The Nok Terracotta Sculptures of Pangwari

Tanja M. Männel & Peter Breunig

Abstract Résumé

Since their discovery in the mid-20th century, the terracottas Depuis leur découverte au milieu du XXe siècle, les sculptures of the Nok Culture in Central Nigeria, which represent the de la culture Nok au Nigeria central sont connues, au-delà des earliest large-scale sculptural tradition in Sub-Saharan Af- spécialistes, comme la tradition la plus ancienne des sculptures rica, have attracted attention well beyond specialist circles. de grande taille de l’Afrique subsaharienne. Toutefois leur Their cultural context, however, remained virtually unknown contexte culturel resta inconnu pour longtemps, dû à l’absence due to the lack of scientifically recorded, meaningful find de situations de fouille pertinentes démontrées scientifiquement. conditions. Here we will describe an archaeological feature Ici nous décrivons une découverte, trouvée à Pangwari, lieu uncovered at the almost completely excavated Nok site of de découverte Nok, située dans le sud de l’état de Kaduna, qui Pangwari, a settlement site located in the South of Kaduna fourni des informations suffisantes montrant que des sculptures State, which provided sufficient information to conclude that en terre cuite ont été détruites délibérément et puis déposées, ce the terracotta sculptures had been deliberately destroyed qui, par conséquent, met en évidence exemplaire l’aspect rituel and then deposited, emphasising the ritual aspect of early de l’ancien art plastique africain. Auparavant, des observations African figurative art. similaires ont été faites sur d’autres sites de recherches. Similar observations were made at various other sites Or les sculptures en terre cuite découvertes à Pangwari we had examined previously. But the terracottas found at nous permettent non seulement d’augmenter nos connais- Pangwari not only augmented our insights into the advanced sances du développement stylistique avancé des sculptures stylistic development of the Nok sculptures, they also exhibited Nok, mais elles révèlent aussi des scènes du quotidien, par scenes of daily life like a relief of a dugout boat with two exemple le relief d’une pirogue avec deux pagayeurs ou bien paddlers, or remarkable details like a marine shell on the des détails remarquables comme une coquille sur la tête d’une head of a human figure – details indicating trans-regional figure humaine – des détails qui témoignent d’un commerce trade and long-distance contacts. Other finds from Pangwari interrégional et de relations de longue distance. D’autres dé- deepen our knowledge of therianthropic creatures among the couvertes à Pangwari approfondissent nos connaissances sur terracottas of the Nok Culture. des créatures therianthropiques parmi les sculptures en terre cuite de la culture Nok.

Keywords: Nok Culture, Nigeria, terracotta sculpture

Tanja M. Männel 8 [email protected] / Peter Breunig 8 [email protected] * Institute for Archaeological Sciences, African Archaeology & Archaeobotany, Goethe University, Norbert-Wollheim-Platz 1, 60629 Frankfurt a. M., Germany

DOI 10.3213/2191-5784-10300 © Africa Magna Verlag, Frankfurt a. M. Published online 15 Dec 2016

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016, pp. 313–329 313

This article is available under a Creative Commons Attribution 4.0 International License (CC BY 4.0) Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig

Introduction Frankfurt Goethe University in the course of which a team of German archaeologists in collaboration with While a number of researchers performed extensive Nigerian colleagues surveyed and excavated numer- studies on the Nok terracottas, their source material ous sites dating to the Nok Culture (Breunig 2009; was based on sculptures kept in museums and private Breunig & Rupp 2010). collections that had not been subject to archaeological excavations or scientific documentation (de Grunne Having focused on aspects of chronology dur- 1999; Boullier 2001; Chesi & Merzeder 2006). In ing the first years of research (Franke & Breunig the 1940s the British archaeologist Bernard Fagg had 2014; 2016, this volume), priority later shifted to the begun investigating sites of the Nok Culture in Central structural analysis of Nok Culture sites. This involved Nigeria (Fagg 1945; 1990), his work was continued by large-scale excavations to comprehend the spatial his daughter Angela Fagg-Rackham (Fagg, A. 1972) organisation and nature of the Nok sites. Part of this and later by the Nigerian archaeologist Joseph Jemkur endeavour were excavations at the site of Pangwari, (Jemkur 1992). It was however not until 2005 that located in the northern section of the project’s key archaeological research was resumed and intensified study area (Fig. 1) which started in 2011. Being neither when a long-term research project was launched at the seriously affected by looting nor by farming activities,

Figure 1. Map indicating the position of Pangwari and other Nok sites recorded in the surroundings since 2005.

314 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Figure 2. Pangwari. Panorama view with plan of excavated trenches A-I and stone circle.

