Context: Colonial Law and the Abolition of Slavery in Western Tanzania

This article examines the practice of freeing slaves through a sequence of public declarations before witnesses followed by the issuing of a certificate of freedom, which was in use in the Tabora region, western Tanzania, during the first three decades of the twentieth century. Although nowhere codified in colonial law, this process became the default means for slaves to obtain their freedom. Drawing on elements of African, Islamic, and German legal practice, it shows that the law played a role in slave emancipation during these years despite German authorities’ failure to legally abolish slavery, and that slaves themselves played an important role in shaping legal practice.

As one of the signatories to the General Act of the Anti-Slavery Conference of Brussels in 1890, and other international treaties on the suppression of slavery, the German colonial state took legal measures in German East Africa to improve conditions of slaves and to provide ways for them to obtain freedom. Nevertheless, the colonial government was skeptical of the ability of slaves to fend for themselves and concerned about the effects of abolition on the social hierarchies it sought to exploit for governance. The legislation was therefore very gradualist.Footnote ¹ Under these German laws, slaves could obtain freedom through ransom or manumission, official grant of certificates of freedom (Freibriefe), by passing 5 years without working as a slave, absence of heirs on the death of an owner, and by marriage (for slave women). Further, in 1905, the German colonial administration issued a decree that all children born to slaves after 1905 would be free.Footnote ² With these conditions, envisioned German colonial officials, slavery would die a natural death, ceasing to exist in German East Africa after 1930.Footnote ³ These legal measures in what is today mainland Tanzania were considerably more cautious than the measures taken under British influence in the Sultanate of Zanzibar, where slavery was officially abolished, with compensation for owners, between 1897 and 1907.Footnote ⁴

The limited effectiveness of German-era legislation is evident from the fact that in Tabora town, only 3276 persons out of about 40,000 residents, thus less than 10%, had received certificates of freedom by 1912.Footnote ⁵ This is in contrast to contemporary estimates that put the share of slaves in Tabora's population at considerably over 10%. Concomitantly, work on the end of slavery in German East Africa, above all Jan-Georg Deutsch's Emancipation without abolition, has focused on social and economic, rather than legal, mechanisms that undermined slave ownership in the German period. Deutsch characterizes the last decade and a half of German rule as a period of lively contestation between slaves and owners, as the former sought to emancipate themselves by exploiting the shifting power balances between themselves and owners. In particular, the development of a wage labor market in the colony and increasing limitations on the use of violence by slave traders and owners shifted the balance in enslaved people's favor. The outcome was a messy situation where continuing small-scale enslavement coexisted with widespread efforts for both de facto and legal emancipation.Footnote ⁶

After the World War I, British legislation on slavery was more straightforward, though its implementation, as will be seen was more complicated than the letter of the law. The Involuntary Servitude (Abolition) Ordinance, passed in June 1922, declared slave dealing a punishable offence under sections 370, 370A, and 371 of the Indian Penal Code. The sections of the Indian Penal Code invoked here prohibited trafficking of persons, exploitation of trafficked persons, and habitual dealing in slaves. The potential punishments ranged from fines to imprisonment for life. This application of the Indian Penal Code in Tanganyika was part of a broader process of the standardization of law within the British Empire, whereby sections of the Indian Penal Code had applied in Tanganyika since 1920.Footnote ⁷ The ordinance further prohibited detention of a person against one's will in service as a slave or the recognition by courts of the status of slavery, and prohibited removal of property from a person alleged to have been a slave.Footnote ⁸ These anti-slavery laws found some practical application in western Tanzania between 1920 and 1926, as local administrators dealt with small-scale cases of enslavement and slave dealing by local chiefs.Footnote ⁹ Overall, though, sources from the region support Deutsch's contention that by around 1920, ownership of persons as a means to extract their labor power had come to an end in Tanganyika.

