A Five-Day Week in Rwanda in the Second Half of the Nineteenth Century and in the Early Twentieth Century

скачать Автор: Tymowski, Michał - подписаться на статьи автора
Журнал: Social Evolution & History. Volume 22, Number 2 / September 2023 - подписаться на статьи журнала

DOI: https://doi.org/10.30884/seh/2023.02.03

Michał Tymowski, Professor Emeritus, University of Warsaw

The aim of the author of the article is to examine the role of the week and the related dimension of serfdom (corvée) in the treasury system of Rwanda in the second half of the 19th century and the beginning of the twentieth century. The author discusses the current state of research and various opinions on the credibility of written and oral sources relating to this issue. Subsequently he quotes and analyzes a hitherto unknown source regarding the functioning of the week and serfdom, i.e. fragments of Jan Czekanowski's manuscript diary from field research in Central Africa, including Rwanda, in the years 1907–1909. Czekanowski not only confirms the existence of the week – corvée system in the central provinces of Rwanda and draws attention to the conflicts related to its operation, but also describes the methods of legitimizing this form of duty used by the ruling group in Rwanda.

Keywords: Rwanda, Early State, second half of the nineteenth century and the beginning of the twentieth century, the week and serfdom (corvée), written and oral sources

Henri Claessen and Peter Skalník classified pre-colonial Rwanda as an early state from the very beginnings of the work undertaken by research group they had created (Claessen, Skalník 1978, 1981; Cohen 1978: 66; Steinhart 1978: 148; Claessen 1981: 59–86). This decision was validated by the research of the same group in the years that followed (Claessen, van de Velde 1987, 1991; Trouwbourst 1987: 127–137; 1991: 97–107; Southall 1991: 75–96). In the works cited above, two aspects of Rwanda's early state organization in particular are examined (with references to earlier studies): the power legitimatization methods analyzed comparatively, and the treasury system.

The social and economic structures and treasury system that existed in pre-colonial and early colonial Rwanda received much attention in publications beginning with the 1950s. The works of Jacques Maquet, Alexis Kagame and Jan Vansina drew attention to the connection between the early treasury system, which began to take shape in the sixteenth century at the latest, and the structures of state governance by individual Tutsi's armies (Maquet 1954: 151–154; Kagame 1961: passim; Vansina 1962. See also Trouwborst 1991: 102–103). During the nineteenth century, especially during the reign of Kigeri Rwabugiri (ca. 1860–1895) and his numerous conquests, royal power, the system of state governance, including the duty, services and tax collection system, were reorganized. This concerned Rwanda's central provinces above all. Areas lying on the periphery were subject to a different type of exploitation – the provision of tribute, and in case of refusal was enforced by armed raids and looting.

The population subject to the Tutsi (mainly the Hutu and Batwa) provided duties and services to the ruler, his representatives in the provinces, and to local chiefs. These obligations were provided in the form of a wide variety of goods and labor: primarily in the form of crops and beer, or in crafts such as pots, mats, bark cloth, and blacksmith products. Others were provided in the form of services or specialized work, such as the construction of houses for the ruler and chiefs, and transportation of goods. Some duties were provided more intensively on a periodic basis. For example, before war expeditions, blacksmiths provided more heads for arrows and javelins, Batwa hunters provided more meat from hunted game, porters had to intensify their goods transport services (Maquet 1954: 123–125; Hertefelt 1962: 34–35).

It was a complicated and flexible treasury system, subject to change depending on the specialization of individual groups of subjects in various works, on the specialization of the regional economies, and on the permanent or changing needs for specific goods and services. Among these various types of duties, taxes and services there was also the work of Hutu peasants in the chiefs’ fields. During the early colonial period, when Rwanda was ruled over by the German Empire, its internal economic structures, including its treasury system, were left largely intact. Scholarly literature devoted to it covers two periods together – the second half of the nineteenth century and the beginning of the twentieth century.