Pangwari proved ideal for studying the archaeological Context of Nok sculptures from Pangwari remains of a Nok community. Until 2014, ten units (feature 10 and 11) comprising 2,617 square metres had been excavated (Fig. 2). 12,978 individual measurement points were Like many Nok sites in the study area, Pangwari is recorded in a three-dimensional grid, of which nearly located on slightly sloping terrain, stretching up to ten percent have been classified as fragments of terra- a saddle between two shallow hilltops (Fig. 3). The cotta sculptures. Among the sites that have so far been saddle constitutes a favourable position, offering level examined by the Frankfurt project, percentages of this ground, distant views to the east and west, and a cooling range are common among inventories of Nok sites con- breeze which is absent on the lee side of the slope. The taining terracotta fragments. Equally common is their advantageous setting either intentionally or accidentally appearance in association with other find materials, coincides with the peculiarity of the archaeological dis- in particular potsherds or waste like broken grinding covery made in trench E which extended along most of stones, fragments of ground stone axes, and charcoal the saddle‘s expanse. In the west-central section of this or charred food remains from fireplaces, which occur trench a rocky ridge covered by a layer of sedimentary side by side without any specific context, apart from deposits was exposed, one of the numerous outcrops of their presence in pits or pit-like structures (Schmidt plutonic bedrock which spreads all over the hilly area 2014: 47–48). While this implies that the sculptures of the site (Fig. 4). mostly ended up in fragments and were treated as ordinary waste, the excavation of one of the trenches Evidence of human activity was found along the (trench E) at Pangwari revealed a different context. entire excavated part of the ridge. In the northern por- A circumstance yet encountered quite rarely (Rupp tion a congregation of charcoal was detected, forming 2010: 75–76; Rupp 2014), there, in feature 10 and 11, part of the backfill of a pit (Fig. 4). South and east of we found several larger fragments of Nok terracotta the ridge two pits containing fragments of terracotta sculptures which had apparently been intentionally sculptures were uncovered. Compared to the surround- deposited. Clearly classifiable as hoards these two ing soil, the southern pit appeared darker in colour due features reflect a ritual aspect that is not recognisable to a concentration of charcoal particles, it also proved in other features where the terracotta fragments seem to be of slightly softer texture (feature 11). In the upper to form part of the settlement waste. Focusing on the level of this feature a therianthropic sculpture, displaying crucial importance of contextual information for the attributes of human and bird was recovered (Fig. 16), understanding of the Nok sculptures, the Pangwari also found was an anthropomorphic head with large ears evidence will be presented here. (Fig. 17) as well as an anthropomorphic sculpture with a four-legged animal on its thigh (Fig. 18) at the bottom of the feature (positions indicated in Fig. 4, feature 11).

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 315

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig Pangwari. Aerial photo indicating the excavated trenches and the position of features mentioned in the text. Blue dots represent fragments of terracotta sculptures. The red stars red The sculptures. terracotta of fragments represent dots Blue text. the in mentioned features of position the and trenches excavated the indicating photo Aerial Pangwari. Figure 3. Figure indicate the position of a model dugout canoe in trench C and cubic statuette D.

316 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Figure 4. Pangwari, part of trench E. Posi- tion of feature 10 and 11 next to rock ridge (centre). Terracotta sculptures are outlined in white. The other objects without outline are stones. Finds in feature 10 are from level -90 cm below surface. Finds in feature 11 are projected from different levels.

Immediately east of the ridge the excavators ex- different to feature 11, many pieces of broken terracotta posed a plaster of hand-sized pieces of local granitic sculptures — hands, arms, fragments of faces, heads and rock (feature 10). Underneath this cover lay miscel- other body parts. The finds themselves turned out to be laneous fragments of terracotta sculptures (Fig. 5a), all just as unique as the context of the finds. in a good state of preservation, due to having been in a protected position under the stone plaster. Already at 37 radiocarbon measurements were performed on this stage of the excavation it became clear that such a samples taken from all over the site (Tab. 1) – eight on concentration of terracotta fragments carpeted by appar- charcoal samples, 29 on charred remains of annually ently carefully selected cobbles was not in line with the growing plants, mainly Pennisetum (details see Franke above-mentioned waste scenario, but pointed to a delib- 2016, this volume). This data and the analysis of date- erate deposition. This assumption was confirmed in the able potsherds allowed for a reconstruction of the site’s excavation process. After the removal of the stone plaster, chronology. Pangwari first was settled during the Early more parts of terracotta sculptures, lying among small Nok period (ca 1500 – 900 BCE). This period is well granite boulders appeared (Fig. 5b). Dark soil indicated represented in the dispersion calibration of the radio- that feature 10, like the neighbouring feature 11, was carbon dates (Fig. 6). While no terracotta sculptures indeed a pit. It contained large amounts of charcoal and were produced during this phase, Early Nok pottery

