Crime, police and public in transitional societies
Mark Shaw considers the nature of crime in societies in transition, and focuses on the causal factors for increased levels of crime during periods of political, economic and social change. He argues that evidence that the fight against crime in transitional societies is being won is far from conclusive, and a series of policy dilemmas that confront such societies are presented.
It has often been said that crime in South Africa is related to the transition from authoritarian rule to democracy. Similar patterns appear to apply in other transitional societies such as those which have moved from military to civilian rule in Latin America, the former communist regimes of Eastern and Central Europe; as well as states in Africa, most clearly illustrated by Nigeria and Mozambique, which have moved from authoritarian forms of rule or civil war to fledgling democracies. While diverse, a review of the experiences of these societies holds important lessons and highlights the importance of co-operation between them. While comparative analysis of policing and crime control systems has been little practiced (and very often derided), such a cross-national perspective holds important lessons in understanding not only the unique features of each society but also the degree to which similar processes of political, economic or social change produce similar outcomes in terms of crime levels and problems of police reform (Bayley 1999).
At the outset it is worth noting that, almost without exception, the growth of crime (or, as is argued below, its greater visibility) in the post-transition period caught new governments by surprise. The ending of military rule in Latin America, the fall of Communism and the ending of apartheid in South Africa brought great optimism for the future, not fear that growing levels of criminal activity would hold out the prospect that the transitions could themselves be subverted. Steps to counter crime, therefore, were in most cases taken too late because the full extent of the problem was underestimated as newly democratic governments battled with the tasks of transformation and institution building in the immediate post-transition environment. [End Page 1]
To be fair to those that assumed power in these new democracies, however, the link between political transition and crime is more complex than often assumed, and while levels of certain types of crime may grow, other forms of criminality may not increase but instead become more visible in the public eye in the post-transition period. This at least suggests that the automatic link often made between the occurrence of political transition and the growth in crime should not necessarily be assumed. What seems clear however is that apart from generating particular forms of criminality (most notably the organised variety) transitions also have important impacts on the social controls present in any society. In many instances a weakening of these provides an environment that may be more conducive to criminality.
While weakening and changing forms of social control are important in understanding why transitional societies are more conducive to criminality, the focus of newly democratic governments is seldom on these issues. Instead, the concentration is generally on attempting to reform the agencies of law enforcement; in the first instance to make them more acceptable to ordinary citizens used to repressive forms of policing, and secondly, to ensure that they can fight crime. In neither case has any transitional society been entirely successful, partly because law enforcement agencies like the police cannot confront factors such as weakening or changing forms of social control which appear to allow higher levels of crime, and partly because police transformation has become bogged down by the difficulties of reforming essentially authoritarian instruments of control in the context of often acute skills and resource shortages.
Public opinion on issues of crime in transitional societies is also critical to factor into this equation. Pressure from citizens in transitional societies is often (ironically since these were the same individuals policed under authoritarian systems before) aimed at increasing the powers of the police, urging more wide-scale crackdowns on crime. The result is in many societies a return to aspects of authoritarian policing in the name of crime control and a continued use of the military for purposes of internal social ordering. Also seldom considered, but of some importance, is the role of foreign assistance to police agencies in transitional societies. Such programmes often result in a focus on issues such as fighting organised crime which diverts attention from other areas of critical importance, as well as ensuring police agencies which may increasingly look like those in advanced democracies but which are seldom applicable for the conditions present in transitional societies. [End Page 2]
All of these issues provide newly democratic societies with a number of key policy dilemmas. These include five key issues consistent across most transitional societies. First, balancing the requirements of transformation of authoritarian police agencies against the more immediate requirement to fight crime. Second, balancing the devolution of policing authority to ensure more local accountability against retaining centralised control to ensure a uniformed process of change. Third, balancing hard won civil liberties against the requirement to fight lawlessness more effectively. Fourth, structuring the relationship between private and public forms of policing. Finally, balancing the requirement to invest in social policies which prevent crime in the longer term, against the short term imperative to control burgeoning criminal activity.
The paper first considers the links between transitions and crime, then turns to the question of changing social controls in transitional societies. The problems of police transformation are then examined as well as the impact that public pressure has had on crime control policies in transitional societies. Finally, the key policy dilemmas faced by new democratic governments in transitional societies are briefly outlined.
Linking transitions to crime
The rise of crime in periods of political (and related economic and social) transition is a complex phenomenon and is difficult to analyse. Statistical data on levels of crime before and after the transition are difficult to come by, and when available, their accuracy may be open to question. In South Africa, for example, there are no reliable crime statistics for the whole country before January 1994. At the same time, our understanding of crime in authoritarian societies is often complicated by the fact that state repression led to a blurring of the boundaries of political and criminal activity and the state itself was often a significant source (although not defined as such at the time) of criminal activity.
