CQ No. 25 SA Crime Quarterly No 46 • December 2013 61 Andrew Faull (AF): Olly, we know performance measurement and performance management are important issues for public services, and as such for the police. During my fieldwork I was often struck by a sense that the SAPS was lying to itself. The data the organisation was generating and communicating to itself was not always entirely honest (as discussed in detail below). How do Nigerian police officers understand performance measurement and how does it play itself out in police practice? Olly Owen (OO): One of the main issues in measuring police performance in Nigeria is that all performance measurement, like policing policy in general, tends to be reactive. Statistical measurement of crime, or cases, is extremely loose and variable in quality and in what it captures, so that it is not a realistic basis for examining performance. This has been shown through recent donor-supported attempts to bring in measurement culture. Currently, cases are often registered (documented) retrospectively, after an appropriate settlement or solution has been found, so as not to create a ‘problematic’ paper trail of unresolved incidents. In cases of petty crime, formally documenting cases can itself be a threat, made to suspects who refuse to reach informal compensation arrangements with complainants. That is, if they refuse the option of resolution, their offence will be made a matter of record, and thus irrevocable, triggering – or at least making more likely – certain state processes. But in cases of more serious crime, there are also some incentives for officers not to create cases that cannot be resolved and would thus reflect badly on them and their superiors. In the places where statistical measurement is used as a performance indicator,1 divisional police officers (DPOs, precinct managers) tend to respond only to the varying rates of headline-grabbing crime such as murder and armed robbery. In such busy precincts, there is a tendency to assume that if the rates of these serious crimes are managed, all is well. These are also in general – along with car- snatching (vehicle theft and/or hijacking) – the only crimes that superior officers will query lower- level managers about. So within policing circles there is talk of incidents where officers have recorded bodies found on roads, or even with parts missing, as ‘sudden and unnatural death’, with only a cursory preliminary investigation into other possible explanations. This is primarily due to police officers being aware of the risk of listing a homicide case with no obvious suspect – doing so will simply create more unsolved cases, which make the officer and his unit look bad. with Olly Owen and Andrew Faull ON THE RECORD ... Olly Owen is a junior research fellow in the Department of International Development at the University of Oxford. In 2012 he completed his doctorate in Social and Cultural Anthropology at Oxford. This involved an ethnographic study of, and fieldwork with, the Nigerian Police Force. Andrew Faull is a doctoral research student at the Centre for Criminology, University of Oxford. His research also involves an ethnographic study of police, the South African Police Service. He completed nine months of fieldwork with the SAPS in April 2013. In this frank exchange Olly and Andrew discuss their observations relating to performance management in the respective agencies. 62 Institute for Security Studies Also, there are no baseline data to measure police performance against, except a general public sentiment or institutional memory of how conditions and performance have been historically. A non-governmental organisation (NGO), the CLEEN Foundation, has for several years collected criminal victimisation data through a national victims of crime survey, but the results are not well-known or well-used at the level of localised policing. Besides, their funder has recently redirected this funding to other areas. Beyond that, since policing policy at both local and national level is reactive – there is no strategic policing plan, either at national level or in any of the 36 states – there are no targets against which to measure performance. And since the NPF is centralised, and upwardly accountable, there is no easy and direct way for localised publics to voice opinions over police performance. Some more progressive officers take care to communicate well with local citizens and create forums for their input, but this is not a required or regularised system. Police Community Relations committees do exist, but function very unevenly, well in some places, badly or not at all in others. And the system for complaints is likewise not uniformly functional nor is it valued within the force as a performance indicator. Often it is viewed as a threat. This means that often the most powerful forms of communication and accountability between police and public are informal – expressed via social and political networks, or in public protest action – and thus unpredictable, creating a high sense of risk among officers wary of public moods, in circumstances where officers’ actions may suddenly spark public protest, complaints, or worse. How powerful would you say is informal accountability and feedback of sentiment about police performance in shaping officers’ behaviour in South Africa? AF: I see a number of similarities but also clear differences between what you’re saying about the NPF and what I’ve seen in the SAPS. Firstly, the SAPS has quite a well-established formal performance measurement system. Each year a strategic plan lists baselines, and targets are set across a range of performance areas. These are made available to the public through hard and soft copy publications so that, in theory, the SAPS