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 317

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig

- 2013) has Bronk Ramsey et al. et

List of all radiocarbon Reimer Table 1. Table dates from the site of Pang wari (2012/13) obtained until March 2014 by the Frankfurt Nok Project. The list includes the uncalibrated C14 age (bp), the calibrated age range (BCE/ CE), information on the sample dated, the laboratory number, as well as context information. the of number the to refers „AB“ archaeobotanical sample, from which the material was taken; „#“ indicates the find number, if measured individually with a OxCal program The station. total 4.2, IntCal 13 ( 2009; been used for calibration with a probability of 95.4% (2-sigma). Remarks test excavation, 60–80 cm (AB 523) 2011 feature 2B, 45 cm (AB 43) feature 3 (possible fire pit), 37 cm (AB 39) feature 2B, 77 cm (AB 32) feature 2B, 73 cm (AB 40) feature 6, 41 cm (AB 50) near terracotta part, 28 cm (AB 48) near pottery (early Nok), 29 cm (#2002) near pottery (early Nok), 14 cm (#2421) near pottery (early Nok), 36 cm (AB 56) feature 8, 50 cm (AB 55) feature 8, 50 cm (#3242) feature 9, 54 cm (AB 54) feature 9, 86 cm (AB 123) feature 9, 123 cm (AB 70) east of feature 10, 124 cm (#6620) feature 10, 53 cm (AB 68) ferature 10, 45 cm (AB 108) feature 10, 90 cm (#6103) feature 10, 96 cm (AB 116) 48 cm (AB 72) near feature 11, middle level, 72 cm (AB 73) feature 11, 74 cm (AB 112) feature 11, 153 cm (AB 126) feature 11, 182 cm (#6692) Feature 11, and 12, 171 cm (#6648) between feature 11 feature 12, 57 cm (#5017) feature 12, 55 cm (AB 78) feature 12, 89 cm (#5932) cm, (AB 107) feature 12, 110 charcoal and terracotta, 26 cm (AB 77) find concentration, 17 cm (AB 105) feature 13, 31 cm (#752) find concentration, 43 cm (AB 104) stone circle, feature 4, 86 cm (AB 44) stone circle, feature 4 120 cm (AB 129) stone circle, 25 cm (AB 37) Lab. No. Beta-313507 MAMS 19299 MAMS 17843 MAMS 17842 49508 KIA MAMS 17846 MAMS 17844 MAMS 17845 MAMS 17847 MAMS 17848 MAMS 17850 MAMS 19300 MAMS 17849 MAMS 19301 Beta-347870 MAMS 17853 Beta-347871 MAMS 19302 Beta-347873 MAMS 19304 MAMS 17852 Beta-347872 MAMS 19303 MAMS 19306 MAMS 17855 MAMS 17854 MAMS 17851 49509 KIA 49510 KIA MAMS 19305 MAMS 17856 MAMS 19307 MAMS 17857 MAMS 19308 MAMS 17859 MAMS 19309 MAMS 17858 2012 2014 2013 2013 2013 2013 2013 2013 2013 2013 2013 2014 2013 2014 2013 2013 2013 2014 2013 2014 2013 2013 2014 2014 2013 2013 2013 2013 2013 2014 2013 2014 2013 2014 2013 2014 2013 Sample dated 2011 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 Sample collected Dated material Pennisetum Pennisetum Seed fragment Pennisetum Pennisetum Seed fragment Seed fragment Charcoal Charcoal Pennisetum Pennisetum Canarium Pennisetum Pennisetum Pennisetum Charcoal Pennisetum Pennisetum Charcoal Pennisetum Pennisetum Pennisetum Pennisetum Pennisetum Charcoal Charcoal Canarium Pennisetum Charcoal Pennisetum Seed fragment Seed fragment Charcoal Pennisetum Pennisetum Pennisetum Pennisetum Calibrated age (BC/AD) BC 754–411 754–416 BC AD 1449–1616 801–771 BC 1257–1044 BC AD 1454–1624 757–431 BC BC 1260–1125 AD 1659–1955 AD 1019–1150 AD 93 BC–5 772–542 BC 770–540 BC 804–771 BC 826–771 BC 790–553 BC 826–763 BC 191–54 BC 774–434 BC 801–773 BC 812–791 BC BC 411–235 1024–919 BC 809–789 BC 776–547 BC 800–597 BC 1416–1298 BC 1381–1056 BC 1001–848 BC 800–772 BC AD 1042–1184 1496–1415 BC 762–540 BC 797–557 BC BC 1356–1211 804–776 BC after 1955 C14 age (bp) 2450±30 bp 2463±18 bp 383±14 bp 2574±15 bp 2940±30 bp 365±14 bp 2461±19 bp 2973±19 bp 199±14 bp 971±14 bp 2037±14 bp 2504±18 bp 2490±21 bp 2591±18 bp 2610±30 bp 2524±15 bp 2600±30 bp 2105±21 bp 2480±30 bp 2581±17 bp 2617±15 bp 2310±30 bp 2826±17 bp 2610±17 bp 2510±16 bp 2553±16 bp 3093±16 bp 2985±40 bp 2780±21 bp 2577±17 bp 904±15 bp 3172±18 bp 2481±16 bp 2545±22 bp 3016±16 bp 2590±18 bp after 1955 Trench Pangwari B Pangwari B Pangwari B Pangwari B Pangwari B Pangwari C Pangwari C Pangwari C Pangwari D Pangwari D Pangwari D Pangwari D Pangwari D Pangwari D Pangwari D Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari E Pangwari F Pangwari F Pangwari F Pangwari F Pangwari SK Pangwari SK Pangwari SK

318 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

was present in all excavation trenches which is clearly distinguishable from later ceramic ware by its elaborate impressed and incised patterns. The décor of Middle Nok pottery is reduced to a decorative band along the rim of the vessel (Franke 2014). In contrast to Early Nok material, remains dating to Middle Nok (ca 900 to 300 BCE) seem to be less prominently spread across the site, but rather concentrate in a few pit-like features. One possible explanation is that people of the Middle Nok period only occasionally returned to the site that their antecedents had abandoned (refer to central section of dispersion calibration in Figure 6, overrepresented due to intensive dating of respective features). Alterna- tively, the settlement could have continued from the late second millennium BCE into the early first millennium BCE, since the 14C-dates show data accumulations in the Early and Middle Nok periods, but no interruption or discontinuance.

Archaeological material that we can identify as Middle Nok (terracotta fragments and potsherds) mainly occurs in a few spots, for instance in pits like features 10 and 11 (Fig. 4), or in features that we interpreted as graves (see Breunig & Rupp 2016 this volume). Either these are the remnants of brief visits possibly only undertaken to deposit sculptures and to bury the deceased in the ground of their ancestors; or Middle Nok people did settle at Pangwari, deposited sculptures and held funerals in the place where they lived. At this point it is not possible to develop a more precise model since the majority of finds from Pangwari such as grinding equipment or un- decorated potsherds cannot yet be attributed to a specific phase. 14C-signals also point to revisits during the Late Nok period (ca 300 BCE – 0 CE) and Post-Nok times. However, the delivered evidence still remains unspecific, so that the 14C-datings do not unequivocally mirror this interpretation. For instance, the dates acquired from the eight samples taken from features 10 and 11 disperse all over the first millennium BCE Fig.( 6) which can hardly be regarded as reliable data to determine the point at which the deposits were made. However, looking at the strati- graphic position of the dated samples, non-conforming dates stem from the upper levels, while those from the basal levels consistently remain in the Middle Nok period (Schmidt 2014: 75, 79, 109). Thus, the lack of chrono- logical clarity seems to be due to a mixture of the upper deposits of the site caused by erosional processes.

