Abstract
This article deals with the complex aspect of French public safety policies and offers a big-picture view of the factors that impact both their design and their implementation processes. Indeed, many players are involved in their development: city mayors and local police forces when they exist, but also local agents of the national police forces who have the main responsibilities in this field, and the Public Prosecutor. Whereas cooperation between national police forces and local authorities is statutory and needed in order to address safety issues, conflicts and tensions are not rare. Each stakeholder pursues their own objectives: mayors are seeking public support and try to meet local expectations, while national police chiefs favour priorities set by the Ministry of Home Affairs. Far from following only a rational choice process, the implementation of local safety policy results from power relationships and the position of each player in the game. To gain the upper hand over a partner, a stakeholder can rely on material resources but also on the control of information about security issues, knowledge and expertise in this field, and its legitimacy among the public. In the French case, two forms of legitimacy are in conflict as far as security policy is concerned. If mayors base their arguments on their status conferred by a democratic election, the police chiefs use the fact that they belong to a national service based on a principle of equality – even if it is not applied: every citizen has the same rights and duties, and is subjected to national policies. Because the French Republic does not recognize communities and local peculiarities, institutions tend to favour their own positions rather than dealing concretely with the safety issues.
Even in such a highly centralized country as France, where policing responsibilities are essentially split between two national forces and countrywide priorities are defined through national public safety policies, policing activities cannot merely consist in blindly following the latter. France being such a diverse country (and the largest state in the European union), local arrangements are a necessity. Field actors such as the mayors, Public Prosecutors and Prefects (the local representatives of the central state) all have a say in the design of local policies, to the point that observers have been hinting at a ‘de-monopolization’ process that has long since led the central government to lose its monopoly over legitimate violence (Roché, 2004). However, not all security issues have been devolved to local government entities, and the central state has certainly not given up on staying in the driver’s seat, including in major cities (Maillard and Mouhanna, 2017). As a result, public safety governance is characterized by a state of structural tension between the central echelon’s wish for consistent, nation-wide approaches, and attempts by local authorities at tailoring policies to address the needs of their own constituencies. Beyond the centralized nature of policing, an analysis of the inner workings of public safety policies shows huge heterogeneity. Diversity of local structures, resources, politics and problems – coupled with central and local officials actively seeking to have their political interests reflected in police policies – has thus far not been reconciled through formal central coordination processes.
The present article deals with this complex aspect of French public safety policies and offers a big-picture view of the factors that impact design and implementation processes. By thinking outside of the box of how these policies are formally handled, I show how they tend to emerge not only from cooperation but also from conflicts and tensions among the players involved. Both the activities and the modes of cooperation of police forces and other stakeholders are highly regulated. Still, the intended harmony often leaves much to be desired, not least because national guidelines may fail to account for the diversity of local contexts, resulting in a feeling of disarray, or at the very least in a heterogeneous set of situations that the centralized apparatus is hard-pressed to handle. This has resulted in a system that lacks either the coherence or the rationality associated with centralized models of governance or the responsiveness (to local issues) associated with local models of governance.
Following a necessarily simplified overview of the local public safety system and its configuration, with a particular focus on the actors involved and how they position themselves in the ‘game’ that is building up among them around security issues, 1 I examine the question of who might be the true master of these ongoing tensions in policy matters. We identify which specific resources allow each of them to individually weigh in on the game and pursue their own objectives, or at least influence the decision-making process (Allison, 1971). More precisely, I seek to understand how everyone’s effort to exercise some influence affects these public safety policy design and implementation processes, ultimately bringing about actions that may not be entirely rational or efficient (March, 1988). I use a wide range of research materials, from the many interviews I have conducted with both police practitioners and local elected officials over the last 15 years (Maillard and Mouhanna, 2017; Mouhanna, 2011) to press articles and legal texts.
Starting from there, I show in particular how complex and difficult it is to get state and municipal security services to cooperate, however indispensable such cooperation may appear to be. Many attempts simply end up creating tensions that prove detrimental to the intelligibility of security policies.
Local public safety policies in France: An overview
Formally speaking, until the late 1990s, local public safety policies allowed for little more than slight local adjustments to the priorities defined in the agenda of the central state, that is, by the Minister of Interior exclusively. Although actual facts reflected this formal position fairly well, the vision was nonetheless subject to some checks and balances, as I will show below. From 1997 onwards, it started being called into question by a political willingness to tailor security policies to suit local needs (Ministry of Interior, 1997). However, the system ultimately brought about by the resulting cooperative process turned out to be rather self-contradictory – in more than one respect.
Heterogeneous local government entities
France’s administrative government structure is a hugely complex affair involving no fewer than five layers of management. Here I do not give a detailed exploration of an organizational system whose various entities routinely spend significant time arguing over who is responsible and who is financially accountable for what. Rather, from a public safety perspective, the most important level is the one closest from the field, that is, mayors, who ‘must use their policing competencies to maintain public order, safety, security, and public health in their municipality’. 2 From a legal point of view, things sound fairly simple here: mayors are in charge of safety and security issues, in the widest sense of the terms, for the communities who elected them. They handle security, manage crime prevention and have the power to ban a number of activities. This generic definition, however, often collides with practical reality, owing to the many different situations denoted by the word ‘municipality’.