The present article seeks to show how legal practice mattered to emancipation despite the gradualist nature of German-era legislation and despite the primacy of political, social, and economic change in undermining the institution of slavery. It examines procedural innovations devised by slaves seeking freedom and their supporters in missions in interaction with colonial officials, which made public declarations before witnesses a means to shed slave status in the Tabora region during the German colonial period and into the British one. In devising these procedures, slaves demonstrated their own creativity in shaping the German and British legal systems. In due course, colonial administrators relied on these pronouncements before witnesses as prerequisite for certification of freedom to slaves who aspired to leave their owners. The declarations served to make emancipation valid to both sides, owner and slave, avoiding tensions which would otherwise arise between them, and between slave owners and missions. By examining the declarations, we show that the legal abolition of slavery was a protracted process involving struggles between slaves and their owners, individual initiatives, and slaves’ own creation of a legal course of action that drew on long-standing indigenous legal practice, in particular the use of oaths. In the process, we rely on archival sources deposited in the archives of Tanzania and the United Kingdom, as well as oral history interviews and mission sources.Footnote ¹⁰

This locally devised legal process is likely to have had important implications for the further trajectories of former slaves. It is clear that even missions which were self-consciously abolitionist institutions often ended up reinforcing hierarchies that kept former slaves in lowly positions.Footnote ¹¹ Moreover, as has long been debated, questions remain as to how to understand the aspirations and aims of former slaves. Rather than a sovereign state of independence, they typically sought reinsertion into social networks, even if in a state of dependence.Footnote ¹² Nevertheless, ex-slaves also had clear ideas of autonomy or, as one contribution to the debate has called it, “freedom from below.”Footnote ¹³ Importantly for the present context, the manner in which a person shed the slave status often shaped the subsequent status as a freed person. For instance, imposed emancipation according to colonial law was typically less socially valid than manumission according to long-current Islamic norms. The negotiated character of public declarations discussed here, with the explicit involvement of owners, is likely to have resulted in a more socially valid form of emancipation than by the fiat of colonial rulers. By the same token, though, it was a qualified emancipation that did not remove all traces of subservience. We return to this issue later.

Milahi: Oaths, Society, and Authority in Nineteenth- and Twentieth-Century Western Tanzania

Statements of assurance or oaths (milahi or kuapiza) abounded in nineteenth- and twentieth-century western Tanzania. Present-day informants characterize them as an important element of the social world, shaping social relations and the practice of authority, because they provided assurance that what was being said before authority figures was true.Footnote ¹⁴ These assertions chime with existing studies of pre-colonial legal and political practice in central Africa, in particular Martin Chanock's work. There is an important difference in emphasis, though: while Chanock is very clear about the capacity of oaths, especially if administered alongside a poison, to terrorize and discipline dependents, Nyamwezi informants focused on their mediating and legitimating capacities. The contrast suggests a degree of nostalgia among Nyamwezi informants, but also serves as a reminder of the political and fungible character of these oaths.Footnote ¹⁵

Circumstances including the fallout from military invasion, epidemics, family squabbles, complications created by men's participation in porterage, and other turbulence prompted the use of oaths of different kinds and in different social fora. Informants assert that public declarations in western Tanzania were used at three levels: family, clan, and territorial/chiefdom-wide.Footnote ¹⁶ Among Nyamwezi families, tensions which could not be solved amicably prompted elders (vahanya va vipande) to take the lead in administering oaths to the irreconcilable parties involved. Clans (vahanya va gunguli), too, brought together heads of families to preside over statements of assurance or oaths on clan membership. At the territorial level (muhanya wa chalo), the chief and his subordinates (wanangwa) directed oaths on issues that had bearing on their communities, such as land disputes, but also suspicions of human involvement in the causation of drought or epidemics.Footnote ¹⁷ An oath recorded from oral sources by Gilbert C. K. Gwassa and Joseph F. Mbwiliza in Buha offers a snippet of how oaths could be used to address land disputes in western Tanzania.

This is a description from the 1970s of what was then said to have been a nineteenth-century practice, and aspects of it, such as the emphasis on a monotheist deity, may be anachronisms. Nevertheless, the account shows these oaths to have been both a public and a ceremonial practice, involving regalia—that is, garb designed to be seen—ritual objects and ritualized speech. In Unyamwezi, informants assert that such oaths were a crucial form of conflict mediation, where, much like in Buha, the oathing parties asserted that “ufi ulamba ugulugulu lyangombe limpute” (if I am lying let God take away my life). This invocation of divine retribution gave the oathing parties’ assertions a great deal of weight.Footnote ¹⁹ Present-day informants are clear that communities in western Tanzania had a strong conviction of the effectiveness of these oaths and took the sanctions associated with making a false declaration very seriously.