Figure 5. Pangwari. Feature 10 (terracotta sculpture deposit) in trench E. a: situation at level -60 cm below surface with fragments of sculptures between and below of cobble-stone pavement (above). b: situation at level -90 cm below surface with large fragments of terracotta sculptures below pavement. Sculptural parts are outline white. All other objects are stones.

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 319

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig

Figure 6. Graph of all radiocarbon dates of Pangwari (above) and only those of features 10 and 11 in trench E (below). The heights of the curves reflect the probability distribution of the respective dates. For calculation and diagramme, the programme CalPal (IntCal13) has been used (Weninger B., Jöris O. & Danzeglocke U. 2014, CalPal (Cologne Radiocarbon Calibration & Paleoclimate Research Package), www.calpal.de, accessed February 3, 2014).

General remarks on the attributes of Nok sculptures upside down, as standing figures with a cylindrical body, as reliefs on the bodies of specific, possibly ceremonial The terracotta finds from Pangwari are in line with vessels, or as figurines in miniature size — either solitary, the general appearance of Nok sculptures as described Janus-faced, as pairs embracing each other or standing by Bernard Fagg (Fagg 1990), style and manufactur- back to back. Where it can be determined, the sculptures ing technique leave no doubt about their Nok origin. seem to represent men more often than women, rarely Constants in the manufacture of the sculptures concern children, sometimes animals or therianthropes. the use of clay mixed with a coarse temper of quartz, feldspar, mica (the components of the local granite) and With a rather high degree of probability, the sex fireclay, as well as the torso being built using coiling of the figures can be assigned when respective char- technique (Beck 2014). At the end of the sculpting acteristics or gender attributes are visible. Thus, some process, a thin clay slip was applied to the sculpture’s figures display clearly female protruding breasts or surface which usually is only partially preserved but flat male chests. Genitals are virtually never depicted. present on most of the Pangwari figures. Without this However, a large number of the statues exhibit no such coating, the coarse temper is exposed and gives the sur- clearly male or female features. As far as garments are face a rough appearance. After drying, the figures were concerned, the sculptures are frequently shown wearing most likely fired at approximately 600 to 800°C in an a loincloth. The loincloth of male figures is styled as a open fire Fagg( 1990: 21; Beck 2012: 256–257), a firing flat, downwardly tapered band falling from the waist, process resembling practises still used in Nigeria today. where it is attached with straps.

Regarding the stylistic features of the Nok sculptures, The women’s loincloth, in contrast, consists of a the disproportionately large eyes are most significant broad strip of cloth tucked under the hip-belt, sticking and the typical “Nok eye” is identifiable even on small out above and below in different lengths Männel( & terracotta fragments (e.g., Fig. 12). The triangular or Breunig 2014: 190). We also observed that female crescent-shaped rims of the eyes are carved deeply into the figures wear two chains or bands, each running from clay, creating an effect of bulging eyeballs Fagg( 1948: the necklace to the right and left hip and then back to 587). The pupils of the larger sculptures are pierced, in the necklace. When bands or chains are shown in male smaller figures they are just dented Männel( & Breunig figures, there is only one band that runs diagonally 2014: 189). Other attributes are subject to remarkable across the torso and is attached laterally to the hip belt. variation. The size of the sculptures ranges from a few centimetres to over a metre in height. They appear either Also depicted are a variety of clothing with plaited as three-dimensionally shaped figures in various posi- cords, knots, pleats in textiles, also lavish hairstyles, tions, often placed on a base that consists of a pot turned headgears, beards, scarifications, a broad range of adorn-

320 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Figure 7. Pangwari, trench C. Relief of a dugout canoe with two persons and load. ments, as well as tools from the daily life of farmers or that it takes less effort to steer possibly ritual objects, for instance axes. Some sculptures a boat than to keep it moving also display symptoms of sickness like eye disease, ulcers forward. An alternative reading and physical abnormalities (Männel 2010: 60). Similar could be that of the rear figure observations have been described for other Nok sculptures as an experienced navigator, since the early discoveries of Bernard Fagg (Fagg 1990: maybe a knowledgeable person 31; Nozais et al. 2000), however, the terracotta finds dur- of older age who could safely ing recent excavations, especially those made in Pangwari, guide the boat through rapids Figure 8. Pangwari, have greatly contributed to form a more comprehensive and over long distances. trench D. Anthropo- picture of the vast range of images in Nok figurative art. morphic cubic statuette. Between the two figures four piles are depicted, repre- The terracotta sculptures of Pangwari senting cargo that cannot be clearly identified. Neverthe- less, the watercraft can be regarded as an indicator that Next to the sculptures found in features 10 and 11 of there was potential for transporting large amounts of goods trench E, some outstanding finds uncovered in other and that in all likelihood an extensive trade network existed trenches must also be referred to. They do not derive in the Nok Culture area. Whether these trade contacts were from specific contexts like the deliberately deposited long-distance, reaching as far as to the Atlantic Coast, as objects from feature 10 and 11, but were found in cul- might seem possible regarding a sea shell shown on the tural layers coinciding with other materials that have to head of one terracotta sculpture (described below), remains be addressed as discarded objects, and are thus classified an open question. Besides the 8000 year old dugout canoe as belonging to the “waste-scenario”. from Dufuna in northern Nigeria (Breunig 1996), the small relief fragment from Pangwari dating to the first millennium BCE is the second oldest known evidence Model of a dugout canoe of a watercraft in Sub-Saharan Africa, thus constituting a matter of particular significance for the history of nautical Of particular interest is the terracotta relief of a boat with endeavour in Africa. two paddlers (Fig. 7), reassembled from three relief fragments found in trench C (position see Fig. 3). Navigable watercourses like the Gurara River, a tributary of the Cubic statuette Niger, are located close to Pangwari (Fig. 1). The shape of the model, tapering to a point at both ends, suggests Among the 185 fragments of terracotta sculptures from a dugout canoe. There are two anthropomorphic figures, trench D is a somewhat unusual three-dimensional stat- one in the front and one in the back of the boat, each hold- uette in cubic form (position see Fig. 3). Compared to ing a paddle. Both wear round caps in a style known from the other figures found in Pangwari, one might gain the other sculptures; more details from the heads or bodies impression that this figure was not crafted by a very able are hard to recognise due to the eroded state the relief is hand. This is not least due to the eyes not being in typical in. The artist seems to have placed emphasis on the fact Nok form, lacking the dented pupils but instead showing that the figure in the back of the boat appears to be more several small arches above the eyes that may portray dainty than the paddler in the front, possibly to indicate wrinkles (Fig. 8). An equally possible interpretation is