In this article, I will concentrate on the part of the duty and obligation system that was associated with the introduction of the 5-day week in Rwanda. The introduction and the assessment of corvée in Rwanda have already been studied and discussed, especially in the 1970s and 1980s. My aim here is to go over this discussion and, while doing so, to make use of a new hitherto unknown source concerning the week.

Whether the 5-day week determined the assessment of the corvée provided and, if so, since when, remains debatable. Maquet only states that, ‘as far as corvée was concerned, the chief of the hill (i.e., an agricultural settlement) was obliged to deliver to the residence (a chief of higher rank) the required number of workers, and could order others to work for him’ (Maquet 1954: 125 ‘en ce qui concerne les corvées, le chef de colline devait fournir à la résidence le nombre de travailleurs requis et pouvait faire travailler les autres pour lui’). The undefined and arbitrary measure of such a form of corvée and of the number of people subject to it allows us to assume that Maquet was describing this obligation as it functioned before the ‘weekly reform’. Marcel d’Hertefelt, on the other hand, justified the distinction between the obligations and services, the collection of which was organized by the armies active in various provinces, and the corvée imposed on Hutu peasants ‘through administrative means’. The latter work lasted for ‘a certain number of days per week’ (Hertefelt 1962: 35 ‘travaux de toutes sortes imposés aux Hutu, à certaines époques ou pendant un certain nombre de jours par semaine’; see also Chrétien 2003: 155). Thus, there were two coexisting corvée exacting systems, one of which took into account the existence of a week.

In 1974, the journal Cahiers d’études africaines published an issue with a set of articles on the society and economy of pre-colonial and early colonial Rwanda. A research team made up of French and Rwandan scholars conducted research in the field between 1967 and 1972, recorded oral traditions and oral accounts about the organization of the pre-colonial state, and early colonial Rwanda, particularly Rwanda's treasury system; determined the existing types of duties, works and services and the manner of their collection. One of the teams goals was to take into account oral testimonies given by Hutus about the system and practices of collecting duties in pre-colonial and early colonial Rwanda (Vidal 1974a: 5).

In keeping with this assumption, Joseph Rwabukumba and Vincent Mudandagizi cite two accounts on the introduction of serfdom. One is from a Hutu informant, the other from a Tutsi informant (Rwabukumba, Mudandagizi 1974: 21–22). Both confirm that Hutu peasants initially paid taxes in crops and supplied beans and sorghum to the chiefs in quantities set by custom. During the reign of Kigeri Rwabugiri, however, an ubuletwa obligation was introduced, consisting of labor to cultivate the land of the ruler and local chiefs during two days out of 5-days week. According to the recorded accounts, the ruler introduced it after the Ankole war expedition. During this expedition conflicts arose over the suitability of troops made up of Tutsis and troops made up of Hutus (the two accounts agree about the existence of the conflict, but differ on the bravery or lack thereof among the Hutus).

In any case, the ruler exempted the Hutus from having to take part in armed expeditions, and in return introduced the ubuletwa obligation. According to the article's authors ‘serfdom was the most hated and most degrading manifestation of [land] dependency, even if it did not apply to all Hutus. It signified the enslavement of the Hutus to the dominant minority’. The author of the account, of Hutu descent, concludes his assessment with a more moderate statement ‘All this did not exist before [the days of] Rwabugiri’ (Ibid.: 21. ‘La corvée fut, de la dépendance foncière, la manifestation la plus haïe et la plus humiliante, même si elle ne frappait pas tous les Hutu: elle signifiait la servitude des Hutu à l ҆égard de la minorité dominante’ and ‘tout cela n’existait pas avant Rwabugiri’).

In the research conducted by Claudine Vidal, numerous accounts confirm that the use of the weekly system of two days of work out of five was extended at a later time (presumably during the colonial period). Initially, from the time of Rwabugiri, work on this basis was done by Hutus from the farms at their disposal. Two days of work for the lord left the rest of the week for the peasant, usually two days of work for himself and one day of rest. Subsequently the same approach was used with regard to Hutu peasants who possessed no land, or even an iron hoe needed for cultivation. Vidal refers to them as journaliers (day laborers). They were lent a hoe by a more prosperous person, in return for which they were required to work two days out of five, or four half-days during a five-day week (Vidal 1974b: 61–67).