For historical reasons, France boasts more than 35,000 municipalities, 3 each of which has a mayor who enjoys the same legal powers as every one of his or her colleagues. However, the vast majority of these municipalities are not large enough to have expert security staff of their own, and not rich enough to pay for a municipal police force; 54 percent of these municipalities have fewer than 500 residents, and 86 percent fewer than 2000. Although located in rural areas and not densely populated, these small municipalities are not immune to security issues (Mouhanna, 2016). About half the French population lives in these areas. Mayors are thoroughly engaged in their community, acting both as a social worker and as the community’s spokesperson when dealing with state agencies. In urban areas, conurbations are split into municipalities, which are just as heterogeneous. Affluent cities with their own municipal police force and staff competent enough to diagnose situations can be found next to vast, poverty-stricken suburban areas, often called banlieues, which are prone to major public safety issues but lack the financial resources needed to tackle them. In such vast conurbations, mayors are further removed from the community, and may be tempted to focus more on the needs of ‘their own’ area and voter base.
Financial resources are not the only explanation for observed disparities in the degree of engagement of local elected officials in matters of public safety. Ideological considerations are at work and explain why different strategies can be seen, some favouring a highly securitarian approach combining a municipal police force with CCTV and even juvenile curfew laws, others a more social, preventative approach of safety and security, implemented through educators for instance. Even the forces vary, depending on the choices made by elected officials. Some carry weapons and are better equipped than the national forces, whereas others merely issue parking tickets. A number elected officials, on the other hand, consider the very concept of a municipal police force as going against the founding principles of the French Republic. Since not all municipalities can afford to invest equally in a force, they believe it is the central state’s responsibility to guarantee that all have equal access to policing. This would be consistent with how such a force should ideally be employed in a democracy (Jones et al., 1996). Their ‘securitarian’ counterparts believe that this is best dealt with at the local level, given the national forces’ shortcomings and general lack of concern for local demands.
Heterogeneity at the local level, whether ideological or financial, is partly balanced, officially at least, by the fact that most public safety responsibilities are in the hands of the still-active national forces. The guiding principle of public action in such matters is that the national welfare state is in charge here, as is the case for public health and social security (Rosanvallon, 2015). Policing affairs are dominated by a device that is deeply rooted in the post-1945 genesis of the French state: each citizen has rights that may be exercised and for which the government can be held accountable, in particular the right to be protected. From this theoretical point of view, everyone is equal but, in return for this, the French state recognizes only individual citizens, as opposed to communities of any kind (Zauberman and Lévy, 2003). For instance, the term ‘community policing’ never sat well with politicians and high government officials, for whom the nation ought to be ‘one and indivisible’, and as such should not acknowledge any groups, whether based on race, religion or common interests. 4 According to this principle, there is no recognized intermediary between each individual French citizen and the government, at least as far as the safety of people and property is concerned. This trend was reinforced in 1941 when municipal police forces were nationalized under Pétain’s dictatorial regime. The legitimacy of mayors to intervene in this domain has therefore long been considered with suspicion in France, in contrast to Belgium, for instance, whose initially local policing model tends to linger on even though the forces have been nationalized.
Since the 1980s these ambitious protective objectives have been lowered: having become ‘bloated’ from accumulating competencies (Donzelot, 1984) in spite of ever-growing budgetary pressure, the government has tried to remain influential while cutting down on costs. As a result, some competencies are now shared with local government entities, which have increasingly been involved in policy design. The central state has become one of several partners in theoretically co-produced policies, and so must negotiate. What I am interested in here is precisely the terms of this negotiation process.
Public safety policies are also characterized by the French institutional system’s extreme personalization of power. The President has been granted enough prerogatives – from pardoning to dissolving the parliament, through to being head of the Army – by the Constitution of the Fifth Republic (Conseil Constitutionnel, 1958) to be nicknamed a ‘constitutional monarch’ (Duverger, 1974). Moreover, as their influence over the economy gradually waned, successive Presidents have been keen to preserve their image as a protective sovereign by reasserting their power in matters of security, and thus their authority over a policing apparatus whose control and loyalty is guaranteed by appointing members of their immediate circle. Thus it is that insecurity has become a major topic in French politics, both locally and nationally. This culminated during the 2001 municipal elections and the 2002 presidential elections. Citizens are reluctant to be directly involved in their own protection, relying instead on the authorities and expecting their elected representatives to guarantee it. This autocratic conception of democracy does not necessarily grant citizens a say in policy design. The understanding of the founders of the Fifth Republic was that citizens expressed their views by and when casting a ballot only – after which the elected officials should be fully entrusted with government. This model prevailed for a long time in all areas, including security (Donzelot, 2003). Still, although citizens have been able to reclaim a voice in other domains – healthcare, the environment, land use planning – the dominant central–local tensions between elected officials and their functionaries gives citizens little means of influence other than periodic ballots.
