Journal of Social Development in Africa (1996), 11,2,33-42 Some Observations on Crime in Botswana 1980 -1992 CHRISTINE & ROY S LOVE * ABSTRACT Since 1980, the pace of economic and social change in Botswana has been rapid. One might expect this to be reflected in patterns of offending behaviour and court responses. This paper examines trends in certain types of criminal offence during this period, making use of police and prison reports and a summary of magistrates' court records in Gaborone. While there are clear rising trends for a number of major offences and differences in sentencing patterns between them, the paper concludes by warning against making simple connections between these observations and structural changes such as the growth of urbanisation or of presumed urban income inequalities. Introduction Much has been written about Botswana's relative political stability since Independence and its rapid economic growth, particularly during the 1980s. The government has frequently been complimented on its responsible management of the substantial revenues generated from diamond mining over the last twenty years or so, and of having avoided the worst outcomes of countries like Zambia or Nigeria which also have had dominant single industry exports. Such cautious management of the nation's finances has also preempted the need for massive structural adjustment, as experienced by so many other African countries since the early 1980s. This fiscal conservatism has been associated with an official ideology which has downplayed the role of the state in the development process, other man to facilitate private enterprise or to compensate for its initial absence. This has also been reflected in a relatively slow rate of expansion in the provision of social services, and although there has been substantial increase in certain areas, such as the number of primary schools, rural clinics, and nutritional surveillance, in other areas progress has been limited. Universal state pensions for the elderly are lacking, support for the 'destitute' is limited, there is no subsidy for urban transport and lowcost housing has not been prioritised. * Former lecturers, Faculty of Social Sciences, University of Botswana. Present address: 9 Birch Grove, Oughtibridge, Sheffield, S30 3FG, UK. 34 C&RSLove It is in areas such as these that indicators of social progress are usually based. Yet a concentration on the provision of services has the effect of excluding other areas of life, which, in a changing society, can generate as much concern amongst the population in their daily activities as do matters of housing and health. The prevalence, whether actual or perceived, of criminal activities such as burglary and theft or violent crime is one such area. Although, therefore, a nation like Botswana can point to significant development in broad economic terms, together with substantial social improvement since Independence, little is known about social breakdown, even as manifested and defined through the relatively narrow focus of the legal system. This paper therefore attempts to provide greater balance in our understanding of social development in Botswana, by examining a number of critical behavioural parameters in relation to crime and patterns of punishment. Using published and research data for a number of selected offences between 1980 and 1992, we shall first of all examine trends in reported crime and sentencing patterns, and then relate these to the literature about crime and underdevelopment. Reported Crime Table 1 shows that between 1980 and 1992 the total number of offences against the Penal Code reported to the Police rose from 25,314 to 47,462, a rise of 87.5% overall, or 5.3% per annum. As with any crime statistics, a bald statement like this leaves a number of questions begging concerning the meaning of such observations. Factors such as a greater propensity for certain types of offence to be reported, police recording practices, changes in the spatial coverage of policing throughout the country, may all be relevant, in addition to a de facto increase in offences. They are, however, real figures: not only do they measure what is deemed by the authorities to be important, but they also provide the main data base for subsequent policy making. Offences against property - which include burglary and theft, common theft, robbery and stock theft - constitute the largest category both in terms of absolute numbers and percentage increase. Of the absolute increase in all reported offences since 1980, some 65 % are accounted for by these offences and their percentage rose from 48.1 % of all reported offences in 1980 to 56% by 1992. By contrast, the next major group, offences against the person, also