These issues raise important questions about whether or not dramatic transitions lead to the more visible appearance of older forms of criminal activity in new guises or whether transitions themselves give rise to new forms of criminal activity. In South Africa, for example, is the increase in crime predominantly the result of a displacement of criminal activity previously contained in townships and which has now spread to (white) suburban areas? Have levels of rape, for example, not always been high but now because they are reported (and viewed with concern by both government [End Page 3] and citizens) it appears that the transition has brought with it remarkably high levels of sexual violence?
The answer lies somewhere between the two positions. Old forms of criminal activity have undoubtedly been displaced (often in new forms) into the new democratic order, while at the same time they have been joined by growth in the overall level of all forms of criminality. Of course, this still leaves the question of why a shift from authoritarian rule to democracy should lead to higher levels of crime, and if so, how?
In order to answer the question it is necessary to begin by examining levels of crime in these societies before the transitions to democracy began. As already suggested this constitutes a serious research challenge, not only due to the lack of data on levels of crime in authoritarian societies, but also because authoritarian rule distorted criminal activities, often using criminals to achieve political goals. Nevertheless, a review of the available evidence in a range of transitional societies suggests that in no case (although this is often contrary to post-transition public opinion) was criminal activity a new phenomenon, although in some cases questions of what was considered 'criminal' and what was not, was a matter of legal definition. But, it is safe to say that while the transition from authoritarian rule to democracy was a watershed moment in the history of all the countries concerned, the roots of criminality were bred before the transition began, were magnified in the period immediately before dramatic shifts in political power occurred, and then became highly visible in the post-transition period.
The power of the mafia in post-Communist Russia, for example, is now widely acknowledged, the common perception being that the establishment of these criminal groups was a direct result of the fall of the Soviet Union. This is in fact not the case. A careful reading of the available literature on the period suggests that in the last 20 years of Soviet power, organised crime had become a silent partner in the black-market economy. When Mikhail Gorbachev came to power in 1985, most major Russian cities already had powerful organised gangs. Their cohesiveness and wealth enabled them to survive the collapse of the old state, and to profit from the disarray of the new one (Handelman 1995: 9). This is not to deny, of course, that massive and often illegitimate fortunes were made when the assets of the Soviet state were 'privatised' into the hands of a few well-connected and powerful oligarchs (Castells 1998:180-90), but rather that organised criminal activity did not simply originate in the period of political transition, although criminal networks undoubtedly took advantage of the chaos and uncertainty characteristic of the time. [End Page 4]
Parallel conclusions can be drawn for other societies. In both the case of Brazil and South Africa, there have been relatively high levels of interpersonal violent crime. Indeed, Brazil and South Africa display remarkably similar crime patterns. The nature, scale and distribution of crime are similar in the two states. While the wealthy are often victims of property crime, the vast majority of violent crime is concentrated on the periphery of the large cities where the problems of unemployment, lack of basic housing and social services (including policing) are acute (Neto 2001).
In neither the case of Brazil nor South Africa is the problem of violent crime new. In South Africa, that data which does exist shows increases in violent crime such as homicide - even excluding deaths as a result of political violence - from the early to the mid-1980s (Shaw 2002). Homicide has subsequently been in slow decline since the attainment of democracy in 1994. Crime such as assault and rape have been stable or show slow increases since 1994 in the reported and consolidated national crime figures. The extent of these crimes have probably not changed in any fundamental way. It would defy logic to suggest that the serious problem of assault and related inter-personal violence in the Western and Northern Cape, which is related closely to patterns of excessive alcohol consumption in the area, have suddenly increased out of all proportion after 1994. Instead, it is likely that they are more visible and probably more likely to be reported.
This conclusion - that there were already high levels of crime in transitional societies before the attainment of democracy - cannot on its own, however, provide a satisfactory explanation. Clearly, on examination of the available data, some categories of crime have increased during the transitional period. Some effort must be made to desegregate different types of crime relating any increases to the transition period. A review of evidence from across a number of transitional societies suggests that increases in crime in relation to periods of political transition are often a factor of two critical and inter-related issues. The first is the availability of firearms and the second is the increased organisation of some aspects of criminal activity.
There is much debate about the link between firearm availability and crime. A review of the societies in question suggests that the more guns are available on the legal market, the more easily they are transferred into criminal usage. This is a particularly serious problem in Latin and Central America and Southern Africa. In South Africa, for example, there is a [End Page 5] direct correlation between the number of firearms lost and stolen in 1997 and 1998 and the increase in armed robbery in the country during that period (see Shaw 2002). Similar patterns can also be discerned for Latin America.
A related factor is of interest here. It is possible that in transitions where a high level of violence has been a dominant feature of the society in the pre-transitional period, and where the nature of the peace settlement in the society is effectively a stalemate, people on both sides of the divide either retain their arms because they do not trust the opposition, or buy new ones because they fear the future. These weapons have the potential to re-circulate on the illegal market. Add to this individuals on both sides who are demobilised but with few opportunities for employment as well as weak law enforcement, and a recipe for increased crime is the result. Thus, conflictual transitions which end through negotiated peace deals when conflicts stalemate provide a particular opportunity for surges in violent crime (Call 1999). The parallels with South Africa are difficult to miss.