In his 1981 article on duties and services provided in the royal domains in pre-colonial Burundi, Émile Mworoha included remarks on Rwanda and the obligation of Hutu subjects to work according to the two out of five days rule (Mworoha 1981: 752–768). This researcher negated the regular use of the weekly corvée system: ‘We are very skeptical about publications which, for example about Rwanda, have presented regular corvée of two days out of five as a customary reality. This figure, repeated in publications devoted to that country, has acquired official status to the point of being recited in ‘oral traditions’ (Mworoha 1981: 763–764 ‘…nous sommes très sceptique devant les écrits qui, par exemple à propos du Rwanda, avaient présenté comme une réalité coutumière une corvée régulière de 2 jours sur 5. Ce chiffre, répété dans les publications consacrées à ce pays, a pris valeur officielle au point d’être récité dans les ‘traditions oralesˮ’). Mworoha further points to the difficulty of reproducing quantitative information, especially when based on oral tradition. He questions the durability of memory, especially with regard to numbers. He reminds us that, in this case, we are studying a society that did not know our modern calculation rigors. Furthermore, Mworoha believes that the term buletwa (ubuletwa) seems foreign to the Kinyarwanda language and ‘seems to us to be related to Swahili rather than Kinyarwanda’ (Ibid, ‘nous semble se raccrocher plutôt au kiswahili qu’au kinyarwanda’). He proposes, therefore, to conduct linguistic studies to evaluate the authenticity of the institution of the week. Jean-Pierre Chrétien (2003: 155) does not deny that the term might have been borrowed from Swahili, but he sees no reason to deny on this basis the existence of the form of corvée introduced in Rwanda in the second half of the nineteenth century.

In conclusion to his reflections, Mworoha added that ‘it is only in the last few years that several authors demonstrated the recent nature of the buletwa corvée royale in this country.’ (Mworoha 1981: 764 ‘Il a fallu attendre ces dernières années pour que plusieurs auteurs montrent le caractère récent de la corvée royale du buletwa dans ce pays’). In so doing, he cites the Rwabukumba and Mudandagizi article I mentioned earlier. Let us recall, then, that only publications from the 1950s and 1960s did not specify the time of the introduction of the ubuletwa. Since the 1970s, the consensus in research has been that it was the period of state reorganization during the Kigeri Rwabugiri era, i.e. the second half of the nineteenth century.

Mworoha's reservations about the reliability of oral transmissions, the imprecise nature of the numerical information contained in these transmissions, and the different computational rigor that existed in pre-industrial cultures than in modern cultures, are worthy of critical consideration. It is true that the attitude to numbers in early state societies differed from that existing in contemporary societies (Kula 1984). This applies to both oral and written accounts. If we read in an early medieval chronicle that the opposing army was a hundred thousand men strong, it only means that in the chronicler's view that army was numerous.

The attitude to numbers such as two or five and to assessing works and obligations and corvée was quite different, however. In illiterate societies, memory worked quite well for such things. Let me cite the example of the early states of Central Europe, which took shape between the ninth and the eleventh centuries. During this period, in the duchies (kingdoms) of Bohemia and Poland, so-called ducal law was recorded in memory and transmitted orally. It addressed, among other things, the treasury system of the duchies. Relying on memory, knowledge of the ducal duties and services, their names, measure, collection method, persons and families obliged to pay duties, and control of the system made the effective operation of these early states possible (Bardach 1965, vol. 1: 73–74, 137–144; Buczek 1958; Krzemieńska, Třeštik 1964; Kučera 1964; Łowmiański 1970: 129–150; Modzelewski1964.

And lastly, why should we assume the existence of difficulties in remembering the functioning of a weekly two out of five day system of duty in the form of corvée in Rwanda, when we do not negate the memory of dozens of other widely different types of duties, services and taxes and forms of labor performed by the population dependent on the Tutsi?

