Criminal Justice in Precolonial Tswana Societies

Published as: Bruce S. Bennett & Maitseo M. M. Bolaane, “Criminal Justice in Precolonial Tswana Societies”, International Journal of African Historical Studies Vol. 55, No. 1 (2022) pp. 69–88.

Botswana is notable for its adherence to flogging and the death penalty in its criminal justice system. These are typically claimed to be part of immemorial tradition, with Botswana’s customary courts being understood as simply a modernized form of ancient practice. However, evidence from early written sources suggests that there was a much more complex and inconsistent situation, in which compensation and revenge were more common features, and murder was often regarded as a private matter. This suggests that effective vertical justice by chiefs may be a relatively recent development of the late precolonial period, representing more powerful rulers and larger political units, and later further accentuated by colonial backing for chiefs who could maintain order. Some support for this may be found in anthropological evidence about the Kgalagari villages in the 1960s, which may be seen as the nearest observed modern analogue.

The present-day Republic of Botswana^¹ is notable, in the Southern African region, for the relatively prominent part played by traditional institutions in official life. British colonial rule was run on a shoestring, and relied on indirect rule. Subsequently, the Botswana Democratic Party, which has won every election since independence, has maintained (under stricter control) elements of traditional rule. Chiefs (dikgosi, singular kgosi) have lost much of their old direct power but remain important and influential figures, presiding over the kgotla assembly, which both discusses public affairs and administers justice. As in a number of other African states, a dual legal system exists, involving the continuation of customary law as a recognized form of law within the overall legal system. Most minor cases are dealt with under customary law, either in kgotla or in urban Customary Courts.

The Customary Courts make frequent use of flogging as a punishment for criminal offenses, and it can be used by other courts as well. The practice, which seems to be fairly widely supported (at any rate it is not the subject of much public controversy) is protected from constitutional challenge by special provisions.^² It is generally assumed that this “traditional” punishment has been the norm throughout Tswana history.^³ We will argue, however, that—while it is true flogging does have a long history—its present frequency is largely the result of colonial rule (and the late precolonial period in some areas) and was not necessarily typical of premodern Tswana society. Tswana society did have concepts of objective standards which can reasonably be translated as “law,” and proverbs such as “Molao sefofu, obile otle oje mong wa one” (“The law is blind, it eats even its owner” illustrate the idea of a law existing apart from any particular case.^⁴ However, premodern Tswana polities often lacked the power for enforcement of such standards.

There is a problem in distinguishing normative statements from what actually happened in a society.^⁵ Codification of customary law did not always address this.^⁶ It should be noted that the major anthropologist of twentieth-century Botswana, Isaac Schapera, explicitly recognized that Tswana society was not timeless, and wrote in detail about colonial influences and chiefs as legal innovators,^⁷ but his focus was on more recent changes than those we postulate. Since tradition is in question, written sources are essential. There are some by contemporary Tswana , but most are from early explorers and missionaries. How accurate were their reports? The answer no doubt varies, but we have concentrated on reports which are either from long-term residents (such as missionaries who might spend decades in a Tswana society), or which can be analyzed in terms of the events reported rather than the writer’s interpretation. We have also included accounts of the Sotho, who are closely related and indeed not clearly distinguished at that time. It should be noted that a Tswana morafe (polity, plural merafe) might include members who were not ethnically Tswana, especially after the mfecane/difaqane (c. 1815–40).

The major offenses in precolonial Tswana society, which were potentially capital crimes, are usually stated to have been murder, witchcraft, and rebellion. However, of these only murder really equates to the modern conception of an individual crime. Rebellion was in reality at least as much a matter of politics as of law, and the procedures described by Schapera seem to transpose political violence into an imagined legal process.^⁸ Rulers sometimes had men who represented a potential or actual threat assassinated in secret. Similarly, witchcraft was understood to threaten the entire community at a fundamental level, and the kgosi might have lost control of the situation if he failed to pass sentence.^⁹ To maintain his authority he had little option but to ride the wave.

There is no doubt that the death penalty was sometimes imposed for murder,^¹⁰ but it is disputable how often.^¹¹ William Burchell, who traveled in Southern Africa in 1815, stated that he was told by BaTlhaping that murder “rarely” led to punishment by the kgosi, though there was a risk of private revenge. (Burchell lacked first-hand evidence and hoped it was not true).^¹² Murder of women by their husbands was sometimes regarded as a private matter. Moffat gives an eyewitness account of those in authority laughing at his concern over the murder of a woman by her husband,^¹³ and Elizabeth Lees Price recorded in her diary that Kgosi Sekgoma I of the Bamangwato^¹⁴ declined to act in such a case on the grounds that the wife was the husband’s property.^¹⁵ It seems unlikely that these accounts, which are consistent, are fundamentally inaccurate. For comparison, the early missionary John Campbell stated that among the Nguni “murder is usually overlooked” except when there were aggravating circumstances, and the usual punishment, if any, was a fine.¹⁶

Moffat noted a case of execution for murder. However, this was a special case. The victim was a missionary, J.M. Kok,¹⁷ and the kgosi was anxious to demonstrate his innocence, sending messengers to the Griqua Town mission station to explain what had happened. It is worth noting that he initially asked the widow to kill the murderers herself using the same type of weapon.^¹⁸ The practice of the kgosi enabling such retaliation is recorded in tradition: it is debatable whether it constitutes a legal punishment or facilitating revenge.^¹⁹

A very interesting letter by a young Tswana in 1883 about witchcraft and murder should be noted. The anonymous writer, writing to Mahoko A Becwana, was impressed with what he believed were the better legal practices of European justice: firstly, in having rejected witchcraft as a crime while prioritizing murder, and secondly, in requiring investigation and credible witnesses (and indeed in penalizing false accusation). “Some of us young Batswana, we are already awake; we are awoken by the trials of you Europeans that we see in your courts.”^²⁰

Murder is the one that I hear you give much legal attention. And if a person who has killed someone with a deadly thing, even if it is poison, I hear that you won’t kill him if the murder was committed without being investigated by credible witnesses.^²¹

In addition, those who made false accusations were sometimes put on trial. But among the Batswana, if there was death, or infertility, or cattle died, there would be claims of witchcraft.