For the personalization process at work at the scale of the Republic is indeed echoed at the municipal level. Far from the collective principles at work in other countries, the French mayor is the central figure of local politics, imbued with a degree of majesty that varies with the incumbent’s means and resources, and a degree of proximity with the residents commensurate with the municipality’s population. The tensions between the local and central level I am discussing here are a permanent feature of French politics, where power has long been split between a centralizing monarch and a number of ‘baronies’, an Ancien Régime terminology still derogatorily used as a reference to local strongholds of centrifugal power. All of this, of course, is compounded by the bureaucratic logics of security organizations and agencies.
The prevalence of top-down management in state policing institutions
The highly centralized nature of both police forces and public safety policies in France is one of the most well-known features of this Jacobin country. Any conurbation exceeding 20,000 residents is handled by the Police Nationale, while the countryside and small towns are under the jurisdiction of the Gendarmerie. Both these highly hierarchized institutions have a General Directorate that reports to the Ministry of Interior. Together, they cover the whole French territory, and their internal organization reflects the traditional administrative subdivisions of the country – from the région (of which there are currently 18), to the département (101), the canton (2054) and the municipalité (35,416). At the local level, a police circonscription (district, 400) covers several neighbouring urbanized municipalities. A brigade of the gendarmerie territoriale (1700) is in charge of several municipalities as well.
Centralization features a highly hierarchical police structure in which obeying orders from above takes precedence over addressing demands from the people, thus neglecting some aspects of democratic policing (Monjardet, 1996). It also relies on a set of management tools and an agenda of priorities established by national policies and assessed with these management tools (Matelly and Mouhanna, 2007). Compliance with top-down orders is a high priority for locally deployed managing officers and is dutifully passed on to the lower echelons, including field practitioners. In spite of the discretionary dimension of their work (Klockars, 1985), the latter are constantly required to bear in mind the Ministry’s agenda – for instance by tackling, first and foremost, crimes that are deemed a national priority. The pre-eminence of national concerns over local matters is reinforced at the organizational level, with brigades specializing in areas considered as national issues. The failure of community policing policies in France must be understood through the prism of this state-centred logic that neither wants nor can make room for citizens in the governance of security.
Another factor is the presence of various authorities – senior civil servants – representing the state and controlling local chiefs at the level of the département – an administrative and geographical subdivision that may encompass hundreds of municipalities. They are the commandant de groupement (group commander) for the Gendarmerie, the Directeur Départemental de la Sécurité Publique (Departmental Director of Public Safety) for the National Police and the Préfet (Prefect), whose competencies cover all aspects of public life, such as spatial planning, economic issues, etc., with a particular focus, however, on public safety issues. Prefects belong to the Ministry of Interior and are expected to abide strictly by state-led policies; as a matter of fact, they may be removed from office at any moment by the government. They are able to weigh in strongly on local policies, because they have a say in how state administrative services work and in funding matters – for CCTV, municipal police equipment – and because the law allows them to substitute for a mayor whose action is considered inadequate.
This is supplemented, at the level of the département, by another authority: the judiciary. It might be considered surprising to classify as agents of the central state the members of an institution that ought, as a matter of principle, to be separate from executive power. French magistrates belong to either one of two distinct bodies: sitting judges (juges du siège), who are independent, and the Prosecutor’s office (parquet), whose hierarchical line extends up to the Minister of Justice and whose members are expected to implement the Ministry’s policies. The parquet is highly and increasingly engaged in local public safety policies, not only because they work with the police on all criminal investigations but also because Prosecutors are more and more often requested, by law, to get involved in local security schemes. They are now expected to deal with prevention, handle public disturbances and help define priorities in the struggle against insecurity. One major difference between Prosecutors and the other local representatives of the state is that the former belong to the Ministry of Justice, as opposed to the Interior, which means that their objectives may differ. In any case, they are as far from the image of the aloof, far-removed decision-maker as can be, being on the contrary heavily involved in local policy-making.
Recognizing that tensions may arise from potential clashes between local logics that vary from place to place and national priorities that, for egalitarian reasons, aim to implement similar policies throughout the country, the authorities have designed schemes to improve cooperation among the various actors.
Highly-organized cooperation
French administrative authorities seem to have a passion for regulation. Since 1997 and the great community policing experiment that ended in 2002, they have developed a number of organizations to arrange cooperation between the players involved in the design of local public safety policies. They can be summed up in a handful of guiding principles. First, the laws of 2003 5 and 2007 6 have confirmed that mayors are at the ‘core’ of security policies, whether preventative or public safety-related, in their municipality. Mayors have ‘the prime responsibility for the safety of citizens’. They must be informed immediately, by local police or gendarmerie representatives, of any breach of public peace in their municipality. They also must be kept informed by the Prosecutor about legal proceedings that may ensue from these or other breaches, possibly notified by the mayors themselves in the first place. Ever since the 19th century, mayors have been endowed with criminal police powers that make them competent to allege infringement of penal laws, to gather evidence and to seek the perpetrators. Although this rule has never really been and still is not implemented, local elected officials nevertheless may use it to support their claim of greater involvement in matters of security.
