Findings
p. 13-25
Texte intégral
1Pottery amounted to 90 per cent of the cultural remains, most of which came from spit level 1 (table 2). Table 1 shows the condition of the sherds, 48.95 per cent being decorated and 40.59 per cent undecorated. The same table also shows that 89.54 per cent are bodysherds while 10.28 per cent are rimsherds. This is quite understandable as it appears that bodysherds are easier to use for pavements than rimsherds. No whole vessel was recovered in the course of the excavation.
2It is obvious that the pottery in level 1 was in relatively good condition, even where the sherds were highly fragmented, worn out and friable. Friability increased with depth and in fact, the state of the pottery below this level was so poor that finds from lower levels were not considered in the present analysis. Research on these other finds is expected to form the thrust of another paper. The pottery remains in Ijesaland need to be properly researched, in the hope that this will establish a ceramic sequence for Yorubaland, which will greatly assist in piecing together Yoruba history.
3Decoration. The decoration is characterized by impression and incision, with the former being dominant (figure 5). Roulette is the most prominent impression pattern and was observed on 60 per cent of sherds. Two types of this roulette were found; twisted cord and maize cob. Quite often, they are found in association with other patterns, like incision. Incised patterns ranged in width from 1 mm to over 5 mm. As in other areas of West Africa which use twisted cord patterns on pottery, the double twist pattern was prominent at Itagunmodi; the size of individual depressions varying from 1 mm to 5 mm. Twisted cord roulette appears to be a characteristic decoration of pots, not bowls (figures 5, 6 and 7). Associated motifs are usually oblique, diagonal or horizontal – a vertical orientation was rarely seen.

Table 2. Stratigraphy of Extravated Remains
Figure 5a. Fragments of potsherd decorations (nos 1-6): twisted cord: 1, 5; maize; 2, 3, 4; punctates; 4; brushmarks: 6

Figure 5b. Fragments of potsherd decorations (nos 7 – 20): twisted cord: 14, 15; maize: 7; incision: 10, 13, 16; punctates: 9; brushmarks: 18; composite: 8, 11-13, 17, 19; channelled: 20