And there will be no question asked or even an investigation. The kgosi will prepare the people who are going to kill the witch. But he will not be simply killed as he is passing:^²² he will be beaten and crushed and his eyes plucked out so that he might tell the names of others.^²³

Those implicated would be similar tortured for names. “Yet still there will be no question or investigation.”^²⁴ The writer condemned chiefs for their actions. In particular, he noted that, last month, Kgosi Sechele of the BaKwena had killed five people for witchcraft on the accusation of a child.

Sechele himself replied to this. “These people, I would not have killed them only on the words of the child if they were not [already] generally known [as witches].”^²⁵ This rather confirms the original writer’s complaint that specific investigation was lacking.^²⁶

It should be noted that under colonial rule witchcraft did not disappear as a matter of law. Cases continued to be tried in kgotla until 1927, when (following a major case) the colonial government made both accusations of witchcraft and attempts to practice it offenses.^²⁷ In 1937 the district commissioner at Serowe (the Bamangwato capital) tried a case of attempted witchcraft,^²⁸ possibly becoming the last ever British judge to do so.

It is also apparent that there was an effective death penalty for some ritual offenses, most notably for revealing the secrets of initiation. The boys’ initiation ceremonies (bogwera), centering on circumcision, were of great importance in forming adult male identity. Andrew Smith, who visited BaTlhaping and BaRolong in 1835, wrote:

The law is very strong on the subject of the circumcision if those who are circumcised speak of it to the persons that are not circumcised. The chief will immediately cause such person to be put to death. Under these circumstances it is very difficult to get information on the subject, and I was forced to promise that I never would divulge the name of the person from whom I got the information before I could hear anything concerning it.^²⁹

The group initiated together formed a mophato (plural mephato), traditionally translated “regiment”. Mephato served a variety of purposes, both in peace and war, and when Christian dikgosi abolished initiation, they continued to create mephato by simply summoning an age group.

Among the BaKwena cruel punishments were sometimes used. Sechele, the Kwena kgosi, is reported by Elizabeth Lees Price to have used “Matabele”^³⁰ refugees as executioners and torturers in these cases. It is unclear how far this was practiced for ordinary crime. In one case, a thief was tortured to death, but the case was atypical as he had stolen gunpowder at a time of threat of war, when it was a vital resource: “His arms & hands were pounded & his eyes were put out before he was released fr. his agonies...”^³¹ This was probably considered as treason rather than ordinary theft, and treason has often attracted extreme punishments in premodern societies (notably Reformation Europe). However, the observations about regular executioners suggests that such acts were not isolated exceptions.

In Schapera’s twentieth-century account of “traditional” law, it is simply stated that murder was punished by execution.^³² This discrepancy from contemporary observation is significant evidence of problems with the way tradition has been used. The importance of this historical issue can be seen from the fact that Botswana Court of Appeal stated in a 2012 ruling that “Available authorities ... indicate that the death penalty has been imposed in Botswana for murder since time immemorial,” citing Schapera’s statement as evidence.^³³ It is unclear whether “has been imposed for murder” should be understood as implying possibility or typicality. Talking of a “death penalty” existing from time immemorial also raises a question of categories. If a murder is liable to lead to revenge by an aggrieved party, but not to action by the ruler, this may have the effect that a murderer is killed, but does it constitute a “death penalty” in the same sense as in a vertical-justice system? Also, while the terminology of a “death penalty” suggests a sentence after a kgotla trial, it is noteworthy how often accounts of execution are in terms of summary orders from the kgosi.

The lower-level sentence of flogging is of particular interest. Schapera describes flogging as a common traditional punishment, and of course there is no doubt that it has been a frequent sentence in criminal proceedings during and after the colonial period.^³⁴ However, early sources seem to describe a rather different situation. Burchell commented that although sentences of flogging were passed, they were almost always commuted to fines, except in the case of “atrocious crime”.^³⁵ He connected this with the socio-economic realities of Tswana society, with richer and poorer classes. “According to this scheme of society, the chief will always be the richest man; for once arrived at supreme authority, he holds within his own hands the power of obtaining property”: that is, a kgosi could use his powers to increase his wealth, making fines an attractive option.^³⁶Tswana tradition holds that killing or injuring another person was an affront to the kgosi, and this seems to have been connected with his right to levy fines for such offenses.³⁷ Similarly, Eugène Casalis, a French missionary with Moshoeshoe’s Basotho from the 1830s, wrote a detailed account of criminal cases which not only noted that fines were the normal punishment but did not mention flogging at all.^³⁸

Tradition is clear that poor men, who had neither influence, nor significant property to be taken, were more likely to be beaten, and that this was regarded as normal.

Fining is usually preferred; but where the wrongdoer cannot pay the fine imposed, thrashing is commonly resorted to .... Ya modiidi ke e nkgwê, says the proverb: ‘(The punishment) of a poor person is a white-backed ox’ (referring to the discoloration produced by the bruises).^³⁹

The suggestion has sometimes been made that seeing flogging as worse than loss of property is a modern attitude. While this cannot be definitely refuted, the sources indicate that flogging was generally reserved for poor and low-status people, and to subordinate ethnic groups; and that it was applied when a fine could not be paid. Records from Seepapitso’s kgotla in the early 1910s repeatedly show those convicted paying fines, or being flogged because they could not pay.^⁴⁰ Thus what evidence is available consistently points to flogging being regarded as worse than fines. Attitudes in comparison to the modern punishment of imprisonment are another matter, but outside our ambit.

The missionary John Mackenzie, who lived with the Bamangwato in the 1860s, also noted the dominance of fines:

Among the Bamangwato a fine is the usual punishment for all offenses. For murder the theory of the law demands the death of the murderer; its practice is usually satisfied with a fine. For theft the theory is to restore fourfold;^⁴¹ but the practice is to be content if you can get back your own property.^⁴²

Note the distinction between the “theory of the law,” which is perhaps the source of tradition, and the “practice”. In explaining this, Mackenzie emphasized the limited powers of enforcement available even to major dikgosi such as Sekgoma I or Khama III, successive rulers of the Bamangwato. It should be noted that Mackenzie was in many ways a sympathetic observer. He commented on the unreasonableness of European visitors who did not appreciate the realities, praised the high quality of analysis in kgotla trials, and drew comparisons with Scottish justice in the relatively recent past.^⁴³ He was a resident for a long period, observing daily life and discussing society with many people including the kgosi.