It would thus seem that Mworoha underestimates the significance and the effectiveness of memory in the organization and the collection of duty in early states during the period of illiteracy. This applies both to early states in Africa and on other continents.

A very important allegation that Mworoha raises concerns the lack of written sources about the functioning of the week in Rwanda of the second half of the nineteenth and early twentieth centuries. Mworoha considers traditions and oral accounts to be an insufficient source base. So I would like to recall one well-known example of the use of these sources and add a new example that was hitherto unknown.

At the beginning of the 1970s, Jean-Pierre Chrétien (Chrétien 1972) made use of several written sources corroborating the existence of the week and the ubuletwa in Rwanda. He made use of them in an article concerning the revolt of Ndungutse, who claimed to be the grandson of Kigeri Rwabugiri. In 1912, Ndunguste revolted against the Rwandan king Yuhi Musinga. He gained the support of the inhabitants of Rwanda's northern periphery, lying between the provinces of Mulera and Ndorwa, in the vicinity of the Bolero Lake (Chrétien 1972: 646–647). He sought the support of different social groups. He promised Hutu peasants the abolition of corvée (ubuletwa). In this promise, we find a confirmation of the introduction of this type of work obligation in these areas of the Rwandan periphery and of the social tensions that this occasioned (Chrétien 1972: 648–649, 667–668, 672). Ndungutse's revolt was crushed by a German expedition supported by units loyal to king Musinga.

The course of these events was observed by the White Fathers from their mission in Rwasa. It was described by Father Dufays, in a work that was published in 1928. In it, the missionary recalled the promises made by Ndunguste concerning the abolition of the ubuletwa (Dufays 1928: 75, after Chrétien 1972). Materials concerning difficulties in the introduction of corvée and the revolt can also be found in the report in the letter resident Richard Kandt from 25 March 1910, and in the report of lieutenant Gudowius, the ad interim resident in Rwanda in 1911–1912, and in documents produced by the mission in Rwasa (Chrétien 1972: 650, 656, 657, 662, 672).

An even earlier, and hitherto unknown, written testimony concerning the functioning of the week in Rwanda in the early twentieth century is the manuscript of Tagebuch der anthropologisch-ethnolo-gischen Kolonne der Zentralafrikaexpedition in den Jahren 1907–1909, written by Jan Czekanowski (the Czekanowski Diary). This manuscript, which forms a part of the family collection of Anna Czekanowska-Kuklińska, is deposited at the Manuscripts Department of the University Library in Warsaw. Only small excerpts have been publicly known until now, but the entire text is to be published in the second half of 2023.

Czekanowski took part in the research expedition to East-Central Africa funded by Adolf Friedrich, Prince of Mecklenburg. The expedition was made up of five research columns (caravans). As the anthropologist of the expedition, Czekanowski headed one of the five columns. During his two-year stay in Africa (1907–1909), Czekanow-ski collected anthropological, ethnographic and linguistic material, including material concerning local political organizations, social structures, economy and culture. He made systematic, day-to-day entries into his Diary, in German and French, with fragments in Swahili and other African languages. It also contains sentences and words written in Polish and Russian. The entire text is over 1,000 pages long. Czekanowski communicated with Africans in Swahili and also understood other local languages. It is a highly credible source.

The week and the measure of corvée related to it are the subject of two excerpts from the Diary. The first, from early August 1907 (Volume 2, p. 174), was written during Czekanowski's stay in Issawi. The village lay south of Niansa, the seat of the Rwandan ruler. The passage interesting us is the conclusion of a long excerpt from the Diary concerning the legends about the creation of the world and people, as told by the Tutsi:

On parle [de] trois différents mutabazi. Le mutabazi de Kigwa est en relation avec le Déluge. Il a sauvé les gens... Le nom de ce mutabazi est Chumweru et en son souvenir on a mis un jour de repos après cinq jours de repos [?].[1] Ce jour-là est appelé chumweru. Les Bahutu ont une autre légende de [la] création et [ne] connaissent pas celle-ci, qui est gardée par les Batutsi et racontée aux enfants avant la mort [des vieux?].