increased rapidly over the same period, but fell as a proportion of the total. The growth in property-related crimes is not so surprising in a rapidly growing economy where general benefits are not distributed equally. In itself, however, inequality is an insufficient explanation. While there are clearly those people at the Some Observations on Crime in Botswana 35 Table 1: Major Categories of Offence Reported to the Police, 1980-92 1992 1990 1985 1980 CATEGORY 957 271 1052 323 1116 323 896 240 Public Order Lawful Authority Public in General 4050 13446 2872 9659 3770 12431 2793 Against the Person 8409 12183 26560 1819 22016 2042 16461 1126 Property Property (Injury to) 673 Forgery Corruption 12 19 0 174 16 87 6 170 6 22 24 35 27 Conspiracy 47462 31520 25314 41898 TOTAL Source: Annual Reports, Botswana Police Service, 1980-92 bottom end of the income spectrum from amongst whom the vast number of reported offences are normally blamed, there has also been a great increase in the volume and variety of consumer goods in the country, which, together with other forms of property, provide a greater reservoir for offences such as theft, for instance. There are also the motivations of offenders to be accounted for and we shall return to this at a later stage. Official data also measure only what is considered to be important to the state, in this case an emergent capitalist one, with a postcolonial legal corpus derived from the European model. Given the importance of property rights in this tradition, it is not surprising that infringements of these will increasingly feature as part of the social control processes which accompany 'development' (Davis & Slabbert, 1985; Mihyo, 1995; Sumner, 1982). However, within the general group of property-related offences there are interesting comparisons which may be made. It is tempting to assume that increased urbanisation may be a factor but this begs further questions and we shall return to possible causes at a later stage. Table 2 (overleaf) shows the trends in reported offences in Botswana since 1980 for the major categories of burglary and theft, housebreaking and theft, common theft, and a more 'traditional' type of offence, particularly in Southern Africa, of theft of livestock. 36 C&RSLove Table 2 Selected Offences Against Property Reported to the Police, 1980-92 Year Burglary and Theft 1980 1135 1012 1310 1162 1981 1982 1983 1984 1985 1312 1508 1986 1987 1516 1999 1988 1989 2061 1800 1990 1991 1895 1951 1992 2358 Source: Annual Reports, Botswana Police, 1980-92 In the first three cases there is a clear upward trend, with minor variations, over this thirteen year period, whereas in the fourth case, that of stock theft, a peak is reached in the mid-1980s with a gradual decline thereafter, though still ending up slightly above the 1980 starting point. It could be argued that with an increase in alternative income-earning opportunities and urban migration, a decline in the relative importance of stock theft is not surprising, but the rapid rise and fall over the middle years of the 1980s is more difficult to explain. This coincides with a period of severe drought (1982-88) and although cattle numbers do not tend to respond immediately to rainfall failure, the evidence in Botswana is that numbers did fall continuously through 1982 to 1987 (BoB, 1994). If the greatest stock reductions were amongst those households with the poorest grazing, the worst access to water and limited alternative sources of income, then an increase in stock theft may well have resulted. But alternative possibilities should not be ignored. Owners of large cattle herds may, for instance, be the first to slaughter animals at the onset of drought, when prices are still H'Breaking and Theft 1302 1450 1661 1474 1598 1669 1955 2558 2668 2755 3601 3533 3881 Stock Theft Common Theft 5309 946 964 5019 5729 1280 1080 5333 6201 1347 1391 1415 6927 7730 1375 1204 8656 8689 1167 1029 8928 9664 1155 1041 11291 12331 Some Observations on Crime in Botswana 37 relatively high, though even for them the more rational decision may be more extensive exploitation of common grazing land - in which case smaller farmers will tend to lose out. This strategy might also create more opportunities for thefts. The relationship between drought and stock theft may therefore be an indirect one. It is also possible that inconsistencies in data collection lie behind such short-term variations, although some variation linked to drought conditions would not be unexpected. At present all this can only be speculation in what is an underresearched area in Botswana. Court Processes That the differences between offences are not superficial or spurious, however, is borne out by examining later stages