In a similar vein to his distinction between theory and practice, Mackenzie also noted that if the culprit was a man of importance, it was frequently very difficult to punish him.^⁴⁴ Analyzing the evidence about precolonial justice is made more difficult by the fact that the power of Tswana rulers varied greatly according to circumstances.

Flogging members of the elite seems to have been abnormal and potentially dangerous; something that could lead to violent conflict. In Bahurutshe tradition, civil war was provoked by Motebele’s flogging of his brother Motebejane and his mophato over an unclear incident involving a baboon.^⁴⁵ In 1926 Tshekedi Khama, regent of the Bamangwato, ordered the flogging of the Ratshosa brothers, following what can only be regarded as deliberate provocations by Tshekedi, who was then a new regent determined to break the power they had built up. They refused to submit, and attacked Tshekedi.^⁴⁶ Perhaps Motswasele II, the Kwena kgosi assassinated c.1821 and remembered as a harsh ruler,^⁴⁷ beat the wrong people too often. He does however seem to have gone unusually far from Tswana norms. He was remembered and condemned in particular for his repeated use of the death penalty, which may suggest that (as with Bamangwato practice reported by Mackenzie) it was normally rare in practice.

Within mephato, discipline was maintained by courts of the relevant section, which could (according to modern sources) order flogging.^⁴⁸ This would be in the context of a mutually dependent group, making offenses a matter of possibly serious general concern.

All this suggests that flogging became more common in the twentieth century, due to the colonial situation. British power now stood behind the kgosi, at least the kgosi who correctly read the new situation, making enforcement easier and altering the balance between ruler and ruled.^⁴⁹ In precolonial times, although a kgosi was “the decider,” he had to consider that if he went too far men could leave, and there was always the possibility of assassination lurking in the background. In 1916 the kgosi of the BaNgwaketse, Seepapitso III, was assassinated.^⁵⁰ The aftermath of the assassination, in which the deceased kgosi’s brother was executed by the colonial authorities for the crime, showed that killing a kgosi was no longer an option, and Sekgoma Khama’s secession from Khama III, early in the Protectorate, was the last large-scale secession to be tolerated by the British. (The episode later faded from memory because Sekgoma was reconciled with Khama before the latter’s death, and succeeded him as kgosi of the Bamangwato.) In the Kgalagari villages the threat of secession remained a major check on autocracy up until the Independence era. However, although it remained possible it was losing its effectiveness even there: on the one hand a village was now a center of resources such as boreholes or schools, and on the other the increase in central authority meant that secession would not change the basic situation.^⁵¹

The social meaning of flogging is complicated by its role in bogwera (male initiation). This involved extensive beating of initiates, to a degree which sometimes led to death, especially in the case of orphans or the very poor. ^⁵² Rituals included one where adult men went to the initiation camp and beat the initiates, especially their own relatives, with flexible whips in order to draw blood and leave large scars.^⁵³ This was believed to confer wisdom as well as to harden them.^⁵⁴ Livingstone noted that in the course of this a younger man taking part in the beating might himself receive a blow from an elder, and that he had seen the kgosi himself (Sekgoma I) receive a severe cut from an older man.^⁵⁵ Initiation is a large subject beyond our present scope, but such deliberately traumatic experiences play a part in creating strong bonds overriding other loyalties. Paul Landau has noted the appearance in some initiation ceremonies of behavior cementing masculine solidarity against even one’s female kin.^⁵⁶ This was suffering embraced as part of status and identity, and it seems unlikely it altered the significance of flogging as a punishment—certainly, the behavior of those who might be subject to the latter (preferring fines) does not suggest attitudes much different from those of people in other societies. Fred Morton has argued that initiation was primarily to graduate “individuals, not fighting units.”^⁵⁷ But the important point here is that the pain was significant precisely because it was endurance of a highly undesirable experience.

The fact that murder could often be settled by fine suggests an element of horizontal justice. In societies where central authority lacks sufficient power to enforce the law, groups such as clans may resort to vendetta or blood feud if their members are killed. In such a situation the peace is kept by agreement between groups rather than by a ruling power able to dominate all of them. To avoid a vendetta, compensation is likely to be offered to the offended lineage. Mackenzie noted that compensation for killing had at one time existed in his own country of Scotland.^⁵⁸ Lesser offenses were even less likely to be considered public matters: in discussing the role of the pitšo (kgotla) meeting Moffat commented that “private thefts, murder, and a host of other crimes passed unnoticed in these assemblies, and were left to the avenger.”^⁵⁹Casalis believed that a fundamental emphasis on property “places all delinquencies in the category of theft.”^⁶⁰

A concept of vertical justice seems to have been developing, but realities lagged behind. It has been suggested that African societies tended to punitive rather than compensation systems mainly when there was a high degree of centralized authority.^⁶¹ Shaka Zulu, a powerful ruler in a centralized kingdom, could impose harsh punishments. By contrast, consider the “stateless” or acephalous societies of West Africa. There was a wide variety of practice in these societies, but some patterns are visible. In particular, there is a de facto division into those offenses which are matters between persons or families, and those offenses which are threats to the whole community. In general, murder was a matter between families and often settled by compensation. Indeed, in some groups, a murderer was simply isolated for a month or so after which the matter was considered closed.^⁶² Although in some cases a murderer might be executed, or sold into slavery,^⁶³ it is notable that if he escaped his family might have to provide another victim, indicating that even a death sentence may not show straightforward vertical justice concerning an individual.

Theft, however, was typically a threat to the community, requiring punishment by the community, often consisting of public humiliation.^⁶⁴ Repeat offenders were often sold into slavery.^⁶⁵ In the case of the commercially-oriented Aro (Igbo), theft in the market was a capital offense.^⁶⁶ Murder of a kinsman affected the community because it was an abomination, which attracted the anger of the gods. Persons considered a public danger, such as repeat murderers, were also threats to the community. Witchcraft was a very serious threat to the community.^⁶⁷

West Africa was also notable for the use of pawns in commercial transactions crossing political boundaries. Even in the modern world, horizontal arrangements are necessary at the international level where there is no authority capable of compelling compliance.