Next, the prevailing organizational principle in matters of public safety is that of co-production. This novel concept is promoted by government authorities who, although unable to pay for the management of security from state funds alone, would rather not relinquish a police apparatus that is one of the main pillars of the state’s sovereignty. Local heads of national agencies and mayors must cooperate and define, together, what their priorities and modes of action will be for combating crime. In addition to the above-mentioned information duties, this cooperation takes the form of committees pooling the various players, in particular the ‘local committees for safety and crime prevention’ (conseils locaux de sécurité et de prévention de la délinquance or CLSPDs). These are chaired by the mayor and are compulsory in any municipality exceeding 10,000 residents or comprising a ‘sensitive’ security area. Other towns may, but are not required to, create one.
Another avenue for cooperation is the development of municipal police forces, which have been granted increasing powers by successive pieces of legislation over the last couple of decades. This is also ruled by a mandatory consultation process. Agreements get signed between municipal teams and state services to regulate cooperation between the various forces. Patrol areas are allotted and tasks are divided up among them, from ticketing to handling protests, demonstrations and various sports or festive events involving a crowd.
Finally, CCTV was encouraged by national authorities since the early 2000s, particularly through the Prefects. In the vast majority of cases this is a tool owned by the mayors who bear the costs. Obviously, such a powerful tool for monitoring not just crime but also how the various forces operate is a major resource in the hands of municipalities. CCTV may be used to control not only the streets but also the way police forces work in these streets: it drives all forces to share information and to build cooperation.
Other structural arrangements, which may or may not involve the mayor, also aim to strengthen coordination among the various state agencies. Prefects, in their capacity as the département’s heads of security on behalf of central authorities, hold weekly meetings with National Police and Gendarmerie chiefs. Prosecutors do the same, on a monthly basis at least, with ‘judicial’ police officers, that is, those Police and Gendarmerie officers who officially qualify as criminal investigators – something municipal police are not allowed to do. The Prefect and the Prosecutor themselves also meet at regular intervals. Additionally, for municipalities located in ‘security priority areas’ (zone de sécurité prioritaire, ZSP) 7 the various national and municipal players involved sit together on special committees.
Everything thus appears to be done for local public safety policies to be designed collectively, combining the advantages of potential access to the resources of national forces, if needed, with a degree of adaptability to local realities, thanks to the presence of mayors.
One major difference exists with cooperation as understood in other countries: the lack of representatives of the people or community organizations. Neither neighbourhood associations nor religious community representatives – not to mention the representatives of ethnic and economic minorities – are part of these consultation entities, which include only national and local civil servants. Although the demands of the population cannot be said to go unheard, they are mostly channelled through elected representatives who reflect them with varying degrees of faithfulness – translating and sometimes instrumentalizing them to suit their own agenda.
The game around cooperation
Although things may seem clear in theory, implementation is fraught with obstacles on a daily basis. Rather than describe each situation observed in France (Maillard and Mouhanna, 2017; Mucchielli, 2017), it is more productive to look at which factors have the most impact on the genesis of a security policy, and to understand how actors may leverage their mastery of one of these factors to improve their position. In other words, what key elements may explain that either Prefects, mayors or Prosecutors take the upper hand and push their own agenda? I am not suggesting that the whole process must always be contentious. Still, as pointed out above, the players involved necessarily have diverging agendas, since their respective organizations appear to have diverging logics: nation-wide priorities versus local demands; policing logic versus judicial logic; administrative and political logic versus professional logic from the Police or Gendarmerie. Stakeholders will thus tend to consider their own particular objectives as a priority, instead of targeting common goals that are fundamentally difficult to define and share. As shown by Allison (1971), decisions based on such systems – which involve multiple actors yet are not necessarily cooperative – may yield sub-optimal responses. At some point, everyone will start jockeying for position in power plays that are unlikely to produce solutions that, though possibly optimal from the point of view of residents, would in fact seriously challenge one’s own organization’s practices. For instance, a chief of police whose nationally defined priority is road safety improvement will not engage in actions against street harassment, even if there is a clearly identified need at the local level.
Current discussions among security ‘experts’ are informed by the following decisive elements, through which I shall ‘read’ public safety policies: material resources, control over information and expertise – all of which constitute the building blocks of the various actors’ legitimacy. All of these resources yield their possessor a degree of ‘power’ (Crozier and Friedberg, 1977). These qualities are not shared equitably among the various players, or within towns even, which means that the balance may vary considerably from one municipality to another.
The question of material resources has been touched upon above and is dealt with briefly here, simply to highlight a point that shows how limited the potential municipal engagement in policing actually is. Most cannot afford to build an autonomous policy based on their ‘own’ police force. In practice and in spite of the many pieces of legislation that have increased their powers, municipal police forces remain limited in terms of their scope (that is, the sheer total number of municipal police officers) and in terms of the number of municipalities concerned. France officially possessed 20,900 municipal police as of January 2015, 8 compared with 98,000 gendarmes and 145,000 officers in the Police Nationale force. This figure has remained fairly stable over the past decade. These municipal police are deployed in only 3400 municipalities, out of more than 35,000; and 80 percent of forces have fewer than five officers. 9 Most of these local forces can be found in the greater Paris area and on the Mediterranean coast. The largest in France is in Marseilles (402 officers, that is, 0.46 per 1000 residents), followed by Nice (372, 1.09 per 1000 residents). Cannes comes fifth (197 officers, 2.64 per 1000 residents). One of the highest ratios in the country can be found in St Tropez, where a force of 37 (87th) means the community enjoys 8.18 police officers per 1000 residents. Legally speaking, these officers enjoy much less power than their colleagues from the national force, especially in judicial matters. They cannot investigate crimes, search houses or put someone in custody. In addition, they can intervene only strictly within the bounds of their municipality, which might be seamlessly embedded in a much larger urban area. Owing to municipal budget constraints, the force is rarely available 24/7. As a result, municipal police are usually perceived at best as auxiliaries to their national counterparts. Still, a sizeable municipal force may enable mayors to implement their own policies, or at least to contribute to the actions of the National Police. Should a municipal beat patrol arrest intoxicated people who are behaving aggressively on the street and take them to the national police station, for instance, something will have to be done – at the very least, a few hours in the ‘drunk tank’ will be called for.