While the availability of firearms and their impact on crime takes similar forms across the societies in question, the same cannot be said of organised crime. In fact, while the label 'organised crime' is used to describe particular types of criminal activity in post-transition societies these criminal arrangements are often very different depending on their organisational histories and the context in which they operate.
For these reasons organised criminal activity in the former Soviet Union is different in a number of important ways from that in Southern Africa. This is because two key conditions for the growth of organised mafias were present in the Soviet Union: excessive bureaucratic power and illegal or parallel markets. Thus, before the collapse of the Soviet Union criminal organisations competed (as part of an underground economy) with the state. After the collapse of the Soviet Union, they dominated the newly privatised structures of the state itself. As a result such groups are often hierarchically organised and rely on extortion and other means to dominate different arenas of legitimate and business activity.
In Southern and West Africa and Latin America (with the exception of Columbia which has a particular history of organised crime), organised criminal activity is more accurately described as a criminal network (see Williams 2001). There is no doubt that the crime is organised, but the interrelationships between various individuals engaged in the criminal enterprise is more dynamic and fluid. These aim largely at the movement and sale of [End Page 6] illegal goods including narcotics. There is a cross-over between the types of organised crime, but the fundamental organising principles of each should be recognised.
The above discussion suggests that the links between crime and political, economic or social transitions are much more complex than a direct link between the process of transition and any increases in crime which accompanies it. Critical to understanding the phenomenon is the requirement of effectively analysing the nature of crime in the pre-transition period, separating these from those types of crime generated by the transition itself. Perhaps most notably in the case of South Africa, it is likely that the country had a high (although hidden) level of inter-personal violence before the transition began, onto which was grafted increases in additional crime categories such as violent robbery, more directly related to factors associated with the transition (Shaw and Gastrow 2001). That goes some way in explaining the overall high levels of crime in the post-transition period in the country. Similar patterns are found, although with different categories of crime, in other societies in transition.
A series of additional factors relating to forms of social control in transitional societies should also be considered. While these do not in themselves cause crime, comparative experience from across transitional societies suggests that they at the very least provide an environment that is conducive to criminality.
Changing social controls and crime control
Ironically, and contrary to popular belief, increases in crime in transitional societies may be less an issue of declining levels of law enforcement and policing than is commonly assumed. A comparison of authoritarian regimes across countries suggests that most citizens were policed as much for crime control as for political control. Yet it is not certain whether policing in most post-authoritarian states has improved in the eyes of the majority of citizens, and if it has, whether this is not because the police are more effective in controlling crime, but rather because they are now less repressive and thereby less intrusive.
An overview of the growth of crime in a number of transitional societies suggests a more complex reason for the growth in crime: the breakdown of community and related principles of social organisation, including the crime control arrangements and reduced risks for punishment, as well as an increase in opportunities, targets and motivation. Thus, dramatic political, [End Page 7] economic and social transitions may be much more disruptive of the internal social organisation (including that of crime prevention and control) of communities than has often been assumed. Three forms of internal social organisation may be dramatically altered in a period of transition.
First, in societies such as in South Africa, the struggle against an authoritarian state produces opposing forms of community cohesion and social control, which keeps criminality in check. In Northern Ireland, for example, criminal activities have been restricted given the vulnerability of offenders (who are threatened with prosecution if they do not agree) to being recruited as informers by the police (Brewer 2001).
Second, in communist countries with centralised political structures such as in Eastern and Central Europe and the former Soviet Union the state itself imposed the organisational network. The collapse of the communist state led to a breakdown of these structures, fragmenting local forms of social cohesion without any immediate replacement (Kvashis and Babaev 2001).
Third, in most societies, quite apart from structures established in a response to, or by, an authoritarian state, some form of community controls remained. These include structures such as the church, community groups, the extended family and neighbourhood groups. A review of societies that have undergone dramatic transitions suggests that these structures are weakened and lose their reach into the community (see Bayley 1985:7). This is a result of the strength of the two new forms of community cohesion outlined above and which are generated by authoritarian states. But it is also a consequence of the disruptive nature of transitions and the violence that often accompanies them, weakening old forms of social organisation which no longer provide an attractive option for increasingly militarised and vocal sectors of the society, often the youth (for the particular case of South Africa, see Marks 2001). In addition, traditional forms of social control are undercut during periods of transition by the emergence of new social movements and non-governmental organisations.
The breakdown of social and state controls is an important factor creating environments more conducive to crime in diverse transitional societies. Traditional forms of internal social cohesion may be replaced by (or mutate into) a different set of organising principles, including criminal organisations or gangs. In communities feeling threatened by a growing group of criminals this may also take the form of vigilante groups which come to play an important role in local community cohesion. [End Page 8]
Changes brought about by the dramatic impact of the political transition are exacerbated by longer term processes of industrialisation and urbanisation which have themselves had a considerable impact on the changing nature of community and social controls. In a post-apartheid society, the effects of HIV/AIDS may already be having a considerable impact, both on family units (through the death of parents and breadwinners) and organisation and on cognitive behaviour (producing people that have little to lose through offending) which may have the potential to impact upon crime rates.