These commitments were related to wider beliefs concerning the power of words and of particular speakers. In the family, for instance, the Nyamwezi people believed that “shangazi ana mate mabaya” (a paternal aunt has bad words), meaning that if a shangazi sought to cause harm to relatives with her utterances, she would succeed. That belief suggests the authority of elderly women (in particular aunts) in officiating oaths among family members. In this context, the intimacy between family members, especially the shangazi's close daily involvement in the life of her relatives, reinforces the sense of danger attendant on being untruthful in an oath.Footnote ²⁰ Indeed, oaths, in particular so-called reconciliatory oaths (viapo vya upatanishi) could be used to restore troubled personal relationships.

Oaths of allegiance to chiefs in times of invasion, meanwhile, had a more public and political character. Before 1840, ritual chieftainship dominated authority in Unyamwezi, but after 1840, as long-distance caravan porterage took hold in Unyamwezi, authority was increasingly based on the accumulation of wealth through porterage and farming. Hence, wealthy individuals (vandeva, sing. mundeva) and chiefs who controlled trade routes began to administer an oath of allegiance (yamini) to porters before they would embark on the grand march (safari kuu) to the coast. The power of this oath corresponded to social hierarchies. The Nyamwezi porters believed that territorial chiefs had the power to pronounce more powerful and consequential oaths. So, they took seriously oaths administered at the chief's palace before embarking on the grand march to the coast and believed that breaching would cause serious problems.Footnote ²¹

Oaths of loyalty also prevailed among buswezi and migawo or migabo secret societies. These healing societies in western Tanzania were related to a geographically widespread ritual complex, typically referred to as Cwezi-kubandwa and most closely associated with the Great Lakes region.Footnote ²² The dynamics of these societies have been interpreted in very varied ways, from everyday form of resistance by women to an element of statecraft, again highlighting the fungibility of their practices, oathing included. Nyamwezi informants characterized them as a concern of male elders rather than chiefs, and emphasized their esoteric, numinous character. Details of the ritual knowledge and power, initiation, and training of members of these secret societies were only accessible to initiated members who had taken the society's oaths.Footnote ²³ Their secrets about healing remained undisclosed to strangers, whose presence at buswezi rituals was considered detrimental.Footnote ²⁴

Another form of oathing, called kupanwa, aimed at elevating personal loyalties to the status of a blood relationship. Oaths of this category involved exchange of blood by two long-standing friends. Exchanging blood (through incision made on the body) meant that they had reinforced “blood relationship” (udugu wa lusale or budugu wa nsale). Further, sharing blood meant that the two individuals became each other's kin and they could claim inheritance as bonafide members of the other's kin group. It also meant that they had to respect rules of exogamy and could not marry within each other's kin group.Footnote ²⁵ In some Nyamwezi societies, long-lasting friendships of two individuals were cemented with the marriage of their sons and daughters. After marriage, the two elderly friends took an oath that prohibited further marriage in the two families because they had transcended the boundaries of mere friends to now become relatives.Footnote ²⁶ Yet, some Nyamwezi communities administered oaths in order to frighten people (kukanga) to refrain from actions that seemed detrimental to the well-being of the family, clan, and the society at large. Threatening oaths (viapo vya kutishia) often aimed to inquire into the truth of disruptive events including theft and witchcraft. Suspects held an object or a wooden piece from the tree called limwavi—believed to cause harm to people—as they took oaths so that that the crowd “would find the truth on the matter” (wangepataa ukweli wa jambo).Footnote ²⁷

There appears to have been an evolution in the administration of oaths in the course of the nineteenth century. Initially, the ritual was essentially verbal. Statements like “ufi ulamba ugulugulu lyangombe limpute” (If I am lying let God take away my life) were used to assure the participants’ truthfulness.Footnote ²⁸ By the second half of the nineteenth century, in the context of increasing exchange and mobility associated with long distance trade, it became increasingly common to reinforce the use of words with that of numinous objects, said to cause harm to anyone untruthful in their oath. In particular, the parties to an oath increasingly held branches of a tree called limwavi. People in Unyamwezi believed the tree to have “immediate effects” (matokeo ya haraka) on people who had broken or been untruthful in an oath. The fear shown by some individuals of holding it as they swore was “a proof that it is working” (uthibitisho kuwa linafanya kazi).Footnote ²⁹ Whatever the effectiveness of the tree, it is evident that it focused attention on how much was at stake in an oath, that it was believed to have “power to heal and harm,” and was a tool for the enforcement of social norms.Footnote ³⁰

