shelter. It is suggested, more studies are encouraged before making a much wider and meaningful conclusions. Future studies should focus also on the meanings, dating and possible interpretation of rock art. It is also recommended to carry out more exploration on petroglyphs so as to know exactly the existence or non-existence of rock engravings in Iringa Region.

It is my hope that information generated from this study will be useful to the national policy makers, heritage managers, scholars and investors especially those interested in the cultural tourism industry.

The local communities should be sensitized and motivated to protect and promote heritage resources in their areas. In this way the management of the heritage sites should follow the collaborative approach whereby the local community, local and central government will work together to promote, protect, conserve and preserve these priceless treasures.

8.5. Recommendation for Future Studies

This is the first rock art study in Iringa Region which has systematically evaluated, documented and recorded the status of rock paintings. It is anticipated that, this piece of information will as well be very useful for management and conservation of the rock painting sites. Moreover, new knowledge radiating from the rock art sites will possibly attract more researchers and tourists who in turn will bring funds in order to promote, protect and preserve these sites.

This study calls for more surveys and documentation of more rock art sites in the southern highlands. Future archaeological initiatives also should try to enhance community outreach programs and other promotion strategies so as to raise community awareness for the sake of proper preservation of the rock art, and for sustainable heritage management and development. Future studies should try to find connection between the rock art of northern central, central, and southern Tanzania with that of Zambia and Malawi to examine their relationships or variations. It possible artists who painted the rock art in these regions were culturally and ecologically connected. I strongly believe that this study has added dots on our current understanding and the distribution of rock art sites in Tanzania.


Anati, E (1983), The state of research in rock art: the Rock Art of Tanzania and the East African Sequence. Bollettino del Centro Camuno di Studi Preistorici Anati, E (2005), Rock Art Sites on the UNESCO World Heritage List. ICOMOS- UNESCO docomentation center-Paris

Bagshawe, F. J (1923), Rock Paintings of the Kangeju Bushmen, Tanganyika Territory. Man 91(2):146-147

Bendera, A. (2011), Documentation of Rock Paintings Sites in Manyara Region, The Case of Babati and Hanang districts. Unpublished MA Dissertation, University of Dar Es Salaam

Bernard, H.R (1995), Research Methods in Anthropology: Qualitative and Quantitative Approaches. London: Sage Publications Inc.

Biittner, K. M., P. M. Bushozi and P. Willoughby. 2007. The Middle and Later Stone Age of Iringa, Southern Tanzania. Nyame Akuma 68: 62-73.

Biittner, K (2011), Characterization and Utilization of Stone Age lithic raw material from Iringa Region, Tanzania. PhD Dissertation, University of Alberta.

Bisele, M. (1993), Women like meat: the folklore and Foraging Ideology of the Kalahari. Johannesburg: Witwatersrand University Press.

Buckle, C. (2007), Landforms in Africa: An Introduction to Geomorphology. Longman Group

Bushozi, P. G. M (2003), Middle Stone Age Occurrences and Hominid Behavioural Patterns in West Lake Natron Region, Northern Tanzania (Unpublished MA dissertation)

Bushozi, P. G.M (2011), Lithic technology and hunting behaviour during Middle Stone Age in Tanzania. PhD Thesis, University of Alberta.

Bwasiri, E. J (2007), The Management of Indigenous Living Heritage in Archaeological World Heritage Sites: A case study of Mongoni wa Kolo Rock Painting Site, Central Tanzania, Unpublished M.A dissertation, University of Witwatersrand; Johannesburg.

Chami, F. A (2008), “The Great Lakes: a complexity of cultural wellsprings”, in Arnold, M. Art in Eastern Africa, Dar-es-Salaam, Mkuki na Nyota: pp.47-64

Chami, F.A (2006), The Unity of African Ancient History, 3000 BC to AD 500. Dar es Salaam: E and D Limited.

Chaplin, T (1974), The Prehistoric Rock Art of the Lake Victoria Region, Azania 9: pp.1-50

Cole, G. H. and M.R. Kleindienst (1974), Further reflections on the Isimila Acheulian. Quaternary Research 4: 346-355.

Coulson D. and A. Campbell (2001), African Rock Art, Paintings and Engravings on Stone. New York: Harry N. Abrams, Incorporated.

Evernden, J. F., and G. H. Curtis (1965),The potassium-argon dating of the late Cenozoic rocks in East Africa and Italy. Current Anthropology 6: 343-385.

Fosbrooke, H. A (1950), Tanganyika Rock Paintings: A Guide and Record. Tanganyika Notes and Records.

Fozzard, P. M. H (1959), Some Rock Paintings of Western Usandawe, Tanzania Notes and Records 65:57-62.

Hansen, C.L. and C.M. Keller (1971), Environment and Activity Patterning at Isimila Korongo, Iringa District, Tanzania: A Preliminary Report. American Anthropologist 73(5):1201-1211.

Harpum, J. R. (1970), Summary of the geology of Tanzania. Dodoma: Tanzania Mineral Resource Division.

Harris, J. F. (1981), Summary of the Geology of Tanganyika. Dar es Salaam: Tanzania Government Printers.

Howell, F, (1962), Isimila - an Acheulean occupation site in the Iringa Highlands, Southern Highlands Province, Tanganyika. In G. Mortelmans and J. Nequin, eds., Actes du IVe Congres Panafricain de Préhistoire et de l‟Etude du Quaternaire. Annales de Musée Royal de l‟Afrique Centrale, Tervuren 40: 43-80.

Kessy, E.T (2005), The Relationship between the Later and the Iron Culture of Central Tanzania. PhD Dissertation. Simon Fraser University

Kimaro, F (2008), Later Prehistoric Cultures of Isimila Site of Iringa Region. Unpublished MA Dessertation, University of Dar Es Salaam.