can be held to account. The SAPS also has an always-on and nationally networked Crime Administration System (CAS), in effect an online crime database. This allows managers to draw data from the system and respond to reported crime with every 12-hour shift change. At the two city stations where I worked this seemed to be the norm. In morning meetings detectives would discuss all reported crime from the previous 24 hours, and patrol commanders could post patrols based on the geographic spread of recent activity (though in practice patrollers often ignored or failed to listen to this information during parades, I believe). At the rural station I worked at crime was so rare that it wasn’t necessarily to pull stats, everyone knew if something had been reported. It is also mandatory that all SAPS stations establish community policing forums (CPFs) comprising local residents and business people, and that these meet regularly. However, like the civilian oversight you describe in Nigeria, these vary in size, strength and impact across the country. At the rural station I was impressed that, despite having to provide transport to members from across the expansive precinct, police managed to maintain relatively good relationships with their CPF as well as other community groups. In recent years there have been concerns, both from within the SAPS and civil society, that the SAPS places too much importance on reported crime as a measure of police success. This is exemplified in the ritual of releasing the national crime statistics from the previous financial year each September, with great ceremony and fanfare, followed by public debate about what the data mean for the country and the SAPS. I forget who said it but there’s a great quote that goes something like, ‘The problem with taking credit when crime goes down, is that you’re also responsible when it goes back up.’ If there’s one SA Crime Quarterly No 46 • December 2013 63 thing we know it’s that police have a limited impact on most types of crime. You mentioned that police officers in Nigeria try to not record certain types of crime. Because of the importance placed on recorded crime in South Africa we’ve seen instances in which the media have reported that police have neglected to record crimes on the CAS to ‘stay in the green’; in other words to keep the number of crime incidents within the designated target. The targets tend to be set lower than the number of crimes recorded in the same month of the previous year. During meetings and briefings there’s a lot of talk of being ‘in the green’ or ‘in the red’, depending on how many crimes have been recorded for the month thus far. A notable change in recent years has been an emphasis on ‘crimes reliant on police action’, such as arrests for drunk driving. In my experience police are generally able to ‘stay in the green’ when it comes to these categories, and so play them up in the media. During my recent fieldwork I became aware, for the first time, of the SAPS recording data I had not previously seen recorded. For instance, the commander of a ‘crime prevention’ unit would open a spreadsheet on a station computer and record the number of people stopped for the evening, the number of cars stopped and searched, the number of taverns visited, the number drugs confiscated, and so forth. What is important about this, however, is that the data would always be captured mid-shift. In other words they were estimates based on the number of actions and seizures the commander hoped that his members would carry out. In a sense he was helping to build an imaginary narrative, part fiction, part fact, which would travel up the national hierarchy, not only guiding future management decisions, but eventually, making its way into the annual reports. I noted something similar during VCPs (Vehicle Check Points) at one of the stations. Shifts would regularly be charged with conducting up to five or six different VCPs across their precinct during a single shift. This involved setting up a roadblock, stopping passing cars, checking licence details and searching for illegal substances, weapons and so forth. These have become a regular feature of South African policing and are something the SAPS proudly reports on each year. However, while at times we established formal VCPs and stopped traffic, most of the time VCP reports were forged. The officials I was with would record the physical description and licence plate information of passing cars, and enter a fictitious name and driver’s licence number to accompany it. This would serve as evidence of a car having been stopped when nothing of the sort had happened. We would always be sure to gather this information in the area in which the VCP was meant to have taken place, so that the car’s AVL (Automatic Vehicle Location system) would show that we had been in the correct area at the right time. We would usually complete one pro-forma sheet over a 20-minute period before heading on. However, as one official said to me, if police officials ran their VCPs for an hour, as they are meant to, they would fill many more pages. Because this never happens, and police tend to only run a VCP (or fake VCP) until one page’s worth of details have been captured, management must know they are being deceived. And yet it didn’t seem to matter. As long as the next person up the hierarchy was receiving data that he or she could report to his or her superior, everyone seemed happy. This is what I meant when I said I’ve decided that in many respects, the SAPS is an organisation that is lying to itself. Of course it is important to have data and the SAPS has good recording systems. However, it is problematic that an agency which, in popular culture at least, is meant to be truth centred and