The witnesses to the oaths, moreover, were extremely important. The main criterion for choosing witnesses was seniority in family, clan, and society. They could either be individual or collective depending on the nature and character of an oath in question. Within families and clans, respected elders were preferred as witnesses. In chiefdoms, people who lived in the chief's courtyard (banikuru/vanikuru, sing. mnikulu) acted as witnesses. They were duty-bound to listen, advise, and ensure that members abided by the oath.Footnote ³¹ Witnesses served a twofold purpose: they confirmed the truth of the claims sworn to, and they managed tensions arising from breaches of oaths or situations where competing oathing parties failed to find a compromise.Footnote ³² Within families, failure to resolve breach of oaths or conflicting oaths was believed to lead to persistent misfortune and the use of curses. Concomitantly, oaths functioned as tools of social control as well as community building and mediation. The widely held belief in the power of oaths meant that they could not easily be defied (viapo vilikuwa havifanyiwi mzaha).Footnote ³³ In the context of the decline of slavery, these beliefs, and particularly the role of witnesses, were redeployed to assert claims to emancipation.

Public Declarations, Witnesses, and Certification of Freedom, 1907–1920s

At the turn of the twentieth century, the culture of oaths took on a new dimension, as slaves sought freedom to join mission communities and actual or fictitious relatives in villages surrounding Tabora. Increasingly, people seeking to emerge from slave status performed public declarations, or statements of assurance before witnesses, in order to obtain certification of freedom from the German colonial authorities. These were public performances that brought together the slave seeking freedom, their owner or a representative of the owner, representatives of a mission station sponsoring the slave's bid for freedom, representatives of the colonial authorities, and witnesses chosen by the slave. Before this audience, as well as interested bystanders, the slave would declare their determination to leave their owner. The owner would acknowledge it, and raise objections, set conditions or demand compensation from the mission supporting the slave. Once a compromise was agreed, the slave would obtain their Freibriefe.

This practice developed in the context of German-era legislation and legal practice on slavery. German attitudes to emancipation were even more gradualist and cautious than those of the British in the nearby Zanzibar protectorate and Kenya. Denying outright emancipation “at the stroke of a pen” and avoiding reparations, legislation aimed instead at letting slavery die out over time by emancipating slaves’ offspring. For those currently enslaved, the law upheld their owners’ rights over their labor power, and in the case of women slaves, their sexuality, in principle, while setting out broad ground rules for dealing with slave flight, maltreatment, and demands for manumission. But, as Deutsch's work has shown, these rules offered limited practical guidance to local officials who were faced with complicated, conflicting claims about different owners’ rights to enslaved people, the history of their enslavement, sales, and marriages, their relationships with others and the status of their dependents.Footnote ³⁴ The colonial government dealt with these complexities by leaving a great deal of leeway to officials “on the ground,” who improvised individual solutions to the slavery-related disputes that came before them.

In Tabora, these solutions came to include public declarations before witnesses that drew on the long-standing practice of oathing. Nyamwezi ideas and practices surrounding oathing, which were embedded in the society and culture, interacted well with notions of procedural correctness and decorum upheld by German administrators and Catholic missionaries in the region. It is therefore no wonder to find that the German administrators and missionaries adopted them as prerequisite for emancipation. Statements of assurance before witnesses in Tabora began, according to available records, to be used in 1907 by slaves who wished to move into mission communities. The statements of assurance served to address increasing concern among slave owners over fugitive slaves and others who went to the mission without their consent. They formed an initial stage for certification of freedom (provision of Freibriefe) from the colonial administration. Slaves’ declarations before witnesses, performed in front of officials of the German colonial state and missionaries, helped manage tensions between slaves and their owners, between the slave owners and the missions, and between missions and the colonial state, before the three parties (slave owners, slaves, and missions) moved on to negotiate the terms of an individual slave's freedom.Footnote ³⁵