At the scale of conurbations consisting of several municipalities, resources are not pooled on the basis of security issues, everyone preferring to have ‘their own’ police, presumably to deter people from neighbouring municipalities with a bad reputation. Dispersal into a multiplicity of small-scale municipalities usually unable to pool their security resources means that they almost certainly are at a further disadvantage to the state (Maillard and Mouhanna, 2017). Again, cooperation will usually be deficient here, and neighbouring municipalities typically refuse to work together, a dispersion that enables the unchallenged central government to impose its views.
The key issue of information flow
Financial resources are not the only criteria influencing public safety policies. Access to relevant information is another determining factor for involvement in decision-making processes. Obviously this point is closely related to the former: a municipality whose budget allows for a CCTV system or a well-equipped police force that patrols the streets and is involved in crime scenes will necessarily be better informed than others that cannot afford those – provided that these human and technical systems work properly. Still, other sources do exist: an elected official can have a network of informants, for instance. Considering how social ties work at this scale, mayors in smaller communities may even find it easier to have a comprehensive view of their residents’ problems than their colleagues from larger cities with specific issues in each of their many neighbourhoods. When faced with a difficulty, people in the countryside will turn to their mayor whereas city-dwellers will more likely resort to specialized police forces (Le Bart, 2017).
The spirit of co-production that officially governs local public safety policies is based on the assumption that information flows perfectly among the various players involved. The law provides for instance that mayors must be able to access all crime-related information pertaining to their municipality as quickly as possible. In practice, though, information flows far less freely than required by law. When citizens call for help in cases of emergency, the Police Nationale and the Gendarmerie – two highly centralized institutions – are usually the first-line responders. Municipal police forces have only an incomplete view of crime-related interventions in their own municipality, certainly less complete than that of the national forces. Should a police operation lead to the involvement of judicial authorities, the mayor does not have to be informed. From a strictly legal perspective, disclosing such information may actually constitute a prosecutable case.
The CLSPD is supposed to be the main information clearinghouse available to the various stakeholders. Yet many municipalities do not have one. Elsewhere, CLSPD meetings tend to be little more than formal ritual gatherings where communication remains conventional and limited to vague, generalist views. Several months, sometimes a whole year may elapse between two such meetings – hardly a favourable context for responding quickly and adapting to the rise of new challenges in security (Gautron, 2010). Most importantly, police and justice officers alike remain quite reluctant to communicate transparently with local elected officials, whom they consider partisan, if not rabble-rousing, whereas they view themselves as honest, unbiased experts driven by a professional ideal (Emsley and Weinberger, 1991). Strategically, national agencies are supposed to defend the state’s interests against a potential resurgence of ‘local baronies’, which explains why national security and justice services strive first and foremost to meet the national agenda’s priorities, as defined by their respective ministries, rather than local demands, which in their view can only be biased. Generally speaking, these agencies do not trust the citizens’ representatives any more than they trust the citizens themselves. Allowing elected officials to intervene would simply be disruptive for their investigative work: secrets might be unveiled and pressure would be applied to influence the investigations (Mouhanna, 2011). Personal and confidential information gathered by police officers and magistrates may not be communicated to non-neutral elected officials who have not been granted proper clearance. Far from being a natural process, cooperating and defining shared objectives involves a level of mutual trust that can be achieved only with considerable effort if everyone’s ‘professional reluctance’ is to be overcome.
The Senate – the second chamber of the French parliament, which traditionally defends the interests of local elected officials – has recently questioned ‘the lack of information received by mayors from security forces’. 10 The Ministry of Interior’s reading of the law was consistent with its officers’ expectations: ‘the mayor’s information is assessed by local representatives of the state’s security forces.’ 11 This interpretation is a significant departure from the law itself. The willingness to ‘hide’ part of the information from mayors can be illustrated in a number of ways. Aside from CLSPD meetings, which are often deemed ‘political’ or ‘PR oriented’, other, more ‘exclusive’ meetings are held by state officers. Elected officials are not invited to those. For instance, Prosecutors hold criminal police meetings to talk to investigators; Prefects convene weekly councils to get updates on the crime situation in the département from the Chiefs of Police and the Gendarmerie. The state has even designed a career advancement policy to ensure the loyalty of these officers and prevent them from bonding with elected officials: the Prefect, the Prosecutor, the Chief of Police and the Chief of Gendarmerie are expected to move jobs every two to three years on average, and may be transferred at will by the government.