Post-authoritarian and post-conflict societies are also increasingly subject to structural changes in their economies. While a number of approaches are followed in this regard, transitional societies tend to share the experience of changes in the ownership structure (privatisation), the multiplication of economic actors and the influences of globalisation. Many transitional societies also seek to redefine the role of the state by reducing or altering its role in economic activity. In such states, the access to newly created opportunities is not equal for all. This factor, combined with the requirement for political legitimacy and the need to attempt to meet popular expectations, creates contradictory pressures. In societies such as South Africa, these pressures are complicated by a political and economic commitment to improve the lot of previously disadvantaged groups.
Such frictions are common to transitional societies. There appears to be an abundance of structural as well as other motivational factors for the involvement of people in 'alternative opportunities' both in the context of the growth of the informal and the criminal economy. Such recourse to the illicit is facilitated by the breakdown and loosening of the mechanisms of formal and social control, including the diminished risks of punitive and/ or resocialising reactions. If, over time, poverty and marginalisation are perceived as a likely reality, the recourse to the illicit is often (particularly for the young) perceived as the most efficient and low risk avenue to live better now rather than wait for the uncertain prospects for improvement promised by the state. This is not to argue that poverty causes crime, but that increasing inequalities in wealth, in a context where the poor (and particular those who are young and poor) have high expectations that this position will change in future, serves as a potentially important causal factor for criminal activity. No amount of political rhetoric about building a new democratic society (except perhaps at the initial stages of collective enthusiasm) matches the economic reality of unequal access to the new [End Page 9] opportunities for wealth. Although these observations are hardly new, they are important in understanding the growth of criminal activity in transitional societies.
New challenges - old instrument
While, as the above discussion outlines, transitional societies are faced with growing levels of crime in the period in which they attempt to consolidate democratic governance, they are seldom immediately equipped with the policing instruments to fight it. 'An efficient police in one age', notes David Bayley, 'is an irrelevant police in another. Crime is a function of social values, hence so is police efficiency' (1975:331). Police agencies inherited from the old political order were designed much more for the control of political opponents than crime prevention. In every case transitional societies face significant problems of reforming the agencies of law enforcement to come to terms with the new democratic context.
A review of ongoing attempts at police reform suggests that results in this area are mixed, with the difficulties of establishing new systems of policing or changing the old, being compounded by the immediate tasks of confronting high levels of criminality. Despite these challenges, focus has generally been, and particularly in Latin America, on the role of militaries in the transitional progress. To some extent there has also been a focus on the reform of intelligence agencies, although this often deals with the information they hold on individuals more than the ins and outs of any transformation process. The same however cannot be said of the police (Tanner 2000).
Why have the police in particular received so little attention? That may be in part because police officers (certainly those that walk the beat, as opposed to those involved in the more political aspects of policing) are generally considered to be the poor cousins of the better equipped and more powerful military. But in part it also represents a prevailing view that police agencies have not had much impact on the way in which transitions have been managed.
While the critical issue of police reform has been dealt with, it is usually in the context of what the ideal post-authoritarian police agency should look like, and some suggestions (often simplistic) of how to get there. There has however been surprisingly little comparative study of the nature of police reform efforts across various societies that have moved from authoritarian rule to democracy. Yet the issue is of some urgency. While the military can be confined to the barracks (although this is not of course [End Page 10] always the case) and the power of intelligence bureaucracies reduced (again, not always achieved) the police in newly democratic states are likely to have more, not less, interaction with the community. This is particularly the case when levels of crime increase, or at the very least become more visible and thus of greater public concern.
Ensuring an effective system of policing in any new democracy forms the cornerstone of the new order. Enhanced feelings of safety and the symbolic presence and engagement of those who are tasked to protect and serve, rather than enforce unpopular laws, are a powerful reminder that democracy has arrived, albeit tentatively. It is little wonder that symbolic changes such as introducing new uniforms and designing new badges attract such attention and often heated debate. For example, and for a particular set of reasons, the debate over new policing symbols is particularly acute in the Northern Ireland peace process (see McGarry and O'Leary 1999). At the heart of these engagements lies the question of who actually 'owns' the police and who are they structured and paid to serve? This lies at the heart of disagreement around how police reform should be structured and controlled. There is however a paradox at the centre of this process; that is, if the public mood turns against policing reform, not because citizens are uncomfortable with the democracy in principle, but rather that they believe that it does little to uphold their safety. Ironically, given that some citizens long for a return to the security of the past, the legacy of past policing hangs heavy on attempts at reform.