justice focused, appears to be feeding itself fraudulent performance data. Am I to assume, from your description of Nigeria, that systems there have yet to become widely computerised or centralised? As Africa’s second wealthiest country, this surprises me. Perhaps this explains your question regarding informal accountability. In South Africa, following apartheid, there has been a great emphasis on building formal accountability structures. I think these have played an important role in shaping 64 Institute for Security Studies police performance, though they have not always functioned fully or optimally, and are always evolving. At a local level I think some CPFs are quite active and capable of encouraging certain types of police performance, though these are not always technically legal (like offering free drinks or food to on-duty cops in order to ensure their presence in an area). South Africa is fortunate to have a range of state and non-state generators of sentiment data about police, including regular surveys. However, I don’t think many police are aware of these surveys and anyway, your question was about informal sentiment, not formal. Here I would remind you of my article ‘Fighting for Respect’ (SACQ 44) in which I suggested that, in certain areas at least, I think many police feel underappreciated and disrespected by the communities in which they work. I think that citizens who complain informally about police performance risk irking police and encouraging police backlash. On the other hand, police are very receptive to the free food and drinks many residents and business people offer them, seeing it as recognition of ‘good relations’. So perhaps unsurprisingly, informal praise of police performance is welcomed while complaints are not. OO: As you rightly assumed, there is currently no effective centralised or computerised system to record crime in Nigeria. Where South Africa has post-apartheid, Nigeria has post-military government, but the effects have been quite different. The civilianisation of politics has put the police centre stage in dealing with national legal and security crises. As a consequence the politics of policing have been more often concerned with the ‘high policing’2 political functions than their ‘low policing’ everyday crime control functions. Where the two interests have not coincided, political interest has usually taken precedence. This has included watering down policing on occasions when it might have restrained militancy or other lawbreaking that served certain political interests, further inhibiting ordinary officers’ confidence in various potentially sensitive situations. Data-led policy making often takes a back seat when institutions feel the need to react quickly, especially given the hierarchical and paramilitary police management culture and structures, which are primarily concerned with issuing and executing orders, rather than holistic planning. What you say about the strategic use of free drinks supplied by the public to the police also puts a different gloss on what in Nigeria is usually portrayed as a more extractive practice. In my question to you about informal accountability I was wondering where the police fear to tread, and the possibilities of political backlash for incidents born of everyday policing. Is that a part of performance measurement and accountability in the crude sense? AF: I think a common and accurate perception in South Africa is that the poor bear the brunt of the police gaze. So I was surprised, at one of the stations at which I was based, to find that middle- class people, even tourists from the developed North, were relatively regularly arrested. I suspect this might surprise many South Africans. However, for every such arrest there were at least 20 arrests of poorer South Africans and Africans from elsewhere on the continent. Regarding ‘high policing’ and the political arena, I think this is an area that is rapidly unravelling to reveal all sorts of complex ties and alliances between senior police and politicians in the country. You might have heard of a former investigative agency in the country known as the ‘Scorpions’, located outside the SAPS. It was largely seen as fearless (or biased) in its investigations but was closed down through political manoeuvring. The ‘Hawks’, its SAPS- based replacement, has taken on some big cases but has been far less courageous than its predecessor in tackling cases involving people holding power in the state. But in terms of everyday policing, I think that many cops are just out to cover their backs and ensure they hold onto their jobs. As such, there are many who would not seek to ruffle political feathers if they know their actions might result in this. Can we measure such inaction at a performance level? I’m not sure. SA Crime Quarterly No 46 • December 2013 65 OO: In Nigeria it is also noticeable that one of the ways in which the public measure police performance, is by comparison. The comparisons invoked are usually between the performance of the NPF today, and the same or other forces in imagining the past, future, or elsewhere. Thus today’s police are unfavourably compared with the folk memory of the colonial police, or the localised Native Authority Police Forces, which existed prior to the 1966-70 civil war. Or the NPF will be compared negatively with the standard of policing elsewhere, with favoured comparisons being the US, UK and Ghana. These comparisons are not always based on actual experience, though given the commonplace nature of trans-nationality in today’s Africa they may be. More often they seem to be informed by media representations, such that the powers and performance attributed to police forces in the developed world may seem to render them omniscient