Cette semaine de six jours est connue seulement dans le pays où les Batutsi ont des prestations et dans le pays où les prestations n’existent pas, n’existe pas aussi la semaine [comme?] dans ce pays de Bahutu – Mulera par exemple. Dans Ruanda le Muhutu travaille 3 jours pour le chef, deux pour [soi] et [i1 se] repose.

[We’re talking [about] three different mutabazi. Kigwa’s mutabazi is related to the Flood. It saved the people... The name of this mutabazi is Chumweru, and in his memory there is
a day of rest after five days of rest [?].[2] This day is called chumweru. The Bahutu have another legend about [the] creation and [don’t] know this one, which is kept by the Batutsi and told to the children before the death [of the elders?].

This 6-day week is known only in the country where the Batutsi have duties to collect, while in the country where duties do not exit, the week doesn’t exist either, as in the country of the Bahutu – for example in Mulera. In Rwanda the Muhutu works 3 days for the chief, 2 days for himself and for 1 [day] he rests].

The second fragment concerning the week comes from an entry Czekanowski made in Volume 3, page 232 on 31 August 1907 in Kapgaye, in the territory of Marangara. This was after the caravan had left Niansa in a northerly direction, on the way to Rwasa. In this fragment, Czekanowski combined what he learned from informers with his own observations:

Pendant l’affichement des tambours au gichurasi,[3] le roi ne juge [pas], ne fait [pas] d’épreuves, ne reçoit [pas] des cadeaux. Après la fin de gichurasi [a lieu] une grande fête pour laquelle viennent tous les ntore. Kukura gichurasi – [signifie] arracher, enlever gichurasi finir ([en polonais] zdjąć żałobę. Wielkanoc? Też święto księżycowe) [Enlever le deuil. Pâques? Aussi fête lunaire.].

On dit qu’entre 8–10 h du soir ordinairement on ra-conte sur le mutabazi?!?!? La semaine a ici 5 jours, 2 jours pour le chef, 2 pour eux, 1 de repos. C’est chez Nshoh-zamihigo – Tandis que chez Kabare et Rudegambia – ils ne cultivent pour les petits chefs sur la colline. Ils font une corvée plus forte peut-être. Ils vont pour la capitale [à] construire des maisons de rois, Rudegambia et Kabare en partant de pombe. Tandis que les petits chefs sont forcés d’acheter de leur cultivateur avec de lait, beurre.

[During the presentation of the drums in [the month of] gichurasi, the king does not hold ordalias, and does not accept gifts. After the end of the gichurasi, a great feast is held, to which all ntore come. Kukura gichurasi – [means] to uproot, to lift the gichurasi (to end, remove the mourning. Easter? Also a lunar festival).

It is said that usually around 8–10 PM at night one speaks about mutabazi?!?!? The week has 5 days here: 2 days [they work] for the chief, 2 for themselves, and one to rest. That’s how it is with Nshohzamihigo. With Kabare and Rudegambia they [Hutu] do not cultivate on behalf of the smaller chiefs on the hill. Perhaps they provide even more corvée. They go to the capital to build houses for the king, Rudegambia and Kabare set off with pombe (beer). Whereas the smallest chiefs are forced to buy from the cultivators in exchange for milk or butter].