Lists of slaves redeemed by the Missionaries of Africa (commonly called the White Fathers), kept under the title of rachats d'esclaves, routinely listed the names of the witnesses called upon by the slaves to facilitate their redemption. They give a vivid picture of the diversity of the slave population involved in these negotiations, and their witnesses. For instance, the list from 1908 to 1909 contain persons originating from various parts of the East African interior, including Rwanda, Burundi, Uganda, Karagwe, and eastern Congo. They were both male and female and aged between 15 and 60 years.Footnote ³⁶ Their witnesses were typically former slaves living in the mission community. The Christian names of some witnesses indicate that they had already become part of the Christian communities in the missions. Some witnesses, though, maintained their non-Christian names, indicating that both Christians and non-Christians could bear testimony to slaves’ determination to shed slave status. Further, some witnesses were relatives of slaves living near Christian missions who could also testify to slaves’ aspirations to get into either the mission community or to live with their relatives as free persons.Footnote ³⁷ Records of the Missionary Sisters of Our Lady of Africa (often called the White Sisters) further add to our knowledge of the people whose emancipation was sponsored by the mission. Between 1909 and 1910 alone, at least forty-six adult women lived at the White Sisters’ convent in Tabora, while the number of girls and children in the orphanage reached fifty-one.Footnote ³⁸ The list of their names of former slave women and orphan girls who sought refuge to the White Sisters’ convent at Tabora includes Nyamwezi, Christian, and Swahili terms such as “Nyamizi,” “Kalundi[e],” and “Kalekwa,” “Sumuni,” “Vumilia,” and “Sesilia.” While the document does not tell us how they became free, it is reasonable to suggest that they worked for the White Sisters to earn cash to buy their freedom, since it was a policy of the White Sisters in Tabora to provide work enabling slaves to earn money with which to redeem themselves.Footnote ³⁹ To be clear, declaration before witnesses followed by the issuing of a Freibriefe was not the only option for slaves seeking freedom. As Deutsch discusses, simply absconding, particularly by joining the trek to growing wage-labor plantations, was also an option. Still the evidence suggests that a great many different people availed themselves of the option of declaration and certification, perhaps particularly if they didn't want to move great distances from where their former owner was located.

There are some further implications of the evolution of these declarations that are worth drawing out. First, that such public declarations “worked” as a means to establish free status suggests that owners’ control over slaves was already fairly weak. Had it been solid, owners would have had little reason to acquiesce into their slaves’ assertion of their desire to become free. As it was, participating in these declarations offered owners a means to influence the terms of the emancipation of slaves who otherwise might have sought their freedom without consultation. They made the loss of slaves who had demanded their freedom more akin to acts of manumission, which under Islamic legal custom were initiated by the owners. Moreover, the declarations chimed with a broader effort by German colonial authorities to perform their functions in public, so as to be seen to be in charge. Legal proceedings lent themselves well to projecting an image of the colonial power as guarantor of order, and German officials took the opportunity.Footnote ⁴⁰

Next, the role of the Freibriefe as the tangible result of these declarations deserves further consideration. Although these documents had no precedents in established oathing practice, they quickly acquired considerable importance. In principle, the certificates were granted to the freed slave him- or herself, but colonial administrators often handed the documents to missionaries who had sponsored bids for freedom, who then kept them in their stations. A total of 130 such certificates of freedom (Freibriefe) from various mission stations were deposited in the archive of the archdiocese of Tabora. In this way, the certificates mediated a new unequal relationship, between officially freed slaves and the missionaries who had helped them pay for their certificates. The cost of manumission, thus of the certificates, was far from trivial, as Deutsch has shown. At the same time, if the certificate was given to its owner, it served as something akin to a passport: it enabled its bearer to move between locations and employers without questions as to their right to do so.Footnote ⁴¹

It is worth elaborating briefly on the financial aspect of obtaining a Freibriefe, since this was a big part of what established the claims of missionaries on the certificates and the loyalty of ex-slaves whose manumission they sponsored. In 1900, the military officers in Tabora recognized the sum of seven rupees or two head of cattle as equivalent to the price for which slaves could buy freedom from their owners. The price of freedom increased after 1907 to twenty or thirty rupees, and at the time of the World War I in 1914, it had increased to forty rupees.Footnote ⁴² As with the declarations, so the financial arrangements for obtaining Freibriefe were a matter of case-by-case negotiation. For instance, in 1905, the slave Luangalla and his wife Kashindye moved into the mission community at Ushirombo, but their owner, Mlasso, complained to the mission that they had gone without his consent. Luangalla paid 30 rupees for himself and 20 rupees for Kashindye to redeem themselves from slavery.Footnote ⁴³ A woman called Maria Wantiho also requested that her freedom be bought because her owner Mtagwa was unwilling to let her move into the mission community. Maria Wantiho paid 15 rupees out of the required 40 rupees to buy her freedom. The White Sisters at Tabora paid the remaining amount.Footnote ⁴⁴ These individual settlements show the parties to them employing notions of personal commitment ritually confirmed, reminiscent of those at play in some forms of oathing, such as blood brotherhood and porters’ oaths of loyalty (Table 1 and Figure 1).