Under authoritarian rule, the police were used to control protest, hunt down and harass those in opposition to the regime. Generally, however, the police in authoritarian regimes did attempt to control some aspects of criminal behaviour, although in most cases there was a grey area between what constituted 'crime' and what was the stuff of 'politics'. Thus, opposition to authoritarian rule was criminalised, while criminals were considered unpatriotic, and thus crime was politicised. In Soviet society, for example, ordinary crimes such as hooliganism, theft or murder were labelled as political offences ('enemies of the people' or 'individual lapses of socialist construction') with consequently higher penalties (Clark 1993:14). In Latin America military regimes relied on the doctrine of an 'internal ideological enemy' which blurred the distinctions between common criminals, guerrillas, subversives, political opposition, the urban underclass and labour organisations (Palmieri 2000). In specific cases, however, the authorities overlooked 'criminal activity' because it is directly or indirectly [End Page 11] beneficial to the state. In the Soviet Union, enforcement agencies 'looked the other way regarding illegal activities considered by the higher leadership to be beneficial to the maintenance of economic production and efficiency' (Clark 1993:91, also Kramer 1977). In apartheid South Africa police encouraged and supported criminal gangs in harassing political opponents of the state (see Schärf and Nina 2001).
This blurring between legal and illegal activities and the wide discretion that authoritarian law enforcement agencies had in determining what would and would not be allowed - the test being always whether it was of benefit to regime maintenance - has had important implications in new democracies. In some cases it has resulted in police agencies with too close connections to criminal groups; the police's relationship with selected gangs in South Africa being a case in point. More importantly it has bred a suspicion for the law and law enforcement agencies, both of which are seen not as impartial and predictable instruments of the state, but as political instruments.
Critically, policing authoritarian and/or divided societies relied more on fear than on building good relations with communities. Policing, particularly of those who opposed the state or people identified as undesirable, was harsh and brutal. There was seldom any due process of law with the important result that police agencies had little tradition of investigating, ordering or presenting evidence to convict offenders in an impartial court of law. As a report on current problems with investigative functions in Latin American police agencies noted: 'Criminal investigations were rendered dysfunctional by governments who did not need proof to eliminate their "enemies" nor want evidence of atrocities or illegal behaviour committed by their "friends"' (Palmieri 2000:2). That statement could as well apply to almost all states with an authoritarian history of policing. Even when court processes were convened they were often charades which were highly unlikely to lead to an acquittal. In all new democracies, the problems of improving the investigative function of the police, and the danger of reverting to old style methods such as torture, has been one of the key stumbling blocks to building democratic police agencies. In Russia currently, for example, acquittal rates are well under one per cent, not because cases are well prepared or investigated but because prosecutors wield enormous power and the culture of the court room differs little from the past, accusation and guilt amounting to almost the same thing (The Economist, July 21, 2001:13). [End Page 12]
Such issues provide an indication of the challenges confronting any police reform effort and also suggest how ill-equipped former authoritarian police agencies are in fighting crime in the context of a democracy (see for example, Shelley 1999). Bayley in a seminal study of the police and political development concludes that: 'Police systems exhibit an enormous inertial strength over time; their forms endure even across the divides of war, violent revolution, and shattering economic and social change' (1975:370). This sobering assessment suggests that the impact of police reform efforts across transitional societies in Latin America, Eastern Europe and Africa which have established democratic systems in the last two decades may be much more limited than many of its advocates would be prepared to admit. The reality too is also that even were police reform to be all encompassing, policing on its own has little chance of undercutting those causes that make transitional societies conducive to criminality.
Police reform and crime control
One key and immediate challenge for those who have assumed democratic control in transitional societies is to secure their hold on the levers of the security establishment. Here there are multiple challenges. The most common across all transitional societies is the desire to legitimate the old forces of order to ensure citizens look to the police for their safety. What appears clear is that no amount of urging by political leaders that the police are now legitimate is enough; agencies of law enforcement have to prove their legitimacy through effective operation, often through direct engagement with citizens on the ground. The growth of crime in transitional societies, however, has in many cases undercut the development of local forms of policing by ensuring more centralised and militarised responses to disorder.
Key to building the legitimacy of the police is to ensure effective forms of local control and accountability - in effect, to make citizens believe that the police are responsive to their needs, and not those of some bureaucrat in a distant capital. Here all transitional societies have had to balance the requirement of ensuring local accountability (which remains weak in all cases) with centralised control - the desire to manage change from the centre to ensure both that it occurs uniformly and that local groups (who may oppose the central state) do not obtain control of the police in their area.
The absence of social and community controls and the establishment of a democracy bring paradoxical forces into play in most transitional societies. [End Page 13] On one hand, the conditions for the growth of crime are enhanced, on the other, citizens look (as they have never done before) to the state for protection. Given the very real constraints on the post-transition state in delivering effective systems of criminal justice (such as low skills levels, lack of representative institutions and poor resourcing) citizens are likely, over time, to seek alternative forms of protection. For the poor community this will include forms of protection such as vigilante groups and for the wealthy (including the business sector), the increased privatisation of policing and crime prevention.
The parallels amongst transitional societies are striking. Probably the most effective means of controlling and preventing crime in the longer term is the one least open to the state: the re-establishment of effective means of community and social control. Key to the process is both the establishment of effective local systems of democracy through which people can exercise their rights and express their grievances as well as the support of institutions such as churches, schools, sport and youth activities that assist in the building of stronger and more cohesive communities. The difficulty of implementing such projects is great, since there is the added problem that such initiatives are often difficult to link to reductions in crime in the short term, thus being harder to sell to policy makers.