Mackenzie recorded, among the Bamangwato, a system of progressively more extreme punishments where someone became identified as a habitual thief, moving from beating, to mutilation of the fingers, and then of the whole hand, by boiling oil. Mackenzie does not make clear how common this was but seems to imply he had seen relatively few people thus mutilated. He thought that the next stage would be death, but seems uncertain, so presumably it had not arisen during his residence.^⁶⁸ Although occasional theft was not uncommon,^⁶⁹ habitual thieves put themselves outside the normal protection of the community by repeated offenses. A murderer was just fined, even when vertical justice applied, because he or she retained some protection from his or her family and allies, but a habitual thief would become intolerable even to his own. Casalis also stated that an “incorrigible thief” might sometimes be killed, and that there was a proverb to this effect, though he did not cite it.^⁷⁰ These observations fit the pattern noted above, though the Bamangwato penalties noted by Mackenzie seem to offer more opportunities than most for reform.

Premodern Tswana practice similarly shows a tendency to a division between interpersonal conflicts and threats to the community. Horizontal justice, with compensation, was commonplace. Vertical justice existed, and in principle could include murder. In practice, however, it was exercised mainly where there was a threat to the community as a whole, as with witchcraft, or troublesome repeat offenders, though Casalis was told by Basotho that a fear of social disorder could deter revenge in favor of recourse to the king.^⁷¹ Execution for theft, especially of low-status or outsider individuals, seems to be mentioned more often than for murder.^⁷² The category of abomination as a general threat does not seem have been as significant as in West Africa: while wrong behavior raised the anger of the ancestors, their retribution was normally conceived as against the particular offender. Burchell’s report in 1815 that among the BaTlhaping murder was unlikely to be punished by the kgosi at all suggests a mainly horizontal system.^⁷³

Tswana polities varied considerably in the degree of central authority, which would be consistent with variation in types of sanction. In small polities, central authority was naturally weak. On the other hand, some larger polities were the result of successful combinations in the face of difficulty. This was a limiting factor in the case of Moshoeshoe’s power in Lesotho,^⁷⁴ and may possibly also have operated in the Tswana polities which grew during and after the difaqane/mfecane, but the latter were more able to subordinate the adhering groups, and the Bamangwato chiefship was a moderately powerful one.

In the twentieth century, Schapera noted a distinction in practice between civil law, which included theft, considered as an offense against property, and criminal law, which dealt with “offences harmful to the tribe generally and therefore deserving of punishment.”^⁷⁵ In this category Schapera mentions homicide, witchcraft, and incest.^⁷⁶ Incest is interesting as a possible example of the “abomination” category. Schapera’s categorization would be consistent with evolution from the premodern pattern we suggest: the original category of things “harmful to the tribe” has increased, thus coming to approximate the European category of criminal law.

The shift towards vertical justice could be a matter of degree. A Tswana account of Khama III’s reforms, in the missionary-run SeTswana-language newspaper Mahoko A Becwana in 1896, lists “practices of his forefathers” that he had banned. These included initiation and rainmaking, but also, strikingly, murder (polaō ea motho).⁷⁷ This suggests that Khama may have introduced, perhaps by practice rather than edict, a clearer law of murder as a personal crime which would definitely lead to prosecution. Such changes did not necessarily meet general approval: when Amraal Lambert, chief of a then powerful Nama group, declared that henceforth murder would be punished by death, it raised significant popular opposition to his rule.^⁷⁸ The process has parallels in early medieval Europe,^⁷⁹ where the new idea of justice as something imposed from above was explicitly linked to royal power: an eighth-century text approvingly states that “the word of a king is a sword for beheading, a rope for hanging, it casts into prison, it condemns to exile.”^⁸⁰

A very interesting passage in Schapera’s study of rainmaking links restraint in sentencing with community harmony. Rain was believed to depend on a number of factors, including peace within the community, especially involving the kgosi. Indeed, the linking of rain with harmony has not disappeared: “If people understand one another, then rain will fall.”^⁸¹ Kgabyana, the daughter of the late Kgatla kgosi Linchwe to whom he had entrusted his rainmaking tools, stated that the regent Isang did (contrary to rumor) know rainmaking. However,

‘whenever he makes it there is plenty of lightning. This is not due to ignorance, but because he neglects the law. The great law is that one must be kind and fond of people, and not quick to thrash them’ (an allusion to Isang’s habit as judge of often imposing sentences of corporal punishment).^⁸²

Kgabyana was speaking in 1934, suggesting that Isang’s increased use of beating was not yet completely accepted as normal. Isang is worth considering because his regency illustrates precolonial patterns persisting into the colonial period. He was in conflict with his nephew Molefi, the heir, with the two representing, as often in Tswana history, a generational divide. Isang was an effective and successful regent who achieved much in terms of development, but he was excessively harsh and had a violent temperament. Molefi by contrast was in many ways irresponsible, yet was popular especially because of his fairness and mildness in the kgotla.^⁸³ After Molefi took over, Isang lost a case against a man who had cheated him, and attacked the culprit himself. He came close to killing him, and dumped the body in front of the kgosi.^⁸⁴ In precolonial times, Isang’s behavior when regent would probably have led either to a secession or to his assassination.

At their installation, dikgosi were traditionally warned not to be harsh in administering justice. In addition, there were formulae which a bystander, at least if someone of status, could utter when he considered that a sentence of flogging should not be carried out, or should not continue, and it was expected that the kgosi should immediately accede.^⁸⁵ If the sentenced man succeeded in running to the house of the kgosi’s wife or mother, he had reached asylum and could not be punished.^⁸⁶Gulbrandsen has analyzed the ways in which kgotla discourse connects with Tswana cosmological understanding.^⁸⁷

There were a number of restraints on the precolonial kgosi. The most important business was conducted in his kgotla, open to all male citizens. Within the kgotla there was open discussion; in Moffat’s description “the greatest plainness of speech,” though he noted that it was the “minor chiefs” whose views were of significance.^⁸⁸ (Indeed, this openness has sometimes been seen as a basis of modern Botswana’s political tolerance: there is a proverb “All the words said in the kgotla are beautiful words”: Mafoko a kgotla a mantle otlhe.) In practice, direct criticism of the kgosi was usually beyond the pale (another proverb holds that “The chief is a little god; no evil must be spoken of him”: Kgosi ke modingwana, ga e sebjwe). But if his misdeeds went far enough, or his policies were very unpopular, open dissent would be voiced. Perhaps more commonly, dikgosi used the meeting to assess the level of possible opposition. Although a kgosi was in principle born rather than made, and might be very much in control, his legitimacy depended on general acceptance: kgosi ke kgosi ka batho, “a chief is a chief by the people” as the proverb had it. His close relatives had to be consulted, and acted as a check, sometimes even secretly fining the kgosi.^⁸⁹Casalis made the interesting observation that chiefs were normally addressed by name, with the honorific formulae used only on “state occasions,” and that they were “interrupted and contradicted without ceremony.”^⁹⁰ Casalis described this as applying to “every one.” Even if that is an exaggeration, it does suggest that status markers originally limited to the most powerful dikgosi later spread down the spectrum.