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 321

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig

remaining terracotta sculptures were recovered from feature 10. Each object has its own peculi- arities but regarding the circumstance that they were found in close association, having been purposefully placed into the ground, they become vital to our aim to understand the ritual aspect of the sculptures.

Large torso

One outstanding piece among the finds from Pangwari is a torso which constitutes the largest three-dimensional Nok sculpture excavated by the Frankfurt team so far (Fig. 9). It was recovered in two fragments, neck and shoulder as well as the lower part of the body, lying next to each other in feature 10 (body in position No. 6 in Fig. 5a, neck and shoulder in position No. 14 in Fig. 5b). No additional parts could be refitted, thus the deposition undoubtedly took place without the head and base. After restoration of the two fragments which had several fractures, and their reassembly, the sculpture measured 82 centimetres in height. Considering the missing parts, the extrapolated total height must have been around 130 centimetres. The body is sculpted in the form of a tubular container, without legs or feet. Of the arms only stumps remain Figure 9. Pangwari, trench E, feature 10. Large male torso. but those indicate that the left arm was kept close to the body while the right arm was stretched that the attributes were intended to illustrate an illness out. The ornamentation is typical for Nok sculptures and or physical abnormality. Depictions of river blindness displays multiple rows of necklaces which are placed on for instance are common among Nok sculptures and are top of each other, covering the whole neck and shoulder known from the site of Ungwar Kura (Männel 2010: area. An equally multi-row belt or chain running diago- 62; Männel & Breunig 2014: 198–199). nally across the torso is attached laterally to the hip belt. The fragmented apron or penis sheath is also attached to the front part of the girdle. This defines the figure as Sculptures from trench E male. The navel is shown as a little bump and the spinal column as a line, another typical but not necessarily con- The sculptures discussed below stem from feature 10 stant feature of Nok figures. On the back of the sculpture and 11 in trench E. The man-bird-hybrid creature, the is a hole measuring two centimetres in diameter. Other head with large ears and the kneeling figure with a large figures also have such a “ventilation hole”, it was quadruped on its upper leg were found in feature 11, the possibly added to prevent the figure to burst during firing.

322 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Figure 10. Pangwari, trench E, feature 10. Torso.

Torso

Another three-dimensional torso (position No. 11 in Fig. 5b) shows a good preservation of the clay slip, thus giving an impression of how the sculptures originally looked like. Apart from the missing head, arms and legs it is almost complete (Fig. 10). This sitting figure is a prime example of the diversity in ornamentation and the creativity of the artists displayed on Nok figures. For instance, there is a necklace composed of several rows of chains or ribbons joint together by four bold rings. At- tached to the necklace with another ring is an elongated pendant which decorates the athletically shaped male Figure 11. Pangwari, trench E, feature 10. Single hand with bracelet. chest down to the abdomen. In the back of the neck the necklace is tied with a knot, the girdle on the right side of the body is a bow. Bow tying has rarely been noticed Position 4). This bracelet shows a decor pattern along its so far. Until now, it has only been observed on a figure conically shaped edge and has ball-shaped ends. It ap- from the nearby excavation site of Ifana 1, radiocarbon pears to be a cuff bracelet which would have had some dated between ca 900 and 800 BCE. It is possible that flexibility when putting it on and taking it offFig. ( 11). we are looking at a regional peculiarity. This is why this attribute deserves more attention in the future. A flat band tied around the fingers, and the hand holding a stick-like object is reminiscent of a bow. Fol- On the left arm of the torso (Fig. 10) there is a lowing this observation, the bracelet could alternatively broad conically shaped and partly decorated upper arm be interpreted as the wrist protection of an archer. The bangle with an incised pattern. The torso has a rope or pattern on the arm ring could then represent the seams a ribbon running across the left shoulder. Attached to on this protective cushion which would have originally it is another decorated element. been made of leather (Wiethase 2007: 42–43, 45).

