Discussion
p. 25-32
Texte intégral
1A total of 3906 cultural remains were recovered from the excavation at Itagunmodi. Figure 9 summarizes table 1, showing the relative distribution of the major cultural remains. Pot sherds made up the overwhelming majority, with 97.9 per cent; other artefacts were not found in any significant quantity.
Figure 9. Summary of cultural remains

2The issues addressed here include an assessment of the success and importance of the Itagunmodi excavation and the follow-up action needed. It is our intention that this should be the beginning of systematic archaeological research in this area. The Ilesa Post Office rescue excavation in 1959 by Willett was the last dig carried out in this area, until this present one. The effort here is also an attempt to resuscitate archaeological excavation as a whole in central Yorubaland, which has been drastically curtailed since the early 1970s. This area has always been considered the hub of Yoruba historical development as supported by previous work in Ife, Iwo Eleru, Owo and Ila (Eyo 1974; Frobenius 1913; Fagg and Willett 1962; Shaw 1965, 1967; Shaw and Daniels 1984 Willett 1962, 1967; Ozanne 1969; and Garlake 1974, 1977).
3First and foremost, it can be clearly seen that the research has been relatively successful, in the sense that it has opened a new chapter in the archaeology of Ijesaland, and central Yorubaland. For instance, prior to this research, no one expected Itagunmodi to yield any archaeological finds, particularly one as important as a potsherd pavement, because it is a very small town of little importance in the oral traditions of the Ijesa. This research has demonstrated that no area should be underrated archaeologically when striving to arrive at an understanding of the evolution of Yoruba society.
4Another result of the research is that additional archaeological sites have been identified in the town. Among these are the cut, hemispherical, buried pots and the exposed skeletal sites. More important is the discovery of some abandoned sites in the forest zone of a pre-Itagunmodi period. It is hoped that future investigation of these abandoned sites will throw more light on the settlement history of Ijesaland.
5Interest in Itagunmodi was aroused by the close similarity of its potsherd pavement to those at Ife. The result of the excavation has confirmed this similarity. First, the pavement is the edge-laid type, typical of Ife pavement. However, there are two types of pavement in Ile-Ife, the dominant type being pure potsherd, while the other is a stone and sherd type. This latter type was exposed by the work of Garlake at Obalara and the Woye Asiri family lands (1974 and 1977). The first type – potsherd pavement – is very common in central Yorubaland. Apart from being edge-laid, the style or arrangement is the same wherever this type is found. It is commonly laid in a herringbone pattern and closely packed. In addition, the sherds used are usually decorated bodies of pots and not rims.
6Besides this correlation in type of potsherd pavement, ceramic attributes are informative, especially decoration and vessel types. For instance, the major decorative technique is the roulette dominated by cord and maize cob types. This is instructive because carved roulette appears to be completely absent in the collection. Carved roulette was considered by Garlake (1977) to be the earliest technique at Ife; prior to the emergence of cord and maize cob decoration. We hope further investigation at Itagunmodi will help to resolve the issue of whether carved roulette was used. Moreover, it should be noted that the materials being analyzed came from the first spit level of the site and the area excavated so far is still very small. It may not be totally representative of Itagunmodi. Nevertheless, the observed correlations and discontinuities among materials from Itagunmodi and other Yoruba sites must be noted. For example, although bowls from Ife and Itagunmodi are similar in shape, the decorations on Itagunmodi bowls are purely incision with dot punctates. This is not so at Obalara or Woye Asiri – the best researched Ife sites. Some relationships between Itagunmodi and Ife materials can be tentatively confirmed; while a measure of local initiative might be responsible for observed differences. It is hoped that this theme will be pursued further in subsequent investigations at the site.
7In the attempt to establish a chronological framework for the site, two C14 dates were obtained from the Centre de Datation par le Radiocarbone, Villeurbanne, France. The two came from the same square but from different levels (table 3). Sample LY -6287 gives a date of 675 ± 200 BP, while that of LY-6286 is 1565 ± 85 BP. When fully calibrated, LY-6287 falls around the 14th century and LY-6286 indicates the 5th century. However, there are some problems emanating from these dates because the stratigraphic context of the dates is not in agreement. For instance, LY-6286 is from spit level 4 (40 cm) and it gives a date of the 5th century A. D. while LY – 6287 is from spit 6 (60 cm), giving a date of about the 14th century A. D. Interestingly enough this is a primary site, because the pavement was in situ, quite undisturbed and the affected square cut through it. The stratigraphy was clearly uniform, save for slight colour and textural differences in the first 5 cm below the surface, which were noted but not considered significant. Only one cultural stratum was recognized and it was compact throughout. It is possible that the discontinuity in materials may have been effected prior to the construction of the potsherd pavement. Therefore, there is not much that one can say at the moment, other than to ask for a reconfirmation of the sample number/laboratory code; in addition to undertaking further excavation not only at the site but also at other identified sites within the township. This further excavation is of utmost importance in the determination of a chronological framework for Itagunmodi.