Mayors, on the other hand, are elected – and often re-elected – for six years, a long mandate. More often than not, they are members of the local community and gather information from their personal network, their voters and municipal staff. They rarely object to informing state departments, whose help is needed anyway to deal with the security problems raised by their residents, because they do not have the skills and means to do it by themselves. The relationship is thus a lopsided one: the police and Prosecutors tend to withhold information whereas municipalities tend to openly share it. Being on the demand side, mayors end up being dependent on National Police services.
Knowledge and expertise
Besides information, other key resources for dealing with security policies are knowledge and expertise, which consist precisely in the ability to analyse the available information. What springs to mind here is the capacity to intervene forcefully against crime. However, I would rather focus on the capacity to produce relevant knowledge that influences public safety policies and even the behaviour of upstream operators (Ericson, 1994), since the endogenous knowledge produced by field police practitioners is not exactly of the same nature as the latter, which necessitates a degree of hindsight as well as more macroscopic data. Attempts at unifying these two levels, through intelligence-led policing (Ratcliffe, 2016) in particular, have consistently shown that practitioners disliked being force-fed their strategies by analysts whom they considered as too far removed from real-world issues, not to mention danger. Nonetheless, such technical and analytical skills are essential supports to strategic policing.
From this perspective, justice is the poor sibling, insofar as Prosecutors have no specialized teams for this kind of analysis. Although they may, on an individual basis, develop specialized knowledge about a given type of crime, or even about certain geographical areas, they have neither the time nor the skills to carry out in-depth analyses. As far as municipalities are concerned, the usual fault lines apply: rich/poor, major city/small town; securing this kind of knowledge in one’s municipal team requires financial resources. For more than three decades, mayors have been trying to cope with a rising feeling of insecurity and fighting the crisis of the inner cities (‘la crise des banlieues’) by creating a range of new positions to tackle urban safety issues. Educators and social workers have been supplemented by people trained in rehabilitation, uncivil behaviour and CCTV management, as well as crime analysts. As a result, major cities are now in a position to offer actual policing strategies. Still, one should bear in mind that the demands of the citizenry are typically taken into account by compiling sources, as opposed to actually involving residents in discussions (Monjardet, 1996; Mouhanna, 2011).
The Police Nationale and the Gendarmerie have also assembled major teams for this kind of analysis. Collating of dozens of reports made by their subordinates as well as numerous complaints filed by individual citizens, and making use of more discreet tools such as intelligence-led policing, these organizations produce lengthy reports that serve as a reference point for subsequent meetings. Again, although state departments’ managing officers do lend an ear to the demands of elected officials, they consider themselves to be the most relevant in terms of data analysis (Lemaire, 2016). Expertise is thus on their side. At the top of this knowledge pyramid sits the Prefect. With the combined benefit of access to both National Police and Gendarmerie analyses, Prefects tend to be able to prevail over elected officials who, although certainly knowledgeable as far as their own constituency is concerned, at least for the better equipped among them, can only have a narrower vision.
That vision, however, is based on the assumption that all police staff work as one. Reintroduce the distinction between field personnel and office staff (Reuss-Ianni, 1983), though, and it clearly appears that knowledge – hence, the ability of police management to analyse situations properly – is liable to be undermined by excessive autonomy from their base and deficient reporting of field information. One could even imagine, as has been the case for a long time with the Gendarmerie (Mouhanna, 2016), a de facto alliance between field police and local elected officials in a bid to shield their territory’s interests against an excessively homogeneous vision from national authorities. This is less and less a possibility, though, since even rank-and-file officers, these days, are subject to frequent job transfers. Experienced officers, in a bid to flee the rural and hard-to-police working-class areas where fresh graduates from the police academy are typically appointed, keep asking to be transferred in the hope of eventually reaching greener pastures. As a result (Roux and Roché, 2016), no bonding can happen with local communities, not to mention elected officials and their ‘excessive’ demands in terms of service to these communities. Whatever information is collected can therefore only be incomplete, since sections of the population, especially in poorer areas with strong ethnic minorities, who are afraid of the police, never complain at the police station or cooperate in any way (Lévy, 2016; Robert et al., 2016).
Another limitation to the analytical capacity of police management teams is the sheer volume of cases they have to cope with. Zero-tolerance policies and an increased demand for prosecution has led both the police and the magistrates to resort to mass processing methods. In order to comply with national authorities’ demands for a reduction in crime figures, they may even be tempted to avoid registering some complaints altogether. This makes it difficult to follow up on offenders and analyse the evolution of the situation. People get arrested and sent to court for serious crimes and/or minor offences, and the police do not have time to check whether or not they have a criminal record or have previously been incarcerated (Matelly and Mouhanna, 2007). This quantitative logic does not suit the needs of mayors, whose vision is more prosaic and stems from very specific, concrete situations for which they expect targeted responses (Goldstein, 1979). Again, the very structure and objectives of these organizations do not match.