Single arm with bracelet A man with shell decoration

However, the find of one single arm and hand from the A three-dimensional figure (position No. 10 in Fig. same feature is not in line with past discoveries — the 5b) with a complete head, front torso and folded arms, arm is adorned with a broad bracelet whose form sug- characterised as male by its chin beard, is unique in gests that it was made of metal, most likely iron (Fig. 5a, the way that its head is decorated with a large sea

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 323

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig

Figure 12. Pangwari, trench E, feature 10. Man with sea shell on his head. shell (Fig. 12). As with most other sculptures from could not be matched with any other terracotta finds Pangwari, the well-preserved slip allows for the iden- from the feature, hence must have been deposited in tification of a lot of detail, not only the shell, but also their fragmented state. This once more sustains the the decorated eyebrows and eyelids or the delicate lines assumption of deliberate destruction and selective representing the hair tightly tied to a bun high in the deposition of merely parts of the sculptures instead of back of the head. It is likely that the shell belongs to the the complete figures. The destruction appears as strong family of marine scallops called „Pecten“ or „Senilia contrast to the outstanding design of one of the heads senilis“ (pers. comm. R. Janssen, August 2013). Since which can be addressed as male because of the chin fresh-water mussels can be excluded, we reckon that beard (Fig. 13) (position No. 13 in Fig. 5b, lowest level the original, after which the sculpture’s shell ornament in feature 10). The head displays an elaborate hairstyle was shaped, came from the West African coast which and an unusually modelled mouth and nose with details is about 700 kilometres away. The distance that had to such as a vertical furrow that runs down from the nose be bridged to transport the shell entails the rarity and through the centre of the bulging upper lip. The skill of high regard that was placed on the item, emphasising the the sculptor is also demonstrated by other details, for value of the object as status symbol and hypothetically instance the indentation in the middle of the upper lip even representing a rare evidence of social inequality. accented by a narrow rim, or open mouth revealing two This idea may be supported by the unusual embellish- rows of teeth. Of the second head only the neck, some ment of the figure consisting of a multi-rowed stone- remains of the hairstyle and the lower half of the face bead necklace and a scarf or decorative plate around the are preserved (Fig. 14) (position No. 5 in Fig. 5a). It neck and shoulders, decorated with comb impressions. also displays a chin beard and two rows of teeth in the slightly opened mouth.

Shattered heads Therianthropic sculpture All sculptures from feature 10 are fragmented. How- ever, two shattered anthropomorphic terracotta heads In the lowest level of feature 10 (position No. 15 in attracted specific attention due to the form of fragmenta- Fig. 5b) an anthropomorphic head was found exhibit- tion. Both heads are smashed, one is fractured vertically ing both characteristics of man and animal (Fig. 15). (Fig. 13), the other horizontally (Fig. 14). The pieces The eyes and the whole facial structure appear human,

324 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Figure 13. Pangwari, trench E, feature 10. Fragment of anthropomorphic head with unusual physiognomic design.

Figure 14. Pangwari, trench E, feature 10. Lower part of anthropomorphic head.

Figure 15. Pangwari, trench E, feature 10. Therianthropic sculpture combining attributes of man and feline predator.

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 325

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig

Figure 16. Pangwari, trench E, feature 11. Therianthropic sculpture combining attributes of man and bird. while strong fangs protruding from the mouth and a wide nose with a fur-like pattern are unmistakeable attributes of a feline predator. A hood is draped around the head, framing the face, evoking the resemblance of a threatening cobra spreading its neck shield. The fore- head is covered by an oval-shaped plate with decorated edges. Similar figures are known Schädler( 1997: 203, Boullier 2001: Fig. 655) but the Pangwari head is the only example from a scientifically documented context.

Man-bird-hybrid creature

Another therianthropic figure was found in the upper level of feature 11. The find consists of a small delicately worked head with human attributes in its upper part and a large beak instead of a human mouth adding a birdlike component. The beak is divided into an upper and lower Figure 17. Pangwari, trench E, feature 11, small male head beak by a furrow (Fig. 16). The eyes starkly resemble with large ears. those of the sculpture of the man with the sea shell decoration (Fig. 12), so that it may be speculated that they were made by the same craftsman. We identified a dimensional headless male figure, which had originally further therianthropic head from the same feature (Fig. been kneeling on an inverted pot. The figure holds an 17) whose chin beard indicates a male figure exhibiting equally headless four-legged animal in its left hand and disproportionally large and pointed ears with deep au- is fully dressed in garment that features jagged braids ricles that emerge from the equally pointed head cover. (Fig. 18) and a male style loincloth. Above the posterior a flat object of unknown purpose sticks out from the body, apparently a part of the dress. Similar clothing as Kneeling figure with quadruped well as the possibly symbolic pose of holding an animal with the hand placed around the quadruped’s neck are Also unearthed in feature 11, almost half a metre below known from fragments of sculptures found on other the position of the man-bird creature, was a three- sites in the research area.

326 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Figure 18. Pangwari, trench E, feature 11. Kneeling person with quadruped.