Table 1. Slaves, Owners, and Witnesses to Freedom in Tabora, 1908–1909

Source: Compiled from AAT 322.399 Rachats d'esclaves, 1908–1909 (December 27, 1919).

Figure 1. Mission Community near the White Sisters Convent in Ushirombo, c. 1900. Courtesy of the Archives of the Archdiocese of Tabora (with permission from the Catholic Archdiocese of Tabora).

In this sense, third, the evolution of these public declarations was part of the process of improvisation that Deutsch traces so vividly in the legal practice of the German period. As it turned out, though, the practice also sat well, in a different way, in British imperial legal practice, which provided precedents for it in the shape of the Indian Evidence Act and the Indian Oaths Act, adopted by the new Tanganyikan government in 1920 (with retroactive effect from April 1919). The adoption both provided consistency for British territories around the Indian Ocean, and put the continuation of the established practice on a legal footing. In this way, oaths of affirmation before witnesses, which had taken roots among slaves, became embedded in the British legal system in Tanganyika.Footnote ⁴⁵ The institution was elaborated further as the Indian Evidence Act and Oaths Act stated the need to establish a commission of inquiry in Tanganyika territory, which were duty bound to administer oaths and affirmations. Further, the commission put in place with few amendments those sections of the Indian Penal Code defining the duties of summoned witnesses for judicial proceedings, indemnity to witnesses, and legal consequences on giving or fabricating false evidence.Footnote ⁴⁶ Since this Indian imperial legislation recognized oaths as a legitimate part of the judicial process, oathing became established as part of legal procedures in Tanganyika.

Public Declarations and Certificates of Freedom in the Aftermath of Official Abolition

The emphasis on publicly declaring a change in legal status that was evident in individual ex-slaves’ declarations is traceable also in events around the Involuntary Servitude (Abolition) Ordinance, which officially abolished slavery in Tanganyika from June 16, 1922. The Ordinance, as mentioned, declared slave dealing a punishable offence, prohibited detention of any person against his or her will, and pronounced slave dealing a punishable offence. It also ruled out the recognition of property in persons or services as a slave by civil or criminal courts in the territory. Alleged slaves could not be dispossessed of properties acquired by their own efforts or by way of inheritance—a provision aiming at the practice of owners inheriting their slaves’ property at their death—and a person convicted of detaining a slave against his/her will could be fined up to fifty pounds or imprisonment for 2 years or both.Footnote ⁴⁷

The senior commissioner of the western province announced the passage of the ordinance in Tabora town before a crowd, stating that from the day of the pronouncement of the decree there could no more be slaves in the area. The overseer of the Catholic missionaries in the area, the vicar apostolic of Unyanyembe Henri Léonard, translated the decree into the French language for his Francophone staff. He called upon his missionaries to communicate the news to all the missions, to familiarize themselves with the law and to speak openly about it, and to make it known in all missions and village outstations.Footnote ⁴⁸ The provincial and district commissioners, in collaboration with native authorities, translated the law into the Kiswahili language as “Ondoleo la utumwa pasipo kutaka” (forceful abolition of slavery) to make it comprehensible to the people who were not conversant in English. Some government officials held public meetings that explained the law and the implications for those who still owned slaves in their households.Footnote ⁴⁹

Ironically, though, the efforts to publish the law on the abolition of slavery fed into a rush on individual certificates of freedom. That demand for these certificates peaked at a time when summary abolition was supposed to have made them obsolete shows the continuing influence of place-specific social and legal norms. As tangible evidence of individual, agreed acts of emancipation and manumission, these certificates provided greater certitude of a change of status than the summary law of abolition. In response to demand, the British colonial officials at Tabora headquarters (boma) had maintained their German predecessors’ practice of issuing certificates of freedom since 1918. Eventually, though, the increasing number of ex-slaves demanding certificates became a challenge, as a shortage of the pre-printed certificates arose. Accordingly, in August 1922, 1 week after making an announcement on the shortage of certificates and about 2 months after summary abolition, the British colonial state discontinued the practice.Footnote ⁵⁰