The quoted excerpts from the Diary, recorded on the basis of the Czekanowski's own observations and the accounts he gathered from local informants, are undoubtedly a written source documentation that Mworoha hoped would confirm the existence of the week and the assessment of corvée. The diary remaining long only in manuscript was, of course, unknown. The quoted passages fully confirm the existence of the week in early twentieth-century Rwanda and the measure of corvée associated with it. They also confirm that it was one form of duties and labor, among numerous other types of fiscal provisions. They are testimony to the slow introduction of this type of obligation. Several decades had passed since it was begun by Kigeri Rwabugini, but it was still used primarily in the central provinces of Rwanda. Even there, the measure of corvée was not yet clearly fixed. In Marangara, a territory not too distant from the center of the state, two systems continued to coexist: among the great chiefs, it was corvée defined as two days out of a 5-day week; while among the lower-level chiefs, it was at the discretion of the one who collected the tribute. Czekanowski estimated the latter of the two systems as being more onerous. He did not settle whether the discrepancy between corvée of three days out of six per week and two days out of five existed in reality (as a regional difference, for example), or is it was a mistake on the part of the informant. In his basic printed work, Forschungen im Nil-Kongo-Zwischengebiet (1911–1927), in volume one, devoted primarily to Rwanda, in the chapter on calendars used by peasants and herders, he included only one sentence at the end regarding the week: ‘Die Wochen haben fünf
bis sechs Tage und sind durch Verhältnis der freien und Frondiensttage bestimmt’ [the week has five or six days and is characteri-zed by the ratio of free days to corvée days] (Czekanowski 1917: 297).

The content of both excerpts from the Diary goes beyond simply stating the existence of corvée and its measure. Czekanowski recorded important data about the methods used to legitimize this type of obligation. It made reference to myths and legends about the origins of the world and people, and the division between Tutsi and Hutu. The first of the quoted passages was preceded in the Diary by just such a legend. However, Czekanowski noted that it was passed down by the Tutsis and that the Hutus had a different legend about Creation. Thus, one can guess that such legitimization of duties and services, including of the ubuletwa, had a limited impact on Hutu subjects. State coercion was of greater importance.

The second account is mostly a description of corvée, its varied measure, and its coexistence with other services, or the collection of different services alternately. But in this case as well, Czekanowski mentions the ceremonial month of gichurasi and the recitation during its course of the legend concerning mutabbazi; a legend, therefore, which among its various functions, legitimized the ubuletwa.

Intense methods used to justify the measure of corvée, still required so many so many years after its introduction, may confirm the existence of social tensions surrounding this type of obligation.


[1] Apparemment, c’est une erreur, ça aurait du être – travail.

[2] This is likely an error, and should have been ‘work’.

[3] Gichurasi – a month of sadness, a description of the ceremony associated with it is to be found in vol. 3, p. 220 of the Diary.


Back, L. R. 1981. Traditional Rwanda: Deconsecrating a Sacred Kongdom. In Claessen, H. J. M., Skalník, P. (eds), The Study of the State (pp. 15–33). The Hague, Mouton.

Bardach, J. 1965. Historia państwa i prawa Polski. Tom I do połowy XV wieku. Warszawa: PWN.

Buczek, K. 1958, Książęca ludność służebna w Polsce wczesnofeudalnej, Wrocław: Ossolineum.

Chrétien, J.-P. 1972. La révolte de Ndunguste (1912). Forces traditionnelles et presion coloniale au Rwanda allemand. Revue française d’Histoire d’Outre-Mer, t. LIX, no 217: 645–680.

Chrétien, J.-P. 2003. L’Afrique des Grands Lacs. Deux mille ans d’histoire. 1st edition 2000. Paris: Flamarion.

Claessen, H. J. M. 1981. Specific Features of the African Early State. In Claessen, H. J. M., Skalník, P. (eds), The Study of the State (pp. 59–86). The Hague: Mouton.

Claessen, H. J. M., Skalník, P. (eds.) 1978. The Early State. The Hague: Mouton Publishers.

Claessen, H. J. M., Skalník, P. (eds.) 1981. The Study of the State, The Hague, Mouton.

Claessen, H. J. M., van de Velde, P. (eds.) 1987. Early State Dynamics. Leiden: Brill.

Claessen, H. J. M., van de Velde, P. (eds.) 1991. Early State Economics. New Brunswick, London: Transaction Publishers.

Czekanowski, J. 1907–1909. Diary: Tagebuch der anthropologisch-ethnolo-gischen Kolonne der Zentralafrikaexpedition in den Jahren 1907–1909, manuscript that is part of the family collection of Anna Czekanowska-Kuklińska, deposited at the Manuscripts Department of the University of Warsaw.