Conclusion to other find categories and brings analysis to another level of understanding. As one result, we regard the Most terracotta sculptures that have been excavated at “waste-scenario”, the assumption that the sculptures Nok Culture sites during the last years affirm hitherto were predominantly treated as waste after an unknown known stylistic patterns and fit into a few basic figura- stage of usage as confirmed. Alternatively, fragmenta- tive categories. Thus, we did not necessarily expect tion of the sculptures could not only have occurred due sensational revelations through new findings but rather to a treatment as rubbish but may have another cause, expected them to fit into known patterns. The discover- for instance their handling during ceremonies where ies we made in Pangwari, however, taught us differently. the figures might have been unearthed, used and buried Here it was clearly demonstrated that just one newly again. However, this explanation does not seem to fit uncovered terracotta object can offer a radical change the circumstance of the general mix of terracotta frag- of perspective, providing details that were unknown ments with all other settlement waste quite as neatly. up to this point and could not be inferred from other archaeological materials. Most outstandingly this holds The bulk of terracotta sculptures was found togeth- true for the sculptural evidence of a watercraft, indicat- er with other ordinary finds — in fragmented condition ing transport of people and goods, or the occurrence of only. However, the fragmentation process remains en- a piece of adornment attached to a sculpture like the igmatic. Because all preserved material was excavated, sea shell, an item which points to far-distance exchange theoretically all parts of formerly complete sculptures or trade and maybe even to the open display of social also should have been gathered, hence it should be status by owning and wearing something that was not possible to reassemble the pieces and reconstruct their available for everybody in Nok society. One may even initial form. Yet, this is not the case. This means that go as far as to conclude a social stratification for which either the sculptures are fragmented in such a way that in the terracotta sculptures despite their elaborate endow- the pieces are too small to be refitted or, which seems ment with jewelry no indicators can be found because more probable because not too many tiny pieces have nearly all sculptures are similarly equipped. Altogether been found — parts of the broken sculptures were taken the excavations at Pangwari have greatly enlarged our somewhere else, possibly away from the settlement. knowledge about therianthropic creatures and have given us a notion of the complexity of the beliefs behind them. To explain the findings regarding the terracotta (fragmented sculptures, missing parts, deliberate deposi- In addition, new insights beyond the scope of tion) we propose a model that is based on the assumption the objects themselves have been gained by analys- that the fragmentation of the sculptures is intentional and ing contextual data. Apart from peripheral sections, marks the end of their biography as ritual objects. This Pangwari has been almost entirely excavated, mean- accounts for the total lack of complete sculptures. After ing that practically all materials left behind by a Nok fragmentation, some parts were discarded alongside community during the various occupation episodes are other settlement waste, while other parts were deposited present in the assemblage and available for research. in isolated positions, possibly sacred places of spiritual Consequently the entirety of preserved finds enables us importance in some distance from the settlements. It is to evaluate the significance of the terracottas in relation equally possible that all broken parts were deposited in

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 327

Downloaded from Brill.com10/08/2021 04:36:58AM via free access T.M. Männel & P. Breunig the settlement or on its fringes. However, the latter is References less likely because no complete figures could be refitted from excavated fragments. The “spiritual-place scenario” Beck, C. 2012. Ton-Quelle. Untersuchungen zur materiellen Beschaffenheit und Herkunft der Keramik und Terrakotta der seems more feasible since it would offer an explanation Nok-Kultur Zentralnigerias. In: Ramminger, B. & Stilborg, why these intentional deposits of broken sculptures have O. (eds), Naturwissenschaftliche Analysen vor- und frühge- only been found thrice during ten years of excavations by schichtlicher Keramik II. Methoden, Anwendungsbereiche, the Frankfurt team. Due to the scarcity or total absence Auswertungsmöglichkeiten. Universitätsforschung Prähis- of other finds that could be indicative, such places are torische Archäologie 216. Habelt, Bonn, pp. 251–270. difficult to discover — unless the deposition took place Beck, C. 2014. The clay’s the thing. In: Breunig, P. (ed.), Nok – on the grounds of abandoned old settlements. Pangwari African Sculpture in Archaeological Context. Africa Magna might represent such a place. The two main phases of Verlag, Frankfurt, pp. 247–254. occupation, identified by radiocarbon dating and pottery analysis as described above, fit a scenario in which Boullier, C. 2001. Recherches méthodologiques sur la sculpture people of the Middle Nok period brought shattered parts en terre cuite africaine: application à un corpus de sculptures of sculptures to the places where their ancestors dwelled. archéologiques – en contexte et hors contexte – de la culture Nok (Nigéria), Vol. II. Unpublished Ph.D. Thesis, Université Here they disposed of the fragments by burying them Paris I – Pantheon Sorbonne. in the ground, possibly accompanied by lighting fire because large amounts of charcoal were found in the soil Breunig, P. 1996. The 8000-year-old dugout canoe from Dufuna among the sculptures, backfilled the pit and sealed it with (NE Nigeria). In: Pwiti, G. & Soper, R. (eds.), Aspects of a layer of cobbles. This indicates the intention to part with African Archaeology. Papers from the 10th Congress of the PanAfrican Association for Prehistoric and Related Studies. the sculptures in such a way that they would not return University of Zimbabwe Publications, Harare, pp. 461–468. into the world of the living. Feature 10 of Pangwari can be interpreted this way, however, alternative explanations Breunig, P. 2009. Die früheisenzeitliche Nok-Kultur in Zentral- are conceivable. Middle Nok finds in other sections of the Nigeria, Westafrika. Neues archäologisches Langfristprojekt site point to the existence of a settlement in the vicinity der Deutschen Forschungsgemeinschaft. Archäologisches of Pangwari, encouraging the idea that the inhabitants Nachrichtenblatt 14, 341–360. repeatedly visited Pangwari for similar reasons or to bury Breunig, P. & Rupp, N. 2010. Outline of a new research project their dead in a place of their ancestors. This is indicated on the Nok Culture of Central Nigeria, West Africa. Nyame by the discovery of at least two features situated a few Akuma 73, 46–54. metres west of where the deposit of terracotta fragments was found, which have been tentatively interpreted as Breunig, P. & Rupp, N. 2016. An outline of recent studies on the graves of the Middle Nok period (XRF soil analysis and Nigerian Nok Culture. Journal of African Archaeology 14 (3), 237–255, this issue. 14C dates are not yet available). The relation between settlement and spiritual place and the model outlined above Bronk Ramsey, C. 2009. Bayesian Analysis of Radiocarbon that explains the dichotomy between terracotta fragments Dates. Radiocarbon 51 (1), 337–360. https://doi.org/10.1017/ as waste or as pars pro toto intentionally buried ritual S0033822200033865 objects will hopefully be tested at already known sites in Chesi, G. & Merzeder, G. 2006. The Nok Culture. Art in Nigeria the future. We have not solved the Nok enigma yet — but 2,500 Years ago. Prestel Verlag, München. the research carried out in Pangwari has brought to light many details suited to further refine the existing models. de Grunne, B. 1999. The Birth of Art in Africa. Nok Statuary in Nigeria. Biro, Luxembourg.