in the criminal justice process. During the same period, the proportion of reported offences of burglary and theft and of housebreaking and theft which led to court summonses averaged between 26% and 28%, while an average of 86% of stock thefts resulted in summonses. This is likely to reflect differences in the tracing of suspected offenders and the strength of the prosecution evidence; it may also be a measure of the seriousness with which stock theft is regarded in a society in which cattle have traditionally had high socioeconomic value. For common theft, which embraces a wide variety of malpractice in multifarious circumstances, the number of summonses as a proportion of offences reported falls between the above figures, at 48.2% averaged between 1980 and 1992. Differences are less pronounced when the proportion of summonses leading to a conviction is examined: on average 79% of summonses for cattle theft led to a conviction while only a slightly smaller proportion (78.4%) for burglary and theft did so. This is not surprising, however, as one might expect a fairly standardised rate of successful prosecutions for offences, whatever their nature, once the decision to prosecute has been taken. Sentencing patterns, on the other hand, do tend to differ more widely. Of those convicted for stock theft, a higher proportion was fined than for both burglary and theft and housebreaking and theft; however, the rate of imprisonment for burglary and theft was higher than for the other offences. In the case of stock theft this may reflect an argument that offenders need to retain their freedom in order to carry on farming, or possibly because of other assets which could be realised, including return of the stolen property. Table 3 (overleaf) summarises the figures. 38 C&RSLove Table 3: Proportion Summonsed, Convicted, Imprisoned and Fined for Selected Types of Theft, 1980-1992 Convicted Summonsed Imprisoned Fined % % 78.4 67.8 60.6 Burglary and Theft H/breaking and Theft Theft (common) 78.2 73.8 Stock Theft % 11.2 15.0 37.1 24.7 28.3 62.2 79.5 Notes % 26.6 27.9 48.2 86.1 • The "Summonsed" and "Convicted" columns are for the the years 1981- 1992 as the 1980 data for Summonses were distorted by a backlog of cases from previous years. This did not affect the columns for those "Imprisoned' and "Fined" which are for 1980-92. Column headings are defined asfolbws: Summonsed: persons summonsed as a percentage of offences reported. Convicted: persons convicted as a percentage of those summonsed. Imprisoned and fined as a percentage of those convicted. Source: Annual Reports, Botswana Police, 1980-92 The absence of any very strong trends in the figures which lie behind these averages seem to reflect a consistent approach by the judicial system to these crimes, although the tendency to punish cattle theft by fine rose from an average of 21 5 % during 1980-86 to 28.5% during 1987-92. The latter practice, once again, may be a response to the ending of the severe drought of the earlier years and a consequent greater ability to pay. The broad conclusion from this comparison of four major types of offence is that a quite significant rate of growth of non stock-related theft has occurred, accompanied, in the case of burglary and theft, by a declining rate of prosecution and conviction, a consequence which one might expect from increased urbanisation other Aings being equal. It is rather odd, however, that the net effect (not shown' in the Table) is of little variation in absolute numbers summonsed and convicted for this offence, varying little from the median values of 423 and 343 respectively, implying a degree of constancy in this area of the criminal justice system which perhaps warrants further research. The clear-up rate for stock theft is, as already noted, much higher, though the absence of a rapidly rising trend in incidents reported may be due not so much to a decline in the relative importance of this offence as other changes take place in Botswana society as to the spread of effective fencing and other control measures on the ranges. Age and Sentencing Patterns In addition to fines and immediate imprisonment, courts in Botswana also make use of partially and fully suspended sentences of imprisonment. A study of the Magistrates Court Records for Gaborone between 1980 and 1989 showed that in 160 of 287 cases of burglary and theft (55.7%) one of these options was imposed. The youngest adult offenders (aged 18-20) were particularly likely to receive partially suspended sentences (over 52%) while almost 60% of 26-30 year olds were given immediate imprisonment. The picture is summarised for all age groups in Table 4. Table 4: Sentencing Patterns for Burglary and Theft, 1980-89 SENTENCE 18-20 (%) Immediate Imprisonment 31 (33.7) Partially Suspended 48 (52.2) Fully Suspended 13 (14.1) Total 92 (100) Source: Gaborone Magistrates Courts: compiled from Court Records by C Love This Table also shows that most offences were committed by those aged 21 -25 who were also most likely to receive prison-related sentences Not only were younger offenders less likely to get immediate imprisonment, but of those who did, the median length of sentence was on average over the period, shorter, being just over two years compared to over thirty-eight months for the 21-25s. Of course, a 21-25 (%) 60(48.0) 57 (45.6) 8(6.4) 125 (100) Some Observations on Crime in Botswana 39 AGES 26-30 (%) 28 (59.6) 17 (36.2) 2(4.3) 47 (100) 31-35 (%) 36^0 (%) 1 (25.0) 8 (42.1) 3(75.0) 10(52.6) 1 (5.3) 4(100) 19 (100) 40 C&RSLove proportion of these would have been imprisoned as a consequence of re-offending while under a suspended sentence. Even so, there seems to have been a tendency to treat those in their twenties more severely than those younger offenders for whom it is more likely to have been their first offence as an adult The option of probation was not available, though its introduction would be expected to have a greater effect on re-offending rates, for certain offenders, than differential suspended sentences (Love, 1992). It should be noted, however, that this exercise of scanning the records of magistrates courts revealed a large number of gaps and incomplete entries. Consequently such figures may only be indicative at best The vast majority of all offenders were male, averaging around 85% of annual totals throughout the period 1980-92 with no discernible trend. Commentary The evidence from a selected number of offences against the Penal Code, taken from Police Reports and from a survey of Magistrate Courts records in Gaborone, is that the rapid development of the Botswana economy over the period since 1980 has been accompanied by a parallel rise in criminal offences noted. To some extent the trends follow changes in the pattern of economic activity, with stock theft, for instance, responding more to cyclical factors than secular trends and with crimes related to affluence and the distribution of wealth, such as burglary and theft, increasing rapidly. Such a pattern would seem to fit in with experience elsewhere in the world and with common explanations involving the uneven nature of economic development. The latter are often claimed to produce resentment, or to stimulate migration which then leads to a weakening of family bonds and support mechanisms, andalso to create greater opportunities for property crime (UNRISD, Approaches to the understanding of offending behaviour which are based upon family break-up and urban migration, or others which are more anomie-based and structural, tend to have their roots in Western models and to focus on the behavioural characteristics of an individual offender, usually young and male (Gilbert & Gugler, 1981). Since the early 1980s Sumner and others have questioned the appropriateness of these models and have stressed the importance of the local nature of social change, particularly in countries which were formerly colonies. This perspective would, inter alia, take account of changes in the cattle economy of the country, of state subsidies for some activities but not others, of changes in household composition over time due to migration, and of a deeper understanding of how the psychological 'world view' of traditional Batswana life has been changing with the pervasive extension of a market-based economy. At a Some Observations on Crime in Botswana 41 more structural level, links must also be made with attitudes towards actual or perceived misuse of state funds and the abuse of political power, and of how this, where it exists, may be seen as condoning offences against the property of the private individual who will often be a direct or indirect beneficiary either of state income or state regulation. There is a linkage here too with the way in which former colonial regimes were regarded as alien, imposing a set of laws and regulations which were not always welcomed by the indigenous people, and whose advice was never sought (Sumner, 1982; Davis & Slabbert, 1985). Furthermore, in a country like Botswana the concept of 'family breakdown' can only be considered in the context of traditional familial survival strategies. Various surveys have shown the continued importance of migrant labour which remains part of the rural household, through remittances and regular return visits, as part of a pattern of behaviour common throughout Southern Africa (Government of Botswana, 1992; Perrings, et al, 1992; Jones-Dube, 1995). Moreover, of those imprisoned in