This action was in keeping with broader official assertions that the effort to end slavery in the territory was completed, but the persistence of public assertions of claims to freedom by ex-slaves into 1923 shows that things were less clear-cut. The British authorities’ reporting on slavery in Tanganyika Territory for the early 1920s claimed confidently that, since British attitudes to slavery were so well known, owners had resigned themselves to the ending of the institution with their arrival.Footnote ⁵¹ The 1922 abolition ordinance had been completely effective: “slavery no longer exists in any form,” as a 1925 document stated.Footnote ⁵² And yet, in 1922 and 1923, the White Fathers conducted a protracted correspondence with British authorities in which the practice of declaration before witnesses was invoked as a means to legitimize former slaves leaving their former owners. This correspondence casts some light on the states of dependence that official correspondence covered with formulaic assertions that there were “not a few ex-slaves who of their own choice continued to remain with their former masters,” but that choice was entirely voluntary.Footnote ⁵³ Moreover, this correspondence also shows the politics that continued around the certification of freedom, about which more later.

Thus in August 1922, the British political officer of Tabora ordered Rev. F. van Aken, Superior of the Catholic Mission at Tabora, to “withhold the certificate” of Paulo, a teacher who was then working at Isenga mission school. The certificate to be withheld was that of freedom; by withholding it, the missionaries were instructed to limit Paulo's freedom of movement. The reason for this request was that one Maganga bin Mzigawe (a Sukuma or Nyamwezi man judging by this name), had accused Paulo of inducing his slaves to leave for the mission. In response, the Father Superior negotiated with Maganga bin Mzigawe a compensation payment to be made by the mission for his former slaves. But he also elaborated, in his return letter to the political officer, that slaves leaving their owners by means of a public declaration of their intention to do so was an established practice, and that the people who had left Maganga had acted of their own accord rather than compelled by Paulo. This judgment, that Maganga's former slaves had left of their own accord, was confirmed by the Senior Commissioner, who van Aken had appealed to. Nevertheless, the Senior Commissioner had demanded that Paulo be taken to the political office for interrogation before he decided in his favor.Footnote ⁵⁴

Five months later, in January 1923, van Aken again had reason to write to the Senior Commissioner about a group of former slaves who had gone to the mission and declared before witnesses that they desired to be free. In his letter to the Senior Commissioner, Fr. Aken described them as “a number of slaves (who formerly belonged to some Watussi) … who declared to myself in the presence of a crowd of natives that they would no longer be slaves and refused to return to their masters…”Footnote ⁵⁵ Again, van Aken pointed out that such public declarations had been in use for some time as part of a process leading to the certification of freedom. Even though the issuing of Freibriefe was by then discontinued, the act of declaration evidently still carried a certain weight as a means to gain recognition for an act of self-emancipation. At the same time, van Aken cited these declarations to defend the mission against allegations of encouraging slave flight (notwithstanding the fact that since official abolition, such flight was technically entirely legal).

These exchanges concerning Paulo and declarations of freedom provide a number of insights into the socio-legal process of abolition. First of all, it shows a situation very much at variance with the letter of the abolition ordinance, which had made owning slave a punishable offence, and with the assertions that colonial officials made in their communications with the League of Nations concerning slave owners’ quiet acceptance of the end of slavery. The political officer in Tabora contacted the mission on behalf of a slave owner, to query the actions of a former slave and client of the mission, in encouraging persons referred to simply as “slaves” (not former slaves, dependents, clients, or any such) to leave their owner's household for the mission. Owners, then, still had the ear of the administration, and administrators did not at this point support unlimited free movement by (former) slaves. van Aken's actions in resolving the situation suggest that he understood the (former) owner's complaint to be a demand for compensation. In turn, the senior commissioner effectively endorsed the compensation deal by exonerating Paulo. Notwithstanding the publicity accorded the summary abolition of slave status in Tanganyika Territory, then, in practice effective emancipation still required a transaction between the former owners, the administration and the mission.