and omnipotent. When Lagos gubernatorial aspirant Funsho Williams was assassinated some years ago, a major national paper ran a front-page story reporting that experts from London’s Metropolitan Police would examine his retinas on which they would find recorded the last thing he saw. Interestingly, this powerful transnational myth has been enhanced by the deliberate promotion of British police involvement in Nigeria over recent decades – using innovative forensic techniques to identify a boy victim of ritual killing in London, or dogged pursuit of high-profile corrupt politicians such as Delta State’s ex-Governor James Ibori, for example. So this reinforces the myth of omniscience, against which an NPF that lacks even the ability to conduct fingerprint analysis, can only compare extremely badly. Likewise, the Ghana comparison is not based on analysis of Ghanaian police performance, but is instead part of a kind of masochistic public rhetoric, in which Nigeria is compared negatively with its smaller West African neighbour. I also mentioned the future. Advocates of autonomous state police forces also point to the accountability and performance benefits they expect from an alternative system as a critique of the status quo. And in addition to these, there are some real institutions to compare with the NPF, in the form both of alternative policing providers such as vigilante groups and youth vanguard-turned security outfits, and of the specialised policing agencies that were split from the NPF under military rule, such as the SSS (State Security Service – a mix of surveillance, intelligence and anti-crime agency), the EFCC (Economic and Financial Crimes Commission, the elite anti-fraud unit) and the NSCDC (National Security and Civil Defence Corps – a state guarding agency that likes to contest police competences). AF: In South Africa the SAPS is often compared to its predecessor, apartheid’s South African Police. Comparisons involve perceptions of the former agency being more efficient but also violent, abusive and lacking accountability. I don’t think there is an established discourse, either in the SAPS or the population more broadly, of comparing the SAPS to our African neighbours. This probably has something to do with beliefs regarding South African exceptionalism, that our experience is incomparable to the rest of the continent. However, the cop of American or British television dramas, and the fantastical world in which they operate, is far more likely to be held up as the everyday benchmark against which to measure police in South Africa. This is of course dangerous, as it sets police impossible standards, and irrevocably links policing and crime in ways that become hard to challenge in popular and governmental consciousness. What is interesting, I think, is that in recent victim surveys many South Africans report being satisfied with the police in their area. This is in stark contrast to regular, loud complaints about poor police service. I suspect this is in part because, despite lots of complaining, South Africans do empathise with the idea that police are central to tackling the violent crime problem. 66 Institute for Security Studies OO: The Nigerian public have traditionally had little sympathy for the police, whom they portray as both underperforming and predatorily corrupt. That is only just perhaps beginning to change, as the public becomes conscious of the risks and losses the police have endured in the Boko Haram insurgency – both in bombings and incidents where this and other militant groups have taken the lives of policemen as a kind of currency with which to send a political message to the Abuja government. I think what ties both situations together, however, is the expectations of police performance and how it is achieved. Perhaps this is inevitable, as the ‘thin blue line’ maintains its bluff of superior strength as part of its job – yet policing, along with perhaps medicine, is one of the areas where a profession’s global media myth can be a stronger shaper of public expectation than its actual lived reality. So I find the Comaroffs’3 argument about the way that ‘detective fictions’ are staged to demonstrate potential police power in South Africa an interesting response to those expectations; and in Nigeria, going forward, I will be as deeply interested in how the police shape their own media image through their working practices. AF: Indeed, the media and mythology angle is a very interesting one. I don’t think there is enough acknowledgement of its power in South Africa, yet the SAPS and government more broadly clearly know how to play the ‘performance management’ game when they need to. NOTES 1. There is no regularised national system of performance measurement based on the number of recorded/reported cases of crime. Such statistics may be used or not used depending on the importance particular managerial regimes place on it. Using performance management in a more regularised way at station/division level is a part of some donor-supported professional tutelage programmes. 2. The term ‘high policing’ was coined by Jean-Paul Brodeur. Brodeur introduced the term in 1983 to refer to the types of policing carried out by intelligence and other state agents not involved in day-to-day uniformed police work. Rather, ‘high policing’ involves actions intended to protect the state. The concept has gained currency over the past decade, particularly in the wake of the ‘war on terror’. See for example Jean-Paul Brodeur, High and low policing in post-9/11 times, Policing, 1(1) (2007), 25-37. 3. J Comaroff and J Comaroff, Criminal obsessions, after Foucault in J Comaroff and J Comaroff (eds), Law and disorder in the postcolony, Chicago: University of Chicago Press, 2006.