In the open grasslands of western Uganda are found impressive earthworks whose makers and functions are yet to be uncovered both archaeologically and historically. The earth works include the Bwogera at Ntusi, Bigo bya Mugenyi, Kagago( Bigaga) and Kasonko, the latter three are long the southern banks of mid-Katonga river, but all in Mawogola, Masaka district; Munsa za Kateboha, Kakumiro in Kibale district and Kibengo south east of Lake Albert in Hoima district.
There are also import hills like Masaka Hill along the northern bank of mid- Katonga river and Mubende Hill in Mubende town itself with its Nakaima tree; canes and rockshleters which are also significant as abodes of Chwezi spirits in the traditions of Kitara cultural region.
According to court and oral traditions, the sites above are associated with the Bachwezi. Two of the Bachwezi rulers are remembered. They were Ndahura whose capital was at Mubende hill and Wamara whose capital was at Masaka hill. Bigo is associated with Mugenyi, one of the Bachwezi heroes, although in some traditions it is said to have been another of Wamara's capital site as well.
There are quite intriguing puzzles at the moment among scholars about the Bachwezi. For example, the Chwezi period to two centuries, 14th-16th in the interlacustrine historiography and yet we only count the reigns of two kings whom, if we take the generally accepted 27 years geneorological generation, they should have ruled for only 45 years because they simultaneously succeeded each other. It would therefore be equated to the Biblical stories of the old Testament where people lived for several hundred years, if we were to accept that the reigns of Ndahura and Wamara lasted for all the two centuries.
The second set of puzzles are, who constructed the massive and impressive earthworks such as the Bigo (forts) and the earthen features such as the Bwogero (basins) and Ntusi; what type of labour was used; was it voluntary collective labour by both the agriculturist and pastoralist populations or was it forced or coerced labour of the agricultural populations by the Cwezi rulers; for how long did the constructions take and most importantly, what were they constructed for? The third puzzle is, were the Bachwezi real people or spirits? Fourthly was there a Bachwezi empire of Kitara or was there a Kitara cultural region in the which a Cwezi spirit possession cult was practised and in which Cwezi heroes spirits such as Ndahura and Wamara were venerated?
In order to analyse the puzzle above, the archaeological record to date of the periods when some of the sites were operational will suffice. Ntusi functioned between the 11th and 15th century, Mubende hill, functioned between the 12th and 14th centureis.
Munsa earthworks functioned between the 11th and 15th centuries and Bigo bya Mugenyi functioned between the 13th and 16th centureis. Secondly, some of them were occupied for more than two centuries.
Oral and court traditions which attempted to tie the origin of the pre-colonial dynasties [to the] collapse of the polities the 16th century, [and] they became shrines as well. It is shrines for the Cwezi spirit-possession cult. To illustrate this point, a look at Ntusi and its features can be of assistance.
From the archaeological record above, Ntusi is one of the oldest sites. It is the largest site with the greatest density of archaeological depositions, including both surface scatters and in the ground itself. There are two features at Ntusi which are of great interest. The first ones are the two mounds locally known as the Bwogero. There are other small mounds throughout the village but the two features above are the most important . The two big mounds were formed out of systematic pile-ups of refuse of broken pottery, bones, grinding stones, ash and soil for a long time. Archaeological orderly; it took place during the same period of three centuries that the Ntusi settlement lasted and it was carried out according to gender with a purpose.
Equally puzzling is the Bwogero the wash basin. It has been interpreted as a dam, a reservoir and large water well. Its construction include the scraping of clay soil from the valley bottom and the piling up of soil along the sides of the basin to create banks. The most interesting part of the basin is the loop which is locally referred to as the Orurembo or Ekikari, all meaning a capital site or royal palace respectively. Archaeological excavations were carried out in the loop itself and in the middle of the basin but no material findings were discovered. There are two openings to the Bwogero, one near the loop and the other at the opposite. There are two dark spots within the depression just near the loop.
An interpretation of the function of the Ntusi site as an important ritual centre of the Chwezi spirit-possession cult can be attempted. The composition of the archaeological remains both in the mounds and the surface scatters throughout the Ntusi village indicates that both the agriculturists and the pastoralists participated mutually and in equal proportions in the ritual festivities which were carried out at Ntusi.
Such festivities could have been initiation ceremonies. At stipulated periods, say every twenty years, the initiates from all over the Kitara cultural region gathered at Ntusi. They would bring food stuffs in form of agricultural produce and animals, mainly cattle. The divines, mediums and other Mbandwa resided permanently at Ntusi. The preparations for the ceremonies include rehearsals into the rituals of the Chwezi spirit possession cult and the scrapping of the bottom of Bwogero.
The festivities entailed much food preparation and consumption. Equally, since the initiates came from all over the Kitara cultural region they were many, approximately 2,000. Initiation ceremonies included dancing, singing and recitals all in the Chwezi language.
At the end of the festivities a rite of passage was performed. The initiatives would stream in procession to Bwogera to be cleansed / bathed of dirt of earth and transformed in Embandwa, - the initiated would enter the Bwogero through the entrance near the Orurembo. There the Chwezi mediums would sprinkle the initiators with herbal mixtures which symbolised cleansing or bathing. The initiates then gathered in the arena of the Bwogero amidst singing and chanting.
There are two dark spots near the entrance to the basin which might have acted as fire places Ekomi. The initiates would spend a night there undergoing spiritual transformation. The Bwogero for this purpose could be regarded as the spirit world. The following day the initiates would again proceed from the Bwogera through the other entrance which acted as the exited Obwegoromorero. By this stage the initiates would be proper Embandwa.
There was the last ritual which had to be performed at the end of the festivities. This was the clearing of the site of the leftovers and the refuse after the festivities. Most of the refuse was piled up on the mounds and gender was observed in that the males damped on he male mound and the females on the female mound. It could be the case that the female and male initiates were the ones who piled refuse on the respective large mounds while the other participants piled theirs on the smaller ones throughout the site.
From the foregoing it can be concluded that Ntusi was a very important ritual centre of the Cwezi spirit-possession cult. It could not have been a settlement in which its population could have practised both intensive agriculture and extensive pastoralism at the same time because the site is relatively small compared to the high density of the cultural and fossil remains in the mounds and scatters along the surface throughout the site of the one square kilometre. The Bwogero could not have been constructed as a dam or reservoir because the catchment area is equally small.
The other sites especially the earthworks took long to construct and the intricate enclosures in the central ditch system must have ensured high standard protocol before one could reach the medium's residence. The central hill-rock in the Kasonko earthworks made of the soil from the ditch construction was a monument around which people could gather to worship Cwezi hero spirits.
The rock-shelters of Bikekete in the Munsa earthworks were sure abodes of spirits not of residences for the mediums. It had been suggested that the earthworks especially Bigo with its sister earth works of Kasonko and Kagago along the southern bank of mid Katonga river were constructed as defences against the Luo (Babito) invasion. However, evidence exists that such an invasion did not take place at all. The Babito dynasty in Bunyoro Kitara like the Bahinda dynasty in Nkore evolved their states from the peripheries of the former Kitara-like cultural region, from Pawir along Nile and Karpkarungi along the Kagera. During the 16th century, a long and devastating drought, which in Nkore traditions is known as Eijuga Nyoza, hit the whole of the Kitara cultural region. The result was that those who survived the drought deserted the region and regrouped themselves along the peripheries and beyond. Hence the erstwhile flouring of Chwezi polities and their ritual centres and shrines were deserted, so much so that the present inhabitants of the Bwera grasslands only hold them in awe and can only ascribe them to the Bwachezi who were nothing but hero spirits of a spirit-possession cult.