Acknowledgement Fagg, A. 1972. A preliminary report on an occupation site in the Nok Valley, Nigeria: Samun Dukiya, AF/70/1. West African Journal of Archaeology 2, 75–79. We are indebted to the German Research Foundation (DFG) for funding the Nok project and to the National Fagg, B. 1945. A preliminary note on a new series of pottery Commission for Museums and Monuments for assis- figures from northern Nigeria.Journal of the International Af- tance and collaboration on a partnership basis. Many rican Institute 15 (1), 21–22. https://doi.org/10.2307/1156826 thanks to all German and Nigerian participants in the Fagg, B. 1948. Masterpieces of early Nigerian art: recently excavation and analysis of Pangwari. Many thanks to discovered figurines of a primitive culture. The Illustrated Eyub Fikret Eyub for the reproduction of the maps, to London News, pp. 586–587. Gabriele Franke and Anke S. Weber for editing this article, to Monika Heckner and Barbara Voss for the Fagg, B. 1990. Nok Terracottas. Ethnographica Ltd., London & The graphics and the drawings and Dr. Ronald Janssen, National Commission for Museums & Monuments, Nigeria. Senckenberg Research Institute and Natural History Franke, G. & Breunig, P. 2014. How old is the Nok Culture? In: Museum Frankfurt, Germany, for his expertise on the Breunig, P. (ed.), Nok – African Sculpture in Archaeological origins of the terracotta shell. Context. Africa Magna Verlag, Frankfurt, pp. 131–136.

328 Journal of African Archaeology Vol. 14 (3) Special Issue, 2016

Downloaded from Brill.com10/08/2021 04:36:58AM via free access The Nok Terracotta Sculptures of Pangwari

Franke, G. 2016. A chronology of the Central Nigerian Nok Culture – 1500 BCE to the beginning of the Common Era. Journal of African Archaeology 14 (3), 257–289, this issue.

Franke, G. 2014. When sherds speak – the pottery of the Nok Culture. In: Breunig, P (ed.), Nok − African Sculpture in Archaeological Context. Africa Magna Verlag, Frankfurt, pp. 169–177.

Jemkur. J. 1992. Aspects of the Nok Culture. Ahmadu Bello University Press, Zaria.

Männel, T.M. 2010. Die Nok-Terrakotten des Fundplatzes Ung- war Kura in Nigeria. Unpublished Master’s thesis, Goethe- University Frankfurt am Main, Germany.

Männel, T.M. & Breunig P. 2014. The Nok Culture’s terracotta figures. In: Breunig P. (ed.), Nok – African Sculpture in Ar- chaeological Context. Africa Magna Verlag, Frankfurt, pp. 188–212.

Nozais, J.P., Boullier C. & Boullier G. 2000. Un malade vieux de 2500 ans... Etude nosologique d’une statue de la culture nok, Nigéria. Bulletin de la Sociéte de Pathologie Exotique 93 (4), 291–293.

Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Cheng, H., Edwards, L.R., Friedrich, M., Grootes, P.M., Guilderson, T.P., Haflidason, H., Hajdas, I., Hatté, C., Heaton, T.J., Hoffmann, K.F., Kromer, B., Manning, S.W., Niu, Mu., Reimer, R.W., Richards, D.A., Scott, E.M., Southon, J.R., Staff, R.A., Turney, C.S.M. & von der Pflicht, J. 2013. IntCal 13 and Marine 13 Radiocarbon Age Calibration Curves 0–50,000 Years Cal BP. Radiocarbon 55 (4), 1869–1887. https://doi.org/10.2458/azu_js_rc.55.16947

Rupp, N. 2010. Beyond Art. Archaeological studies on the Nok Culture, Central Nigeria. In: Allsworth-Jones, P. (ed.), West African Archaeology: New Developments, New Perspectives. Archaeopress Publishers of British Archaeological Reports, BAR International Series 2164, Oxford, pp. 67–78.

Rupp, N. 2014. Communing with the ancestors. The mystery of Utak Kamuan Garaje Kagoro. In: Breunig, P. (ed.), Nok – African Sculpture in Archaeological Context. Africa Magna Verlag, Frankfurt, pp. 214–231.

Schädler, K.-F. 1997. Erde und Erz. 2500 Jahre afrikanische Kunst aus Terrakotta und Metall. Panterra Verlag, München.

Schmidt, A. 2014. Studien zur Struktur der Nok-Fundstelle Pangwari in Zentralnigeria anhand der Befunde und der Fund- verteilung. Unpublished Master’s thesis, Goethe-University Frankfurt am Main, Germany.

Wiethase. H. 2007. Àta Épe. Bogen und Pfeile Afrikas. Wiethase Verlag, Wesseslinden.

Journal of African Archaeology Vol. 14 (3) Special Issue, 2016 329

Downloaded from Brill.com10/08/2021 04:36:58AM via free access