4The same observation applies to maize cob roulette patterns at Itagunmodi, most of which are well-preserved. The shaft of the cob was used, which creates a raised geometric surface pattern.
5When twisted cord and maize cob roulettes were found on the same sherd, the decoration type was classified as composite. The same associated motifs prevail. Other composite decoration types were roulette-incision and incision-punctation. Both were evenly distributed; although associated with different vessel types. For instance, roulette-incised decoration is usually of the twisted cord type with grooving, thin incision or channels – grooving being more prevalent. This decorative pattern was found on potsherds from pots of different sizes and shapes. On the other hand, incision and punctate designs are restricted to bowls. The incision, in these cases, is very thin, like a knife cut, and may be horizontal, vertical or diagonal. The punctation is either horizontal or vertical-dot. (figures 5, 7 and 8). This dot punctate pattern is similar to the decoration on isaasun lids at Old Oyo (Agbaje-Williams 1983; 287). Dot punctates appear to be mainly characteristic of bowls, especially when in association with incision. The type of incision rarely goes beyond 1 mm in width.
6Incision patterns were classified by their width. Incisions as thin as a knife’s edge (less than 1 mm) were considered to be fine and thin. They are found with dot punctates on bowls (figure 8, nos. 2 and 3). Others, more than 1 mm, but less than 3 mm wide, are classified as grooves and are commonly found on pots. Those between 3 mm, and approximately 5 mm in width are described as channels (figure 5, no. 20). Apart from these, there were larger, cruder incisions made by a sharp, short, deep gash with a thick wide cutting edge. The first three incision types have also been identified at Old Oyo, though the thicker one has not been seen (Agbaje-Williams, 1983). The curved incision pattern was identified, but was not as prevalent as in the site at Old Oyo (figure 5, nos. 11, 16 and 17).
7Another type of decoration, fine brush-marking or ‘cloth drag’, made by using a damp cloth or possibly a brush-like tool to wipe or rub the leather hard surface of pottery when not yet fired, produces a myriad of lineal striations. It is only found on pots and not on bowls at all (figure 5, nos. 6 and 18). Brush-marking in its true form is yet to be identified at Itagunmodi; a small number of cloth drag examples were found.
8Vessel Types. In the identification of vessel types, a total of 393 rims (10.28 %) were identified, from the pavement remains, along with associated vessel parts such as lid and base fragments. It is interesting to note that no whole vessel was found, nor were we able to reconstruct the sherds into vessels. Thus, to a great extent the analysis of vessel types depended on rims, of which 151 were decorated. As in other Yoruba sites, the two major vessel types recognized at Itagunmodi are pots and bowls, which have been further sub-classified into smaller types. Three major attributes of the rims were used in this analysis. These were the in-turned, out-turned and straight/vertical rim profiles. We also used the wall thickness of the rims as an identifying factor, as it is greater in pots than in bowls. For instance, the wall thickness of pots around the ‘lip’ area may be as much as 18 mm, tapering down to less than 10 mm below the neck area. The average wall thickness of pots is 7 mm, (figure 6, nos. 1, 2 and 7) while that of bowls is around 3 mm (figure 6, nos. 3 and 6).
9The out-turned necked pots were the most common and were of different sizes. They were often globular in shape, with an average diameter of 23 cm. These pots are similar to those identified by Garlake at Obalara (Ife) as pots with spherical bodies (1974: 134). Some of the sherds were decorated while others were not. The way and manner the lip was formed varied. Some rims have ‘flip over’ lips where the lip hangs like the ears of a dog, others are rolled (rounded and thick), and a few are flat at the lip.
Figure 6. Pots: necked, out-turned and globular: Fluted: 1, 2, 4; Flat lip: 3, 7; Rolled lip: 5, 8; Flip over lip: 6

Figure 7. Bowls: Open bowls, incised and punctate: 1, 2, 4, 5; Carinated bowl fragment: 3; Undeveloped burnished open bowl, 15mm rim thickness: 6; Stool fragment incised: 7; Decorated bowl lids: 8, 9,10, Undecorated constricted bowl: 11; Shallow bowl, plate: 12