Comparative experience suggests also that while the state is good at breaking down forms of local social control and cohesion it is notoriously bad at reconstructing these. What is clear, however, is that a concentration of improved law enforcement alone (however necessary) will not stem the long-term crime problems of states emerging from periods of transition. This suggests that the implementation of crime prevention projects as understood in the developed world (and often marketed in transitional and developing societies) is not the most appropriate route. Instead, comparative experience indicates that much greater debate and effort is required to seek alternative ways of rebuilding the 'social fabric' in post-conflict and post-authoritarian societies.
The danger, however, in societies in transition is that responses to crime become increasingly militarised. In many post-authoritarian societies this results from the dual pressures of increased public insistence on government to be seen to act against lawlessness as well as pressures from within the security establishment. In respect of the latter, policing organisations which have undergone dramatic processes of transformation seek security in operations which they know and are comfortable with, and in any event [End Page 14] may have been urging as an appropriate response to crime. The dangers of such approaches in post-authoritarian states is that important gains in respect of the protection of human rights may be undercut over time. Such militarised responses to crime control, while they may often be sold as such, should not be seen as the same as problem solving and/or saturation policing in more advanced democracies. The case of the decline in crime in New York in the recent past and the policing approaches used in this respect, are, despite inconclusive evidence as to their success, drawn upon in a surprising number of post-authoritarian states as potential solutions to domestic crime problems. Yet implementation in often fragile democracies carries the danger of a return (or at least perceptions to this effect) to authoritarianism.
Some distinction should be made, however, between increasingly militarised responses to crime and more specialised responses such as the establishment of high profile national investigative units. Given the complexity and sophistication of criminal organisations, such units are an essential addition to the tools available to the government to fight crime. The establishment of such units highlights an additional problem for many countries moving from authoritarian rule to democracy: poor co-operation between law enforcement personnel and prosecutors. Given that one of the outcomes of democratic policing is the presentation of evidence before an impartial court of law, improving these mechanisms is essential to the long-term success of post-authoritarian policing.
Citizen responses, non-state policing and external impacts
It has already been suggested that building stronger links with the citizenry is an important requirement of legitimising the police in post-authoritarian societies. In the majority of transitional societies, however, the police are viewed as ineffective by the citizenry. There are clearly exceptions to this and in South Africa, for example, data suggests that while the public appear willing to work with the police, little or no advantage is being taken to build better community confidence in policing by the police through the provision of better levels of service delivery (Louw 2001).
Overall, public dissatisfaction with the police and levels of crime in transitional societies is widespread. High levels of insecurity and perceptions that state responses in this regard were weak ensure that for many citizens such issues are key to their judgement of the success of democracy as a whole. For example, a recent survey of a series of Latin American countries showed low or falling levels of support for democracy in such transitional [End Page 15] societies such as Brazil, Argentina and Chile. The survey found that four out five respondents said that both crime and drug addiction has 'increased a lot' in their countries in the past three years, up from around 65 per cent when the same question was asked in 1995. Citizens also showed low levels of confidence in the capability of institutions of policing and the judiciary to solve these problems (The Economist, July 28, 2001:49-50).
Such finding are replicated elsewhere. In South Africa a series of surveys have shown crime to be one of the key issues undercutting confidence in the new order, although whites and blacks viewed the problem from diametrically opposite positions: blacks seeing the problem of the failure of the new order to consolidate, whites seeing the cause of lawlessness as the breakdown of the instruments of control of the old regime (Shaw 2002). In Russia too public dissatisfaction with high levels of crime has prompted calls for a return to harsher forms of law enforcement with, as one study concluded, the Russian people considering 'their security from criminals to be far more important than their procedural rights' (Knight 1996:98). Overall cross-national survey evidence from a number of countries in transition in Africa, Latin America, Eastern Europe show low levels of satisfaction with police performance, as well as higher public support for harsher measures to counter criminal activity (Alvazzi Del Frate 1998, Zvekic 1998).
In many societies citizens have begun to take their own initiatives against crime. Most commonly this takes two forms - the growth of the private security industry and the emergence of vigilantism. Again, the degree to which this manifests itself in any society is dependent on its history and traditions. In Eastern and Central Europe and the former Soviet Union cases of vigilante type activities appear to be less well developed given a historic over-reliance on the state under communism for the delivery of services. While citizens are dissatisfied with the general level of service provision by the criminal justice system they remain reliant on the state. In countries in Africa and Latin America where communities have, often in response to authoritarian rule, sought community responses to ensure local safety, vigilante groups in various guises have become reasonably common. State responses to these vigilante groups range from attempts to co-opt them to direct attacks upon them. It is also instructive to note that vigilante groups themselves in some countries were likely over time, and as they developed dominance in any particular area, to become involved in the illegal accumulation of resources to ensure their own survival. [End Page 16]
The development of relatively sophisticated private security industries in the majority of transitional societies should be viewed with concern. In most cases the employment of private security is designed to protect the rich from the poor, perpetuating old divisions or extending divisions within the societies along class lines. The cross-over between the security agencies of the state and private security companies is a worrying development. This occurs through police officers or other state security officials moonlighting as security guards, despite the fact that this was illegal in most transitional societies. Or, because former members of the security establishment who had left to establish or work in private security operations retained some links to the state's security forces. The existence of active private security sectors is a reality that governments will have to accept. Ironically, effective regulation of the private security sector requires law enforcement capacity; precisely the reason why private security operations were in a position to exploit the market in the first place. In both the case of vigilantism and private security it appears that the issue is now less how these security instruments that competed with the state could be eradicated, but rather how they should be managed.