Botswana in 1993, for instance, over half (51.5%) had been in paid employment before their arrest 31% were educated to Standard Seven or above, and 45.8% were detained in prisons outside the main urban centres of Gaborone, Francistown and Lobatse. Over 85% claimed some form of religious affiliation (Prisons Report, 1993). Furthermore, the prevalence of stock theft and the continued importance of customary courts, which dealt with 70% of trials in 1992, demonstrate that urbanisation per se may not be the important factor it is so frequently made out to be. The error here may be in a false dichotomy between urban and rural and a consequent neglect of peri-urban environments and smaller urban centres serving a rural hinterland. It has also been shown that structural trends such as inequality in the distribution of income as urban concentration increases are not readily predictable (Alperovich, 1995). The relationship between crime and concentration of population in a country like Botswana therefore calls for further research. More fundamental is the social and cultural context of custom and law which both defines criminal behaviour and determines what is deemed to be an 'appropriate' sentence. Amongst other functions judicial systems usually serve to perpetuate inherited privilege, whether that be in the legitimisation of registered land and property ownership in the Dutch-Roman system or in the sorts of issues decided in customary courts where the influence of traditional chiefs frequently remains strong. The tension evident in social theory between structure and agency consequently permeates any analysis of offender behaviour and its consequences where the individual in his or her 'deviance' and the judiciary in their 'rectitude' make personal decisions based upon external reference points of a long term structural nature. In this context 'urbanisation' is merely a spatial parameter which may make certain types of offence more or less likely but is not in itself a determining factor. 42 C&RSLove Conclusion There is no doubt that distinctive and predictable trends are evident in Botswana, but a proper understanding of their causes and relative growth depends at least as much on an accurate specification of the local nature of social change as on the ability to fit an externally introduced model derived from Western experience. Thus, any explanation based upon 'urbanisation' or even superficially upon aspects of'modernisation' will be inadequate and misleading in its implications for policy. A study of 'Offences against the Penal Code' therefore leads to consideration of issues at the heart of the 'development' debate in which behavioural, structural, and single discipline approaches by themselves are likely to provide misleading, or at best only limited, understanding. Unless all such factors are taken into account the consequences in the longer term on the relevance and effectiveness of official measures of 'social control,' such as imprisonment and corporal punishment, as well as on the most appropriate directions for policy on housing, education, and social support systems, are likely to be negative. References Alperovich, G (1995) "The Relationship between Income Inequality and City Size: A General Equilibrium Model of an Open System of Cities Approach," in Urban Studies, 32,6. Bank of Botswana (1994) Annual Report 1993, Gaborone. Barclays Botswana and Department of Economics, Economic Review (various issues 1991-92), University of Botswana, Gaborone. Davis, D & Slabbert, M (eds) (1985) Crime and Power in South Africa, David Gilbert, A & Gugler, J (1981) Cities, Poverty and Development: Urbanisation Government of Botswana Department of Prisons, Annual Report (various years), Philip, Cape Town. in the Third World, Oxford University Press, Oxford. Gaborone. Government of Botswana, Police Department, Annual Report (various years), Gaborone. Jones-Dube, E (1995) "Non-Metropolitan Migration in Botswana with and Emphasis on Gender,' in J Baker, & Tade Akin Ama The Migration Experience in Africa, Nordiska Afrikainstitutet, Uppsala, Sweden. Love, C (19920 "Court Sentencing in Botswana: a Role for Probation?," in Journal of Social Development in Africa, Vol 7, No 2, pp 5-17. Mihyo, P M - Book review of Cutshall, C R "Justice for the People: Community Courts and Legal Transformation in Zimbabwe," in Development and Change, Vol 26, No 3, July 1995. Perrings, C; Gyekye, A; Love, R; Osafo-Gyimah, K (1992) Income and Employment Generation in Botswana, Report, University of Botswana, Gaborone. Sumner.C (1982) Crime, Justice and Underdevelopment, London. United Nations (1994) Social Development in the Third World, United Nations Research Institute for Social Development (UNRISD), Rome.