Kohl-Larsen, L. and M. Kohl-Larsen (1938), Felsmalereien in Innerafrika. Strecker and Schröder, Stuttgart 29: 21-2

Kothari C.R (ed) (2004), Research Methodology: Methods and Techniques. New Delhi: Wishwa Prakashwan.

Kwekason, A. and Chami, F( 2003),‘The Archaeology of Muleba, South west of Lake Nyanza: a preliminaryreport’ , in F. Chami, G. Pwiti and C. Redimilahy (eds), Climate Change, Tradeand Modes of Production in sub-Saharan Africa (Studies in the African Past, no.3) Dar es Salaam:Dar es Salaam University Press pp: 59- 85.

Kröner, A., A. S. Collins, E. Hegner, A. P Willner, S. Muhongo, and K. V. W. Kehelpannala (2001), The East African Orogen: New zircon and Nd ages and implication for Rodiania and Gondwana supercontinent formation and dispersal. Gondwana Research 4 (2): 179-181.

Leakey, L. B (1936), Stone Age Africa. Oxford: Oxford University Press

Leakey, M. D (1983), Africa’s Vanishing Art: the Rock Paintings of Tanzania. New York, Doubleday and Company, Inc.

Leakey, M.D and Leakey L.S.B (1950) Excavations at Njoro River Cave. Oxford Clarendon Press

Lim, I.L (1992), A Site- Oriented Approach to Rock Art: A Study from Usandawe, Central Tanzania. Unpublished PhD dissertation, Brown University Lyaya, E C (2008), Archaeological field research in Njombe, Tanzania. Nyame Akuma.70:21-29

Mabulla, A.Z.P (1996), Middle and Later Stone Age Land-Use and Lithic Technology in Eyas Basin, Tanzania. Unpublished PhD Thesis. University of Florida Mabulla, A.Z.P (2005), “The Rock Art of Mara Region, Tanzania”. Azania XL 19-42.

Mabulla, A.Z.P and Gidna, A (2013), The Rock Art of North-Central Tanzania (Forthcoming).

Mahudi, H (2008), The Use of Rock Art in Understanding of socio-Economic Activities and Cultural values; The Case of Matongo-Isanzu in Iramba Disrict, Tanzania. MA Dissertation. University of Dar Es Salaam.

Masao, F. T (1976), Some common aspects of rock paintings of Kondoa and Singida, Central

Masao, F.T (1979), The Later Stone Age and the Rock Paintings of Central Tanzania. PhD thesis, Simon Fraser University

Masao, F.T (1982), The Rock Art of Kondoa and Singida: A Comparative Description. (National Museum of Tanzania Occasional Paper no.5)

Masao, F.T (2007), The Rock Art of Singida and Lake Eyasi Basin in Tanzania, London: Duggan Foundation.

Mehlman, M.J (1989), Later Quaternary Archaeological Sequences in Northern Tanzania. Unpublished PhD Dissertation, University of Illinois, Urbana.

Minja, T.R.A (1991), A Visit Report on Kising’a Lugalo Forest Reserve, Catchment Forest. Unpublished paper

Mjema, E (2008), The Later Stone Age of Endadu Rock Shelter, Babati District, Northern Tanzania. Unpublished MA Dissertation, University of Dar es Salaam.

Msemwa, P (2002), Archaeology of the Ruaha Valley, Iringa District. Report on File with the National Museum of Tanzania.

Mturi, A. A (1998), Archaeology of Tanzania. Dar-es-salaam: The Open University of Tanzania.

Mturi, A. A (2001). The field research report of Lake Victoria Basin Rock Paintings, submitted to the Department of Antiquities.

Mturi, A. A and Bushozi, P.G.M (2002), The documentation of Lake Victoria rock art in Bukoba area: A preliminary report (unpublished). Department of Antiquities, Dar es Salaam.

Musa, S (2011), Archaeological Excavation of Rock art Shelters in Kasoma Village, Mara region. Unpublished MA Dissertation, University of Dar Es Salaam.

Namono, C (2010), Surrogative Surfaces: A Contextial Interpretative to the Rock Art of Uganda. PhD Thesis, University of Witwatersrand.

Odner, K (1971), An archaeological Survey of Iramba Tanzania. Azania 6: 151-98.

Rasolondrainy, T.V. R (2011), Archaeological Study of the Prehistoric Rock Paintings of Ampasimaiky Rock Shelter, In the Upper Onilahy, Isalo Region, Southwestern Madagascar. Unpublished MA Dissertation, University of Dar Es Salaam.

Regional Commissioner's Office, Iringa. 2007, Socio-economic Profile of the Region. Tanzania Government Printer.

Shorter, A. E. M (1967). Rock Paintings in Ukimbu. Tanzania Notes and Records 67:49-55.

Willcox, A (1984), The Rock Art of Africa. New York University Press.

Willoughby, P. R (2007), The evolution of modern human in Africa: A comprehensive guide. Lanham: Altamira Press.

Willoughby, P. R (2009), From the Middle to the Later Stone Age in East Africa. In Camps, M., and P. Chauhan, eds., Sourcebook of Paleolithic transitions. New York: Springer, pp. 301-314.

Willoughby, P.R (2012), The Middle and Later Stone Age in the Iringa Region of southern Tanzania. Quaternary International 270: 103-118.

113 of 113 pages


The Rock Art of Iringa Region, Southern Tanzania
A Descriptive and Comparative Study
Catalog Number
ISBN (Book)
File size
4449 KB
rock, iringa, region, southern, tanzania, descriptive, comparative, study
Quote paper
Makarius Peter Itambu (Author), 2013, The Rock Art of Iringa Region, Southern Tanzania, Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: The Rock Art of Iringa Region, Southern Tanzania

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free