Beyond the day-to-day level, there was always the possibility of secession, though this would frequently require fighting off an attempt at recapture. Labor rather than land was the key scarce resource, and the Tswana settlement pattern of large villages has often been interpreted as based on the need (from the kgosi’s point of view) to control population. It is highly significant that after independence, when the powers of the dikgosi to control settlement lapsed, there was a trend toward “lands settlement” in which people moved away from the classic large villages and settled at their agricultural and pastoral lands.^⁹¹ The implication is that ordinary people often did not perceive the large village as being to their personal benefit. (This trend was later reversed since development made the villages attractive for new reasons.)

Secession was so common in the precolonial period that one suggested derivation for “BaTswana” is a verb meaning “to separate from one another”.^⁹² The BaNgwaketse and Bamangwato split from the BaKwena at some time in the past, and the BaTawana split from the Bamangwato in the late eighteenth century. Apart from secession, violent opposition was also possible, and dikgosi were sometimes deposed as a result of dynastic disputes,^⁹³ popular discontent,^⁹⁴ or seizure of power.^⁹⁵

Colonial rule altered the situation, though in doing so it was pushing a trend which was already visible with the growth of large and more organized merafe. British power increased that of the kgosi, inasmuch as the kgosi could make himself the necessary agent of British rule, while the incentive to use fines to gain wealth was reduced. Commercial farming made ordinary accumulation easier, especially when coupled with judicious use of customary rights.^⁹⁶ Attention to cases where dikgosi fell foul of the colonial authorities, such as those of Sebele II or Gobuamang, who were deposed, has tended to obscure the advantages for the skillful ruler.

Perhaps some idea of the precolonial situation can be obtained from the BaKgalagari villages in which the authority of major dikgosi or the central government remained distant into the time of Independence.^⁹⁷ The BaKgalagari^⁹⁸ are a group closely akin to the Batswana, but who were historically reduced either to subordinate status or to remote areas including, as the name suggests, the Kalahari. Judicial business was conducted in the lekgota, and several points stand out. Firstly, although cases dealt with individuals, the lekgota treated the family group as a unit in which the head was responsible for the other members. Secondly, fines were the normal penalty for full citizens, with flogging only common for groups such as San (“Bushmen”) serfs. In a village observed by Adam Kuper, there were only two cases of citizens being flogged between 1964 and 1967. These were because of repeated refusal to pay overdue fines, and took place only after the district commissioner became aware of the overdue fines and demanded action.^⁹⁹ Citizens preferred to seek compensation rather than penal sanction. Traditionally fines were redistributed in the lekgota, though by the period of Independence they were usually in cash and paid to the Tribal Treasury. The distribution of fines is a difference from Tswana tradition, and may reflect the even more limited nature of the chief’s authority, relying on majority support and consensus.^¹⁰⁰ Thirdly, the extent to which penalties were imposed reflected the political realities of the village, and there were a number of situations where action was difficult.

The importance of family meant that men without close male relatives were more likely to be troublesome, since the usual structure for intervention through such relatives was absent.^¹⁰¹ As with Mackenzie’s account of the Bamangwato, it was very difficult to act against an influential man.^¹⁰² In some cases the village might simply be afraid of someone as a violent individual, as a speaker in the lekgota explained (square-bracket insertions are Kuper’s):

There are five kgota’s [in Ghanzi District]. Men are thrashed in Tsekwe’s kgota [Kalkfontein] but not in Ramoswane’s or Keakopa’s [Kuli and Nojane]. Here people drink khari and destroy the kgota. They have knives and would knife any man who tried to thrash them. Will we give R. to the police tomorrow? And when he takes out his kierie [club] we will say, No, it is the headman who called the police, go and club him.^¹⁰³

Most of these aspects are consistent with the nineteenth-century observers’ picture of Tswana justice. The individually dangerous man might have been less of a problem in a larger village where he would be more easily overwhelmed, though his capability for subsequent revenge might have caused hesitation.

Significantly, there was one headman in the region who did not have these problems, the sub-chief of Hukuntsi, who the colonial government had given a status above the “headmen” of the other villages, and police backing. “He favors imprisonment for many offences, and since he rules with police backing he can dispense with the approval of his subjects.”^¹⁰⁴ Here perhaps we can see the appearance of the new order that came with large merafe and colonial backing. Kuper specifically notes the impact of colonial (and later Botswana government) backing.^¹⁰⁵ In this case the modern phenomenon of imprisonment is involved, but the principle is clear.

This perspective is important because, when “the Tswana” are mentioned in broad studies, the data usually refers to the large polities which came to dominate other groups during the nineteenth century. This is perhaps hard to avoid in terms of availability of secondary sources, but can lead to an unbalanced picture.

Modern Batswana, however, tend to regard flogging as an unproblematically “traditional” punishment. This is not peculiar to Botswana. In Malawi, for example, opinion in favor of a traditional justice believed to be harsher and less prone to manipulation than western law seems to have been strong from the mid-colonial to the early independence periods, at least among those whose views received publicity.

In 1967 Dr. Banda [first president of Malawi] expressed his explicit disapproval of such foreign importations as the principle that a man was innocent until proven guilty, the need for corroborative evidence, and the notion that intention was important in cases of legal [sic] killing.^¹⁰⁶

By this time Dr. Banda was effectively dictator (and notoriously intolerant), so the enthusiastic support he received for his views should not necessarily be taken at face value, but such opinions had been widely expressed before.^¹⁰⁷

The discrepancy between modern belief and what independent evidence suggests is not exactly a case of “invention of tradition.” The changes were a real development rather than an imposed interpretation. Rather, a relatively modern development has been projected back into an earlier period, creating a belief in an “immemorial” tradition.