Next, the events show that in the immediate aftermath of the announcement of the abolition ordinance, the Freibriefe, although never part of the British strategy for slave emancipation and at odds with summary abolition, retained a certain weight despite their lack of official legal standing. Otherwise, the political officer's request to withhold Paulo's certificate would have been meaningless. Conversely, by ending the issuing of Freibriefe, the administration removed what had by then become an established strategy to obtain an owner's acquiescence in the departure of a slave and make the ending of their slave status socially valid. The official rationale for this is expressed in the statements made to the League of Nations: slave status was no longer recognized, therefore confirmation that any particular individual was not a slave was not necessary. Yet both the evidence discussed here and much work on the aftermath of slavery elsewhere in Africa suggests that official abolition “at the stroke of a pen” was less effective in rehabilitating an ex-slave's social status than the individual, negotiated ending of slave status with the former owner's acquiescence.Footnote ⁵⁶ In consequence, the ending of this procedure was a significant change to the options available to ex-slaves. It is likely to have made it more difficult for them to assert an improved social status in the place where they had been enslaved, and thus to have increased the motivation to migrate for those who did not want to remain at a nearby mission station. Such an outcome is suggested also by the report of van Aken concerning the ex-slaves who asserted their freedom by declaration in early 1923 (thus at a time when Freibriefe were no longer issued); “A few indeed went off, but did not return to their masters, but to their wadugu (relatives) elsewhere, the remaining unanimously declaring that they wished to stay and to dwell close to the Isenga Mission.”Footnote ⁵⁷

This also implies that ex-slaves who were freed without certificates after August 1922 did not necessarily have much better options than those who had had their certificates financed and retained by the mission. The regimented character of life in the missions and the difficulty of characterizing it as “freedom” have been noted. Sources from Tabora show that, in addition to mandatory attendance of church services, residents in Christian villages performed different activities that bound them together. Girls in the orphanages at Ushirombo, Tabora, and Ndala performed work in the fields under the supervision of the White Sisters. They also learned domestic skills such as sewing, washing, ironing, mending, and cooking, while men worked as carpenters and manufacturers of bricks.Footnote ⁵⁸ Missionaries also owned large tracts of land near mission stations where they employed residents of nearby village communities as part of the “mission workforce” to grow a variety of crops for consumption and partly for sale.Footnote ⁵⁹ Nevertheless, for many people, mission communities became refuges and homes, and for some, the schooling they provided enabled considerable social mobility.Footnote ⁶⁰

Despite the salience of public declarations in the correspondence between van Aken and the administration in 1922–1923, the practice appears to have faded in subsequent years. Mentions of slavery in the Tabora archives later in the 1920s mostly concern isolated cases of slave trafficking which resulted in prosecution of the perpetrators.Footnote ⁶¹ How the victims in those cases asserted their free status and whether public declarations still played a role in the process is not recorded in these reports. That owners now were punished rather than compensated, though, makes clear that procedures had moved on from the practice of the German and early British period. Oathing, meanwhile persisted as a part of Islamic legal procedures, and was admissible as such in the Liwali's court of Tabora.Footnote ⁶² There is every reason to believe that it also continued in buswezi societies. As a means to establish political and professional loyalties, though, the way it had been used by caravan leaders and chiefs, it lost function in an age of appointed colonial chiefs. Elsewhere, new forms of public declaration arose, such as the confessions of sin by “wordy women” in the Christian revival movement that so concerned elders in the 1930s.Footnote ⁶³ The peculiar conjunction of German bureaucratic notions of certification, Islamicate notions of manumission by owner's consent, and the indigenous African practice of public declaration before witnesses that had come together in slaves’ declarations between ca. 1907 and 1923 was clearly over.

This, of course, is not to say that the slave past was simply over for the formerly enslaved. Rather, the problem of (former) slavery moved on to negotiations over resource access, marriageability, dependency, and autonomy of the kind just mentioned in connection with former slaves who remained at the missions.Footnote ⁶⁴ Some former slaves, for instance, re-emerged as commercial beer brewers in Tabora town, while others migrated to Ugogo, depopulated by famine during the World War I. Owners, too, had to adapt, reinventing themselves as intermediate crop-traders and small-scale commercial farmers, whose encounters with the law most often had to do with indebtedness.Footnote ⁶⁵ There again, social and legal struggles were intertwined, but in a very different way.