8In spite of this drawback, the dates are still informative, because both fall within the chronological range of Ife (figure 10). The 5th century date correlates well with the earliest Ife date of the same period, while the 14th century falls within the age sequence of potsherd pavement in Ife. The earliest date for Ife pavement is the 12th century, while that for the Benin pavement is the 14th century (Connah 1975). Therefore, considering the nearness of Itagunmodi to Ife, it is obvious that its pavement, if actually derived from Ife, must be earlier than those at Benin and other pavements in the forest zone. Thus, we can tentatively date the Itagunmodi pavement site to between the 12th and 14th centuries A. D., pending the time when the issues raised above are resolved.
Figure 10. C14 from southwestern Nigeria

9There is no doubt that the research has achieved a great deal. First, it has demonstrated the need for a shift of focus from well-known ancient Yoruba towns to smaller ones, if we are to have a proper understanding of Yoruba cultural history. A similar call was made earlier by Obayemi in 1976. There is a need to collect information from both types of towns so as to have a complete picture, not only to know the spatial distribution and extent of pre-colonial Yoruba settlements, but also the various associated activities. It is when we are armed with this information that proper and relevant historical questions can be asked.
10Secondly, the research has expanded our horizons in relation to an important urban marker, the potsherd pavement, the spatial distribution of which is yet to be fully understood. Knowing this will surely give us an idea of the configuration of precolonial Yoruba urbanism. Almost all currently identified pavement sites (Ife, Itagunmodi, Ilare, Imesi-Ile, Ikirun, Ila, and Oyan) are in Osun State; the finds imply that these sites were once urban settlements.
11Apart from Ife, the Itagunmodi pavement site is the second to be excavated. The significance of Itagunmodi as an urban centre should not be underestimated, because the potsherd pavement has been identified as a feature of urbanization (Agbaje-Williams 1991). This was earlier borne out by the work of Eyo on Ife (1974). Eyo modified Willett’s typology of the cultural evolution of Ife (pre-classical, classical and post-classical), based on Ife terracotta and bronze heads. He opted for the use of one of the features, ‘pavements’, to designate the stages of cultural development in Ife, thereby identifying pre-pavement, pavement and post-pavement periods.
12The importance of potsherd pavements in the culture of the ancient Yoruba should be recognized. The construction of pavements is a good marker of the zenith of Yoruba civilization because it is an involved and time consuming activity, which marks a level of sophistication in the culture concerned. For instance, more people were involved in pavement construction than in sculpting terracottas and bronzes – which implies that a centralized and powerful political organization was in existence. The emergence of potsherd pavement technology and the possible existence of strong political authority in Ife at an early stage need to be researched.
13According to Ife tradition on potsherd pavement, it was initiated by a female Ooni, called Luwo. The exact date of her reign is as yet unknown, but if available radiocarbon dates on Ife potsherd pavements are considered, it could not be earlier than the 12th century A. D. She was a beautiful, sophisticated and powerful woman who was loved and strongly supported by her subjects despite her firmness. Tradition has it that because she did not like her feet smeared with mud, she ordered the whole town paved with potsherd pavements.
14Based on the current distribution of this feature, it is likely that potsherd pavements were made both in villages and towns, in pre-colonial times, throughout the whole of modern Osun State. The veracity of the tradition is not the object of our discourse, but the fact that the remains of this feature exist today and Ife appears to be at the core of the phenomenon; which would appear to confirm the antecedence of Ile-Ife over other Yoruba towns.
15On the other hand, we may critically examine the situation that led to the evolution of pavements. It is plausible to believe that constructing a potsherd pavement was a cheap ecological device for arresting erosion in the city, because wherever a potsherd pavement has been found, it appears to have minimized the problem of erosion in this humid and highly leached environment. Also, there can be no doubt that potsherd pavements had the same function as our present day tar/concrete pavements and roads.
16Moreover, it should be noted that pottery was the most common utensil of any household in those days. Because of the delicate and fragile nature of pottery, the accumulation of broken pieces must have constituted a problem to any community. Thus, the use of sherds as pavements would be a sound ecological strategy, making use of waste material. This theory is supported by the fact that the sherds used in pavements were ordinary ones, not specially made. For instance, almost half of the excavated sherds were decorated.
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Universal Declaration of Human Rights: English, French, Hausa, Igbo and Yoruba
Foreword by Wole Soyinka
1993
Creative writing, writers and publishing in Northern Nigeria
Abba Aliyu Sani, Jibrin Ibrahim et Emmanuel Babatunde Omobowale
1997
Archaeological investigation of Itagunmodi potsherd pavement site, Ijesaland, Osun State, Nigeria
1991-1992 season
Babatunde Agbaje-Williams
1995
Guerre et société au royaume de Fetu
Des débuts du commerce atlantique à la constitution de la fédération fanti (Ghana, Côte de l’or, 1471-1720)
Yann Deffontaine
1993