Although there is a fairly clear record of the limits of flogging in the late colonial period, it is striking that there have been attempts to extend it into changed contexts. In 1982 an Act of Parliament introduced a new sentence (for burglary) of flogging at repeated intervals of months.^¹⁰⁸ This was held by the courts to be unconstitutional; its novelty meant it was not protected by the constitutional derogation.^¹⁰⁹ In 2010, Kgosi Kgafela II of BaKgatla expanded the use of flogging by authorizing mephato to flog people for various offenses outside the normal kgotla setting.^¹¹⁰ This followed his revival of traditional initiation, and was intended to counter perceived problems of crime and social disorder. Kgafela claimed “traditional” authority for his actions, was unwilling to accept the government’s attempts to rein him in, and was eventually prosecuted and lost control of his morafe. These incidents illustrate that transformations of “traditional” punishment are possible even now. Interestingly, Kgafela had previously been a notable human rights lawyer.^¹¹¹

At this juncture the case of Unity Dow v. Attorney-General (1990) should be mentioned. Under the Citizenship Act 1984, children acquired Botswana citizenship from their father, but could only acquire citizenship from the mother if she was unmarried. Unity Dow, a Botswana citizen married to an American, challenged the Citizenship Act 1984, and the courts found that the law was unconstitutional.^¹¹² In its case, the state drew attention to customary law, noting multiple examples of gender-discriminatory aspects. “The whole fabric of the Customary Law in Botswana... is based upon a patrilineal society, which is gender-discriminatory in its nature.... It is not unfair to say that if gender discrimination were outlawed in customary law, very little of customary law would be left at all.”^¹¹³ There was a tendency on the part of the government to portray the women promoting change as an élite, out of touch with the ordinary women it claimed to have consulted in dikgotla.^¹¹⁴ One of the members of the Court of Appeal responded that the derogation clauses showed the limits of customary law. Old customary law could continue, but nothing new could be done which did not meet constitutional standards.^¹¹⁵ The court did not go into the fact that customary law was not in fact static, and changes had been made concerning the rights of women, notably in rights of inheritance.^¹¹⁶ Change has been made in the recent past in some parts of the region: it is important to note that women often prefer the accessible customary system despite the theoretically greater equality of the modern system.^¹¹⁷

In conclusion, therefore, the common understanding of flogging and the death penalty in modern Botswana as having “immemorial” traditional roots is problematic. The premodern situation seems to be vaguely imagined as an analogue of present justice, but the evidence suggests rather a much less consistent environment, in which compensation or revenge, rather than sentence by a court, often prevailed. All this raises questions about how Botswana understands and relates to its past. Traditional society is widely appealed to as a source of practice and culture for the present, and it is noteworthy that this is a recurring theme in Botswana fiction, from Love on the Rocks^¹¹⁸ onwards.^¹¹⁹ However, is the tradition to which appeal is made really what existed before the coming of colonialism?

1 Practice in the region is to use the form with prefix: e.g. “Mosotho,” Sotho person, plural “Basotho”. However “Motswana” now means citizen of Botswana, and “Tswana” is therefore preferable when the sense is Tswana ethnicity.

2 Constitution of Botswana, s.7. These “derogation clauses” provide that no punishment which was lawful immediately before Independence can be held to violate the prohibition of inhuman treatment.

3 Isaac Schapera’s classic, A Handbook of Tswana Law and Custom Compiled for the Bechuanaland Protectorate Administration (1938; reprint, Münster-Hamburg: International African Institute & LIT Verlag, 1994) was compiled for the colonial government and has remained a standard work in Botswana courts.

4 See I. Schapera, “Tswana Concepts of Custom and Law,” Journal of African Law, 27, 2 (1983), 141–149, for a rebuttal of arguments that the Tswana lacked a category of law distinguishable from behavioral norms in general.

5 See Bruce S. Bennett, “Women Chiefs and Pre-colonial Tswana Patriarchy,” Botswana Notes and Records, 51 (2019), 38–48.

6 C.M.N. White, “African Customary Law: The Problem of Concept and Definition,” Journal of African Law, 9, 2 (1965), 86–89.

7 I. Schapera, Tribal Innovators: Tswana Chiefs and Social Change, 1795–1940 (London School of Economics Monographs on Social Anthropology no 43) (London: Athlone Press, 1970).

8 Schapera, Handbook, 274. See Paul S. Landau, Popular Politics in the History of South Africa, 1400–1948 (Cambridge: CUP, 2010), 236; Bruce S. Bennett and Alison Wallis, “Khama v Ratshosa Revisited: The Privy Council Ruling of 1931 on House-burning,” Botswana Notes and Records, 44 (2012), 25–33.

9 In 1995 there were riots in Mochudi and Gaborone over a case in which it was believed that a girl had been murdered for body part for witchcraft, and that the authorities were failing to act.

10 Robert Moffat, Missionary Labours and Scenes in Southern Africa (New York: Robert Carter, 1843), 151. There is a proverb “The head is the chief’s,” understood as meaning that only he could pronounce the death sentence: I. Schapera, “Tswana Legal Maxims,” Africa: Journal of the International African Institute, 36, 2 (1966), 122.

11 In O. Lekorwe, “History of Capital Punishment in Botswana” (unpublished research essay, University of Botswana, 2009), it is notable that modern oral informants describe the death penalty as having been standard, but other sources are more ambivalent. See Andrew Novak, “Capital Punishment in Precolonial Africa: The Authenticity Challenge,” The Journal of Legal Pluralism and Unofficial Law, 50, 1 (2018) for a broader overview.

12 William J. Burchell, Travels in the Interior of Southern Africa (London, 1824), II, 554. The passage is quoted in Moffat, Missionary Labours, 308.

15 Elizabeth Lees Price, journal (5827), Shoshong, 23 November 1863, The Journals of Elizabeth Lees Price Written in Bechuanaland, Southern Africa 1854–1883 with an Epilogue: 1889 and 1900 ed. Una Long (London: Edward Arnold, 1956), 147. See also Margaret Kinsman, “‘Beasts of Burden’: The Subordination of Southern Tswana Women, ca. 1800–1840,” Journal of Southern African Studies, 10, 1, (1983), 39–54, 52.

16 John Campbell, Travels in South Africa Undertaken at the Request of the Missionary Society (London: Balck & Parry, 1815), 519

17 Kok was a controversial figure for trading to support himself; often discussed together with William Edwards.

20 Anon., May 1883, letter to editor, in Words of Batswana: Letters to Mahoko A Becwana, 1883–1896, translated and edited by Part T. Mgadla and Stephen C. Volz (Cape Town: Van Riebeeck Society, 2006), 193.