Generally speaking, in national agencies, the ‘street level bureaucrat’ policeman – who tends to be wary of institutional logics and keeps some distance from management, preferring to directly address the demands of the citizen he is responsible for – is being upstaged by a more ‘institutionally correct’ officer. In other words, the institution has taken whatever steps were necessary to make its agents increasingly compliant with internal goals. As a result, in the eyes of the citizenry, national police agencies in France have lost their legitimacy as far as neighbourhood service is concerned (Lévy, 2016). In the eyes of national police management, this community-relations task ought to be picked up by municipal forces, which is far from being the case everywhere, especially in cities with large municipal forces where similar logics are at work and officers are expected to distance themselves from the community to focus more on repression and zero-tolerance policies.
Conflicting legitimacies instead of well-oiled cooperation
However messy the governance of local public safety may appear with this description – following the shift from a national police monopoly situation to more network-based management of security (Loader and Walker, 2004) involving negotiations between the police or gendarmes, local elected officials and Prosecutors – our analysis shows that the state keeps the upper hand over the design and implementation of local public safety policies, being the only actor possessing the requisite knowledge, resources and expertise. Still, can this form of state-led policing be considered fair and democratic, as more generally defended by Loader and Walker (2001)? That is the argument put forward by those who serve the state, and the rhetoric is the same for all public services in France, from healthcare to social security through to the railways. However, in the French case, the state happens to be a central one, governing from Paris. Local government, as epitomized by the mayor and other local elected officials, is perceived as an obstacle for civil servants aiming to combat local idiosyncrasies.
In practice, several legitimacies are competing, making cooperation-based governance, or co-production, structurally problematic. In spite of recent legislative developments, state departments, whose funding and career prospects are at stake, continue implementing central government plans, regardless of whether these meet the particular interests of the area they are in charge of. This pre-eminence of the national agenda over local needs can be illustrated by several examples. In particular, the rejection by national police officers of several reforms aiming to (re)introduce some forms of community policing (Mouhanna, 2011) is proof of a refusal to take into account the specific needs of people and areas, especially in terms of security. Police officers do not want to have their conduct dictated by local demands. For them, equality thus means standardized treatment for all, at least theoretically, but certainly not adjusting to local realities.
This top-down legitimacy granted by the central state to its local representatives finds itself pitted against another, that of elected mayors, supported by the people or at least by those who voted for them. National police officers are very critical of what they view as a state of dependency of elected officials toward voters, which may lead the former to favour parts of the population over others – another reference to equality as being defended by national services (Livre blanc, 2016). Yet national police officers can be criticized as well in this regard, and are: many studies (Jobard et al., 2012) have shown that physical appearance and geographical (that is, neighbourhood) origin strongly influenced how people were treated by the police. Clearly, the discourse on equality has become an ill-founded self-justification tool that is mainly used to reject control from local elected officials.
At the end of the day, negotiations between these two groups of partners, antagonized as they are by their fundamental objectives, is always complicated and rarely productive. Far from producing rational policies, it often generates instead inadequate or unimplemented decisions based on compromise and self-interest, generating mutual frustration. Although elected officials are answerable to their resident population on matters of public safety, they can act only through state agencies, whose management does not necessarily share the same agenda. The voter-issued legitimacy of mayors is pitted against that of the police, who physically protect the lives of citizens by drawing on their monopoly of the legitimate use of violence. Both, however, end up being weakened by the lack of cooperation. Mayors will be criticized for talking big but no getting results, when in fact the police do not implement their policies. The voting citizen will feel betrayed and may even give up voting, or choose mayors with an increasingly radical stance on security issues. The police, on the other hand, will be blamed for not being close enough from the community, for lacking understanding and for failing to adjust to specific contexts. The end result is a vicious circle: theoretically speaking, elected officials need the police and the police need the officials, but their diverging objectives mean that they cannot – or can hardly and imperfectly – manage to build true cooperation. This lack of cooperation in turn produces misunderstandings and contentious situations that make future cooperation prospects even less likely.
Conclusion: Where do citizens and democracy belong in this picture?
These ‘non-cooperative’ governance mechanisms have multiple consequences. ‘Non-cooperative’ refers both to difficult communication and coordination between national and municipal forces, and to the lack of integration of the citizenry in the design, implementation and impact assessment processes of public safety policies.
The first consequence is that local public safety policies lack effectiveness in the municipalities that experience the worst issues, that is, those with the poorest and youngest populations. The more serious the problems are, the less consideration is given to the community’s voice, and the worse the blame game becomes among actors who find themselves unable to offer coherent, concerted responses. In the other municipalities, forces seem to be deployed more as a form of muscle-flexing than to meet a real need. The idea is to reassure older citizens in high-class neighbourhoods, who do not really care about actual levels of insecurity but are scared that ‘others’ – from other towns or areas of the city – may enter their territory. This is clearly illustrated by the over-representation of municipalities located on the French Riviera in the sample of towns with large municipal forces. The second consequence is that police forces tend to get stuck in the rut of repression and control, instead of trying to solve problems and cooperate with the community. The end results are a set of behaviours that are perceived as extremely aggressive and off-target actions that only reinforce perceptions of unfair and inefficient policing in sensitive areas. Thirdly, this pattern of ineffective cooperation fuels a general trend that, by discrediting the elites, governments and institutions, is weakening them all. Paradoxically, even though they consider security to be a pillar of their power, and voluntarism to be a method, they find themselves unable to produce responses that satisfy their voters. Fourth, as far as France is concerned, these local dead-ends collectively contribute to pushing security issues into the national limelight, with similar consequences. The central government’s belief that they can solve public safety issues without resorting to local expertise yields inconclusive national plans that foster the rise and electoral successes of right-wing radicalism, which thrives on such topics as insecurity and the inattention of the ruling elites, not to mention local people being let down by national priorities.