10Occasionally, grooving at the lip occurs (figure 5, nos. 13-20). Most of the rims are too fragmentary for detailed analysis.
11From the variety in thickness of rims and sherds of pots at Itagunmodi, it can be deduced that these pots must have served different functions in the past. For example, figure 5, no. 20 is similar to the ape type of pot which can be used to fetch water and also to cook. This type has also been recognized at Old Oyo (Agbaje-Williams, 1983: 272-273). Globular pots could have been used to fetch water, but not to cook, because of the narrow necks and rims. Another type of pot is the open-mouthed neckless pot which has a large rim diameter. This type is generally put in one place, usually in the cooking area, and used as a water container. The final type identified was oru, a medicinal pot, or sometimes called a children’s pitcher. Unfortunately, only basal parts were recovered from the excavation, the rims were missing. This type of pot is still common in Yorubaland.
12Criteria used to classify the bowls were diameter as against depth, an average wall thickness of 3 mm and the surface finish. The sherds from the bowls were fewer in number, but finely made, with a good (well burnished) surface finish; their colour ranged from dark grey to greyish black.
13Unlike the pots, which were primarily decorated with roulette patterns (twisted cord and maize cob), the bowls were incised with dot punctates. The incisions are not more than 1 mm in width. Generally, the motifs fall within the following classifications – vertical, horizontal and diagonal (slanting).
14Among the sub-types identified are carinated bowls, isaasun, open bowls with out-turned rims, open bowls with vertical/straight rims and open bowls with in-turned rims (closed mouth bowls). Of these, the open bowls with out-turned rims appear to be the dominant form (figure 5, nos. 1-13). Flat-bottom bowls (plates) were also identified in the collection even though the rims were missing. Other sub-types included lids, which are typical of isaasun (soup bowls), with striated dot-punctate decoration. The incisions on lids are similar to those on the bodies of bowls. Attached to the centre of the lid is a small knob for lifting it. The decorations on the lids are similar to those from Old Oyo (Agbaje-Williams 1983: 287, plate xx). Itagunmodi pottery is typical of Yoruba pottery in general, because both in form and decoration, it compares favourably with known pottery from Ife and Old Oyo.
15The other attributes to be discussed are surface finish, colour and fabric. The first two appear to be functionally induced because they are peculiar to certain vessel types. For instance, nearly all the pots are unburnished and dark brown; while the bowls are smooth surfaced, fully burnished inside and outside, and shining like glazed ware. The bowls vary from dark grey to greyish black in colour. These functional differences are also reflected in the fabric of the ware; the bowls are always fine grained but the pots are coarse. Some of the inclusions in the pots were easily identified as quartz and laterite. Identification of inclusions in the finer grained bowls will however require detailed microscopic analysis; such analysis is also necessary before definite conclusions can be made on the attributes described.
Lithics
16The second most common form of material culture recovered from Itagunmodi were lithics. In all, sixty-two lithic pieces were recovered from six of the nine squares (table 1) and can be divided into three main categories – worked/used, semi-worked/used and waste. All were made from rocks found around Itagunmodi environs, such as granite-gneiss, banded gneiss, quartzite, quartz and sandstone.
17The worked/used lithics found were upper grinders (omo olo), but they are very small indeed. Four of these were made from granite-gneiss, the dominant rock of the area. These grinders are finely formed, with a good smooth grinding surface – one of them has two grinding surfaces. Considering its size and weight, it must have been used by a young person; or, perhaps it was functionally specific – that is, used to grind a particular item, for which no other grinder would have been effective.
18On the other hand, the semi-worked/used lithics were mostly banded gneiss in the process of formation, perhaps into grinders. They are very crude in every aspect, with an undeveloped grinding surface.
19The third category is waste and came from quartzites and quartz pieces – a few pieces were laterite. (table 1 and figure 8).
Faunal Remains
20Few faunal remains were found in this excavation; perhaps because of the humid, acidic nature of the soil and because of the eating habits of the people, which until the present day include the thorough chewing of small animal bones. This still prevalent practice deprives archaeologists of zoo-archaeological information, as the bones are usually fragmented and easily decompose.
21In spite of these problems, fifteen faunal pieces were recovered (table 1) and although they were too fragmentary for any detailed analysis, some broad types were identified – snail shell, cowrie shell, a few lower bone fragments of ovicaprids, and the toe of an antelope (Cephalophus rufilatus).
22The majority of bone pieces were unidentifiable, but it is likely that they belonged either to ovicaprids or to antelopes because these constitute the majority of the fauna of this region. The greater number of shell pieces were snail shell pieces, some of which were burnt; while there were only fragments of cowrie shells (Cyprae sp.).
Miscellaneous Remains
23The last classification of cultural remains were assorted materials put together under the group title, miscellaneous. Like the faunal remains, they were found in fragments. Among these were nails (2), metal scraps (1), enamelware (1), and glass (1). All were found within the first spit level of squares 21 and 25.
Figure 8. Various excavated material – Globular pot, composite design: 1; Punctate incised open bowl: 2, 3; Globular pot with twisted roulette: 4; Decorated stool fragment: 5; Incised sherd: 6; Semi-worked upper grinder; onto olo: 7; used mini-upper grinder; omo olo: 8.

24In addition, charcoal pieces and some burnt palm kernels, were recovered from square 21, between spit levels 4 and 6. Two of these were sent for radio carbon testing. The results of the tests, which came in January 1994, are presented in table 3.
Table 3. C14Dates

•IM = Itagunmodi Methodist Church
92 = Year of excavation
21 – Square
4 = Spit Level
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