22 This phrase is interesting. Potential rivals to the kgosi were assassinated in this manner, but it could suggest a more general idea of the common pattern of retribution.

25 Sechele Motswasele, September 1883, Words of Batswana, 197. Interpolations: “already” ours; “as witches” editors of Words of Batswana.

26 This episode provided the inspiration for Lauri Kubuitsile’s novel But Deliver Us From Evil (Cape Town: Penguin, 2019).

27 Bechuanaland Protectorate Witchcraft Proclamation, 1927 (No. 17 of 1927) (Bechuanaland Protectorate Orders in Council and Proclamations, 30th June, 1890 to 31st December, 1929 (Kimberley: Government Printer, 1930), p. 471.) See Isaac Schapera, “Sorcery and Witchcraft in Bechuanaland,” African Affairs, 51, 202 (1952), 41–52, 42.

29 Andrew Smith, unpublished manuscript, quoted in Fred Morton, “Bogwera and Mephato,” Botswana Notes and Records, 43 (2011), 39. Initiation is still very secret, even if there is no longer a physical threat.

31 Elizabeth Lees Price, letter (5921), 8 November 1880, Journals of Elizabeth Lees Price. Price reports that on one occasion a young Englishman was so moved by “boiling indignation” at the torture that he seemed about to attack Sechele and had to be forcibly removed by his companions.

33 Kgafela II and Another v. The Attorney General and Others In Re: Gabaokelwe v. The Director of Public Prosecutions, (2012) 1 BLR 699 CA p. 717.

34 One of the most famous colonial controversies included a case of flogging: see Michael Crowder, The Flogging of Phinehas McIntosh : A Tale of Colonial Folly and Injustice: Bechuanaland 1933 (New Haven: Yale University Press, 1988).

35 Burchell, Travels, II, 347. Burchell unfortunately does not define how “atrocious” a crime had to be to qualify.

36 Burchell, Travels, II, 347. See also Diane Wylie, A Little God: The Twilight of Patriarchy in a Southern African Chiefdom (Hanover NH: Wesleyan University Press, 1990), 27. Incidentally, such observations suggest that there is a danger for modern historians of allowing the complex property systems of redistribution and patronage to obscure the basic fact of de facto ownership and inequality. That the kgosi kept the whole of fines is supported by Schapera, “Tswana Legal Maxims,” 123. Giving some to the victim as compensation was a modern innovation. By the late colonial period all animals were sold for the Tribal Treasury: I. Schapera & John L. Comaroff, The Tswana (1953; rev. ed., London: Kegan Paul: 1991), 51.

37 Schapera “Tswana Legal Maxims,” 123. Eugène Casalis, The Basutos, or, Twenty-three Years in South Africa (1861; reprint, Morija: Morija Museum & Archives, 1997), 225–6 may refer to this concept.

40 I. Schapera, “Early European Influences on Tswana Law,” Journal of African Law, 31, 1/2, (1987), 151–160, 153. Seepapitso, the Ngwaketse kgosi, was a noted modernizer and kept records of kgotla proceedings in the 1910s.

42 John Mackenzie, Ten Years North of the Orange River: A Story of Everyday Life and Work among the South African Tribes from 1859–69 (Edinburgh: Edmonston & Douglas, 1871), 375.

44 Mackenzie, Ten Years North of the Orange River, 374. This is, of course, hardly restricted to premodern societies.

45 J. Mpotokwane, “A Short History of the Bahurutshe of King Motebele, Senior Son of King Mohurutshe,” Botswana Notes and Records, 6 (1974), 38; J.T Brown, Among the Bantu Nomads. A Record of Forty Years Spent Among the Bechuana (London: Seeley, Service & Co. 1926), 263. The detail about the mophato is probably anachronistic, if Fred Morton’s dating of the institution is correct: Fred Morton, “Bogwera and Mephato,” 42; Morton, “Mephato: The Rise of the Tswana Militia in the Pre-colonial Period,” Journal of Southern African Studies, 38:2 (2012), 390.

46 Bruce S. Bennett & Alison Wallis, “Khama v Ratshosa Revisited: The Privy Council Ruling of 1931 on House-burning,” Botswana Notes and Records, 44 (2012), 26.

47 I. Schapera, “Notes on the Early History of the Kwena (Bakwena-bagaSechele),” Botswana Notes and Records, 12 (1980), 83-87; 85; Schapera, Government and Politics in Tribal Societies, 152.

49 See Christian John Makgala, “Taxation in the Tribal Areas of the Bechuanaland Protectorate, 1899–1957,” Journal of African History, 45 (2004), 279–303 for ways in which British rule empowered dikgosi and their officials against subjects. Some dikgosi, of course, failed to read the situation correctly and lost out.

50 Yonah Hisbon Matemba, “The Assassination of Kgosi Seepapitso Gaseitsiwe of the BaNgwaketse 1916–17,” Botswana Notes and Records, 34 (2002), 25–36.

51 Adam Kuper, Kalahari Village Politics: An African Democracy (Cambridge: Cambridge University Press, 1970), 73.

52 J. Tom Brown, Among the Bantu Nomads: A Record of Forty Years Spent Among the Bechuana, A Numerous and Famous Branch of the Central South African Bantu, with the First Full Description of their Ancient Customs, Manners & Beliefs (London: Seeley, Service and Co., 1926), 80–82.

53 David Livingstone, Missionary Travels and Researches in South Africa, Including a Sketch of Sixteen Years’ Residence in the Interior of Africa, and a Journey From the Cape of Good Hope to Loanda on the West Coast; Thence Across the Continent, Down the River Zambesi, to the Eastern Ocean (London: John Murray, 1857), 146–7; Mackenzie, Ten Years North of the Orange River, 376.

55 Livingstone, Missionary Travels, 146–7. This part of the ceremony was evidently not performed in secret.

62 O. Oko Elechi, “The Igbo Indigenous Justice System,” in Viviane Saleh-Hanna, Colonial Systems of Control: Criminal Justice in Nigeria (Ottawa: University of Ottawa Press, 2008),. 410.

64 Elechi, “The Igbo Indigenous Justice System,” 412. A case of public humiliation of a thief among the Bamangwato involving stripping him naked in the kgotla is described by James Chapman, Travels in the Interior of South Africa, Comprising Fifteen Years’ Hunting and Trading; With Journeys Across the Continent from Natal to Walvisch Bay, and Visits to Lake Ngami and Victoria Falls (London: Bell & Daldy, 1868), I, 112.