As can be seen, accountability is not working in the context of public safety policies whose objective may be more influenced by the political authorities’ need to build a protective father figure image for themselves than by actual problem-solving considerations. The French language does not even have a word to properly translate ‘accountability’. Referring to interpretivism as defined by Mark Bevir (2013) and his approach to the state as a set of tradition-based cultural practices, national political authorities and the upper echelons of police management are caught in a longue durée vision of political action that has failed to fully integrate current paradigm changes, especially the shift from a purely state-centred model to a logic of co-production.
French political life is characterized by two particularly active myths that are further accentuated by the prominence of the presidential election, and are especially well suited to security policies. One is the myth of the saviour (Garrigues, 2012; Girardet, 1986): voters are waiting for a providential politician to come and take them back to the golden age of a problem-free society. This myth is inseparable from the Fifth French Republic founded in 1958 by General de Gaulle, who thus became the republic’s saviour for the second time (after the 1940 period of defeat). Nicolas Sarkozy subsequently built on this myth in the field of public safety to create a crime-buster image that helped him win the presidential election. The second myth – already mentioned above – is that of centralizing technology: proper solutions may be found only by enlightened senior civil servants, and ought to apply to the country as a whole. This Jacobin approach is the exact opposite of what can be seen in England and Wales, for instance, where local security policies are managed by Police and Crime Commissioners (Lister and Rowe, 2015). Whereas, in those countries, responses to growing concerns about crime will always be based on more counter-powers and democracy, imperfect though the latter may be, in France, the same concerns will always bring about more centralization. This is exemplified by the case of the greater Paris area (Easton and Mouhanna, 2014), where crisis situations in destitute suburban neighbourhoods have mainly been met by a reform that had all national forces report to the Paris Prefect of police, which ultimately failed to improve the level of service offered to the residents of these areas.
These two myths might seem contradictory, one pertaining to the charismatic model of the Weberian typology and the other having to do with the rational-legal model. In France though, the two are intricately enmeshed since most potential saviours graduated from the same technocratic higher education system. The stability of this model, despite its limited effectiveness regarding public safety policies, is a function of these people’s ability to play both sides of the field and ultimately ignore the public discussion that increasing numbers of citizens are demanding. In this model, the citizenry’s place, as we have seen above, is extremely limited. The public – and that includes mayors – has to be ‘managed’ rather than treated as a partner.
Still, some mechanisms do exist that may somewhat soften these trends. Although they vary depending on the context and available resources, they usually look quite unorthodox compared with cooperation-based rules. Small towns with scant resources or expertise, feeling dependent on the whims of national policing organizations, strive to build direct relationships with the Police or Gendarmerie officers stationed in their community. Major cities leverage the political influence of their mayor in national spheres, hoping to gain direct access to the Ministry of Interior and contribute to national policies in order to gain at the local level. Finally, the media can be used to raise awareness on local issues and trigger a nation-wide response. For instance, some mayors from suburban areas have recently campaigned in this way in an attempt to make the government pay attention to a crisis hitting their constituencies. Still, such incidental arrangements cannot pass for a judicious, long-term, consistent policy.
One possible – and apparently rational – response to these local dead-ends would be to decentralize policing functions at the municipal level, as is the case in many other countries. This ‘common sense’ notion, however, hits a number of snags. Most municipalities are too small to have their own police force: there are too many of them, and the country’s territorial organization needs to be redesigned first. Another problem is that the central state would have to fund these forces while relinquishing control over their actions, which sounds like wishful thinking, especially since the risk of mayors using the force for their own profit is a non-negligible one. Finally, there are currently enough examples of municipal forces in France to allow us to conclude that their being local is no guarantee that they will take local demands into account any more than the Police Nationale does. Reconstructing a national police force that truly serves the public entails not just a complete overhaul of the professional ethos of today’s police officer – which is another subject – but also a thorough redesign of the local governance of local public safety policies, which should be based not on obsolete municipal boundaries but on a more relevant territorial scale. In rural areas, the governing body should include enough mayors to reach a critical threshold that would enable a truly sectoral logic to emerge. In major conurbations, on the other hand, the municipal scale is too big: the various neighbourhoods are too different to warrant a similar approach, and policies should be tailored at the level of the neighbourhood itself.
As can be seen, the implementation of local public safety governance is excessively dependent on how the institutions that are producing it are organized. The scale of these organizations, in particular, may not be entirely adequate to enable them to take action in the best possible way in matters of public security. This governance needs to be reshaped and rebuilt – in the true spirit of community policing – at the scale of security issues, that is, the neighbourhood, and should more clearly include all types of residents, who are the first in line – something that current players are reluctant to do, even though it is, for all practical purposes, the inescapable prelude to any democratic form of policing.