Another little considered dimension of external influence on police reform efforts is the question of foreign assistance. One reason for the high level of foreign technical assistance in the area of police reform is the lack of civilian experts on security issues in many post-authoritarian states (Neild 2001:27). Many transitional societies have had quite close links with foreign funding organisations and the law enforcement agencies of foreign governments (particularly the United States) who are assisting in both the reform process and the fight against crime. These interventions have not been uncritically accepted in the recipient countries. There is some consensus amongst policy makers on the receiving end that many assistance programmes were not designed with the recipient country in mind and that assistance such as training was often offered simply because it was available rather than because it was relevant to the needs of transitional societies. In particular, the focus of the United States on issues of trans-national organised crime and drugs (and now on terror) has the ability to distort local law enforcement agencies and focus scarce resource on areas that, while of concern to foreigners, were of little value to the immediate safety needs of the local population (Nadelmann 1997).
Extensive foreign training programmes might have two inter-connected impacts. The first is to undermine the development of innovative local [End Page 17] responses to crime problems, given that foreign assistance often points to particular sets of solutions, despite the fact that these were developed in other more developed societies. Second, while there were some advantages in this regard, law enforcement and criminal justice agencies in transitional societies could increasingly begin to look the same, drawing on similar training programmes and heavily conditioned by foreign experience (especially that of developed democracies). While this holds out the possibility of better co-operation between agencies who would be in a position to understand their counterpart's actions and systems, it also might mean that local initiatives will not be encouraged.
Regional responses to problems of criminality are essential but, in the main and with the possible exception of Southern Africa, are reasonably under-developed. The development of mechanisms of regional co-operation are often, although not always, dependent on the strength of political cooperation amongst states. Regional co-operation holds out the prospect of improving the ability of countries to prevent crime and combat criminality by ensuring that neighbouring 'safe havens' were eliminated or made less attractive for criminals. While an international convention against transnational organised crime has been agreed upon at the 10th United Nations Congress on the 'Prevention of Crime and the Treatment of Offenders', once transitional countries sign the convention, in practice many will have limited capacity to implement it.
These are sobering conclusions given the importance of crime control and police reform in transitional societies. What is certain is that the public police in transitional societies are now only one player in an increasingly complex system of security maintenance, including now local, national, regional and international players. Public pressure on safety issues too is unlikely to recede as long as citizens feel insecure and so pressure for harder and short term responses to crime are likely to outweigh any investments in long term crime prevention initiatives. The concluding section examines these and other dilemmas of achieving safety in new democracies.
Conclusion: policy dilemmas
The above review of the challenge of reducing high levels of crime highlights the importance of social responses to crime control in transitional societies as well as the critical necessity of achieving satisfactory police reform. While the challenges of facing high levels of crime in newly [End Page 18] democratic states are many, five key and inter-related policy dilemmas can be isolated.
First, the requirement to balance efforts at police transformation which by their nature require substantial (and often disruptive) change, against the requirement to fight crime, often using the law enforcement instruments of the old order to achieve this. While the authoritarian police instruments inherited by the old order often lack public trust as well as the skills to fight crime, they are generally the only police available to confront the immediate requirements to tackle lawlessness. The result in most transitional societies is a mix of old and new instruments aimed at fighting crime. The balance is critical here as such a mix leads to conflicts over turf and new forms of competition (often based on political access) that are not necessarily conducive to effective law enforcement. The conflict between the police and the new investigative unit 'The Scorpions' in South Africa is a case in point. On the whole however, a review of police instruments in new democracies suggests that in most cases the heavy hand of authoritarian policing has remained unchanged and indeed is difficult to shake off. Policing remains centrally controlled, often paramilitary in nature and with continued problems of community relations.
Second, and related to the above issue, is the difficult balance between centralised control of the police and the requirement for greater devolution of policing to ensure citizen input at local level. Authoritarian policing is generally centrally controlled with little citizen input. In most new democracies (as in South Africa) splitting up police agencies into separate regional or local components was not considered an option, not only because of the high level of disruption that this would have caused but also because of the political requirement at the time to retain centralised control of the police transformation process. While often necessary in the immediate aftermath of the achievement of democracy the result has been the retention of centralised systems of policing with few means at local level in which citizens can impact upon the way in which their area is policed. Achieving greater accountability of the police at local level in post-authoritarian societies remains a key policy challenge.