67 For examples elsewhere in Africa, see Novak, “Capital punishment in precolonial Africa,” 72. See also e.g. J. Kenyatta, Facing Mount Kenya (London: Secker & Warburg, 1938), 230, quoted in Suzette Heald, “Witches and Thieves: Deviant Motivations in Gisu Society,” Man, New Series, 21, 1 (1986), 65–78, 65, grouping habitual theft and witchcraft as public dangers meriting the death penalty.

68 Mutilation for habitual theft is mentioned as historical in Schapera, Handbook, 50. Livingstone noted that Sechele introduced laboring on the roads as a punishment after witnessing it in the Cape Colony. Livingstone, Missionary Travels, 121.

72 E.g. see Samuel Broadbent, A Narrative of the Introduction of Christianity Amongst the Barolong Tribe of Bechuanas, South Africa: With a Brief Summary of the Subsequent History of the Wesleyan Mission to the Same People (London: Wesleyan Mission House, 1865), 82–5.

77 Rratshosa M. Segokotlo, April 1896, Words of Batswana, 259. Old SeTswana orthography. Polao may mean murder or killing.

81 Interview with Kgosi Shaw Moraka, Lerala, 24 April 1990, in Paul Stuart Landau, The Realm of the Word: Language, Gender, and Christianity in a Southern African Kingdom (Portsmouth, NH: Heinemann, 1995), 207.

82 I. Schapera, Rainmaking Rites of Tswana Tribes (Leiden & Cambridge: African Studies Centre, 1971), 29.

83 Fred Morton & Jeff Ramsay (eds), The Birth of Botswana: A History of the Bechuanaland Protectorate from 1910 to 1966 (Gaborone: Longman Botswana, 1987), 87. Differing opinion about the two can still be seen in secondary literature.

84 Fred Morton, When Rustling Became An Art: Pilane’s Kgatla and the Transvaal Frontier (Claremont: New Africa Books, 2009), 276. Had the victim died Isang would probably have been charged with murder; see Rex v. Zanekile alias Martin, 1908, DCG 23/1, Botswana National Archives; Bruce S. Bennett, “Mogotse’s Goats and Other Cases Before the Gaberones Magistrate in 1908,” Botswana Notes & Records, 40 (2008), 27–8.

85 Schapera, “Tswana Legal Maxims,” 130. The formulae were ritualized and varied between groups; for example among the Kgatla Aefologe setlhare kgabo, “Let the monkey [Kgatla totem] descend from the tree.”

86 Schapera, Handbook, 294–5. The role of the mother of the kgosi is insufficiently explored; Sekgoma I’s mother was a powerful and to some extent public figure (Mackenzie, Ten Years North of the Orange River, 436).

87 Ørnulf Gulbrandsen, The State and the Social: State Formation in Botswana and its Pre-Colonial and Colonial Genealogies (New York: Berghahn, 2014), 166–77.

91 R.M.K. Silitshena, “Chiefly Authority and the Organization of Space in Botswana: Towards an Exploration of Nucleated Settlements among the Tswana,” Botswana Notes and Records, 11 (1979), 55-67; 55–6. See also Kuper, Kalahari Village Politics, 19, 54, 73.

92 Desmond T. Cole, An Introduction to Tswana Grammar (1955; reprint, Cape Town: Longman, 1995), xxi–xxii. The correctness of this etymology is disputed but that is not the point.

93 I. Schapera, “Kinship and Politics in Tswana History,” Journal of the Royal Anthropological Institute of Great Britain and Ireland, 93, 2 (1963), 159-173.

94 Thomas Arbousset, Missionary Excursion into the Blue Mountains: Being an account of King Moshoeshoe’s Expedition from Thaba-Bosiu to the Sources of the Malibamatšo River in the Year 1840, ed. & trans. David Ambrose and Albert Brutsch (Morija: Morija Archives, 1991), 112; Transvaal Native Affairs Department, Short History of the Native Tribes of the Transvaal (Pretoria: Government Printer, 1905), 12, 15.

95 In the case that the deposed kgosi was admitted to be the legitimate heir, his lineage might continue to hold some of the kgosi’s ritual prerogatives (Schapera, Handbook, 57). This is suggestive in terms of its recognition that law may exist but not be enforceable.

98 The SeTswana form of the word is Kgalagadi, but with increasing recognition of SheKgalagari as a separate language its own orthography is now common. “Kalahari” is an old spelling of the same root.

104 Kuper, Kalahari Village Politics, 162. This Sub-Chief seems to have used this power aggressively against opponents (80).

105 Kuper, Kalahari Village Politics, 72. However, it is true that this backing was still sufficiently remote for it to be risky for chiefs to rely on.

106 Martin Chanock, “Neo-Traditionalism and the Customary Law in Malawi,” African Law Studies, 16 (1978), 81.

109 Petrus and Another v. The State 1984 BLR 14 (CA). The same derogation was cited in Ntesang v. The State 1995 BLR 151 (CA) to the effect that the Court could not review the death penalty despite concerns as to whether it might be inhuman.

111 See his book Kgafela Kgafela II, The King’s Journal: From the Horse’s Mouth (Bloomington IN: Authorhouse, 2014) for a very interesting account of his life, which however ends with the initiation.

112 Simon Coldham, “Unity Dow v. Attorney-General (Botswana),” Journal of African Law, 316, 1 (1992), 91–92.

113 Respondent’s Heads of Argument, in Unity Dow ed., The Citizenship Case: The Attorney General of the Republic of Botswana vs Unity Dow: Court Documents, Judgements, Cases and Materials (Gaborone: Lentswe la Lesedi, 1995), 20. This book gives the entire case materials.

114 Judith Van Allen, “What Are Women’s Rights Good For? Contesting and Negotiating Gender Cultures in Southern Africa,” African Studies Review, 58, 3 (2015), 100.

118 Andrew Sesinyi, Love on the Rocks (London: Macmillan, 1981). This romantic novel, something of a pioneering work in popular Botswana English-language fiction, has been enormously popular in Botswana.

119 See Mary S. Lederer, Novels of Botswana in English, 1930–2006 (New Rochelle and Lagos: African Heritage Press, 2014), for a comprehensive study of Botswana fiction in English.