Czekanowski, J. (1911–1927), Bd 1–5, Forschungen im Nil-Kongo-Zwischengebiet, Leipzig, Klinkhardt & Biermann.

Czekanowski J. 1917. Forschungen… Bd. I, Ethnographie. Zwischenseengebiet. Mpororo-Ruanda. Mit 4 Tafeln und einem musikalischen Anhang von E. M. von Hornbostel. Leipzig: Klinkhardt & Biermann.

Dufays, F. 1928. Pages d’épopée africaine. Jours troublés. Souvenirs d’une mission en fondation au Rwanda belge. Bruxelles: Librairie coloniale R. Weverbergh, (after Chrétien 1972).

d’Hertefelt, M. 1962. Rwanda. In D’Hertefelt, M., Trouwborst, A. A., Scherer, J. H. (eds.), Les anciens Royaumes de la zone interlacustre meridionale – Rwanda, Burundi, Buha (pp. 9–112). Tervuren: Musée royal de l’Afrique Centrale.

Kagame, A. 1961. L’histoire des armées bovines dans l’ancien Rwanda. Bruksela: Académie royale des sciences d’outre-mer.

Krzemieńska, B., Třeštik, D. 1964. Služebná organizace v rane středovekych Čechách. Ceskolovensky Časopis Historycky 12 (5): 637–667.

Kučera, M. 1964. K problemu včesnostredovekej služobnickej organizácie na Slovensku. Historicky Časopis 12 (4): 552–571.

Kula,W. 1984. Les mesures et les hommes, Paris, Editions de la Maison des Sciences de l’ Homme (Polish edition – Miary i ludzie (1970). Warsaw: PWN.

Łowmiański, H. 1970. Początki Polski. T. 4. Warszawa: PWN.

Maquet, J.-J. 1954. Le système des relations sociales dans le Rwanda ancien. Tervuren: Musée royal de l’Afrique centrale.

Modzelewski, K. 1964. La division autarchique du travail à l’échelle d’un état: organisation ‘ministériale’ en Pologne médiévale. Annales ESC 6: 1130–1150.

Mworoha, E. 1981. Redevances et prestations dans les domaines royaux du Burundi précolonial. In Le sol, la parole et l’écrit. Mélanges en hommage à Raymond Mauny. T. II (pp. 751–768). Paris: Société francaise d’histoire d’Outre Mer.

Rwabukumba, J., Mudandagizi, V. 1974. Les formes historiques de la dépendance personelle dans l’Etat rwandais. Cahiers d’études africaines 14 (53): 6–25.

Southall, A. 1991. The Segmentary State: From the Imaginary to the Material Means of Production. In Claessen, H., van de Velde, P. (eds.), Early State Economics (pp. 75–96). New Brunswick, London: Transaction Publishers.

Steinhart, E. I. 1978. Ankole: Pastoral Economy. In Claessen, H. J. M., Skalník, P. (eds.), The Early State (pp. 131–150). The Hague: Mouton Publishers.

Trouwborst, A. 1987. From Tribute to Taxation. On the dynamics of the Early State. In Claessen, H. J. M., van de Velde, P. (eds.), Early State Dynamics (pp. 129–137). Leiden: Brill.

Trouwborst, A. 1991. The Political Economy in the Interlacustrine States in East Africa. In Claessen, H. J. M., van de Velde, P. (eds.), Early State Economics (pp. 97–107). New Brunswick, London: Transaction Publishers.

Vansina, J. 1962. L’évolution du royaume Rwanda des origines à 1900. Bruksela: Académie royale des sciences d’Outre-Mer.

Vidal, C. 1974a. Présentation. Cahiers d’études africaines 14 (53): 5.

Vidal, C. 1974b. Economie de la société rwandaise. Cahiers d’etudes afri-caines 14 (53): 52–74.

Zajda, A. 1979. Nazwy staropolskich powinności feudalnych, danin,opłat (do roku 1600). Kraków: Uniwersytet Jagielloński (Resumé – Noms des servitudes féodales et des impôts en vieux polonais (jusqu’à 1600).