Third, across almost all transitional societies there have emerged important debates about the effectiveness of democracy itself in fighting crime. This has often led to questioning of whether civil liberties should be curtailed in order to ensure that the state is more effective in countering criminality. Such debates are often driven by public outcries around crime [End Page 19] (particularly, but not exclusively, from the middle classes or those who previously benefited from authoritarian rule) and generally entail suggestions as to how law can be altered so that - and the phrase is used across many transitional societies - 'criminals do not have more rights than victims'. In many cases however the debates are initiated by particular incidents (terrorism in the Western Cape being a good example in South Africa) that then spread to other aspects of law enforcement. The role of the police in such debates is important.
Fourth, in most transitional societies there is some debate about the balance between long and short-term responses to crime. While initially there appeared to be some sympathy with approaches that suggested that more money should be invested in social programmes which would hopefully undercut crime in the longer term, in almost all cases the emphasis now falls on harder and more immediate law enforcement responses to the problem. The rhetoric of 'crisis' and 'war' are now common across transitional societies as political leaders and security officials seek to respond to public concerns about criminality. As is suggested earlier, however, such approaches are unlikely in the longer term to reduce crime levels significantly, given that they do little to undercut causal factors for criminality. What is clear now is that the emphasis is almost exclusively on law enforcement, partly because earlier attempts at undercutting the social causes of crime were seen as too long term with immediate progress being difficult to ascertain.
Fifth, all transitional societies have seen a remarkable growth in non-state forms of policing such as vigilante activity and private security. In all cases such responses are now accepted as a relatively permanent feature of these societies, giving rise to debates as to how public, private and citizen initiated forms of policing should inter-act. The potential for conflict here is manifest. In particular vigilante type activities often arise under or assume later a political guise, seldom making them neutral arbiters of law and order. At the same time the growth of private security often allows the reinforcement of the divide between the rich and poor in states with high levels of economic inequality. The inter-relationship between these three forms of policing (which may often involve the same individuals) is by no means resolved. Ongoing attempts at regulating private security and co-opting or crushing vigilante groups continue. The outcome of these processes will be key in determining how transitional societies will be policed in future. [End Page 20]
These five factors point to the difficulties of managing safety issues in transitional societies. In particular, at least one important danger of high levels of crime in the unstable period that follows the end of authoritarian rule is that military-type law enforcement initiatives are likely to increasingly dominate responses to crime in fledgling democracies with an authoritarian past. However, such interventions may be inevitable given the scale of the problem in some societies and vigorous public demands that a more assertive law enforcement approach be taken. Such operations must, however, be balanced with programmes which seek to undercut many of the causal features of high levels of criminality in the societies under consideration. Nor should a more specialised (and often nationally driven) approach to crime control undercut the critical necessity of improving the service delivery functions of the local police. What is required is a more balanced debate on the correct mix between law enforcement and prevention; and locally and nationally driven interventions.
Thus, it should be noted that what is required is a blend of crime prevention, service orientation, the involvement of local communities, local accountability, professionalisation and specialisation of the police, as well as adequate 'law and order' responses (with due respect for human rights). Determining the proportion of, and the relationship between, each of these factors depends on the domestic context in any transitional society.
Police reform in transitional societies must be informed; feasible and manageable; strategic; structured but flexible to allow for short-term adaptions; and under constant evaluation and scrutiny of democratic control. Moreover, it must be emphasised that police reform (or for that matter crime control) is not the exclusive preserve of the police and must involve external actors such as the private sector as well as community inputs. In this respect non-governmental organisations have played an important role in issues of police transformation in many transitional societies, including South Africa.
It is perhaps of concern to note by way of conclusion that there is no clear example of a post-authoritarian state that has been able to achieve significantly enhanced levels of safety for its citizens. In some states in particular the challenges are enormous. Police reform in Nigeria, given both the scale of the country and the size of its police agency (approximately 120 000 for 140 million people) and the extent of lawlessness characteristic of Nigerian society, remains a daunting prospect (Chukwuma 2001). [End Page 21]
This does not mean however that the battle is lost. Absolutely key to success in fighting crime in transition societies is an understanding that innovative local solutions which will rebuild social and community ties are required to solve problems that were often shaped by powerful national influences. Despite the fact that autonomous paths to development are restricted within the context of an increasingly globalised world order, police reform and crime control policy which build on local potential is often the most viable strategy. This is even more important in view of another recognition, that is, while different transitional societies existed in different contexts, success in one might hold the promise of some success in others. The role of analysts working in the area is not only to scrutinise government pronouncements on crime and policing policy but to communicate both successes and failures to their counterparts in other transitional societies.
Mark Shaw is currently with the Crime Reduction and Analysis Branch of the United Nations Centre for International Crime Prevention. At the time of writing this article he was a Research Fellow at the South African Institute of International Affairs, Johannesburg.
The conclusions of the paper are the result of an ongoing research project on the nature of crime and the challenges of police transformation in societies that have experienced transitions from authoritarian rule to democracy. The work is funded by a grant from the Ford Foundation.