Navigation – Plan du site
English Version

“What Works in Reducing Crime” (November 2004)

What Works? A very unFrench question?
Anne Wyvekens

Entrées d’index

Haut de page

Texte intégral

1During a meeting of experts in crime prevention, an Italian speaker describes a programme to fight truancy that was cited in his country as a "good practice". The British representative enthusiastically exclaims, "Oh yes indeed, it will surely work!". The French expert then speaks up "Practically speaking, without a doubt, but will it work in theory?"

2This little story delights English-speaking audiences, while the French manage only a forced smile. What works? Is this question definitely unFrench? The attention paid to evaluation/assessment, the notion of good practices and usages, are these all just Anglo-American curiosities, unfathomable to the Gallic mentality? This was the first idea that came to mind when I heard this joke and the reactions it provoked. A chasm between the French and those they call "Anglo-Saxons" whether they be British, American or even Scandinavian. How can we explain this? Because the pragmatic note that frames this question doesn't really find favour in the French mind, which prefers abstract and general reasoning. Because the attention to evaluating public policies is not the main concern of French institutions. Because, more generally, the French are almost viscerally sceptical of what arrives from the other side of the Atlantic, and are quick to criticise whenever they can.

3But maybe we should delve into the question a bit further. I'm not an expert in American civilisation, and I am not going to speak here about what America is or isn't. Rather, based on a comparison we made a few years ago on local public security policies in France and America (in particular community policing as implemented in Chicago)1, I would like to take the French expert's remark - at first glance paradoxical - seriously: "In theory, will it work?"

4The paradox is only superficial. Why? Because, practices always consist in theories that are tested, in explanations that will prove to be relevant or not, and on this basis, operational. What works to reduce crime only works when and because we have managed to give an at least partially relevant explanation for the origin of crime. More broadly, the paradox is only superficial because we cannot blithely dissociate theory from practice or vice versa.

5Since we were dealing with local security practices our comparative work brought out two models:

6 Chicago's community policy model involves pooling the efforts of the police, municipal services, and the residents of each neighbourhood. At monthly beat meetings the police listen to the residents' concerns about safety, regardless of whether they are directly related to crime - serious or not - or to the quality of the local environment. The local nature of these meetings allows the specificity of each neighbourhood to be taken into account, because safety for one group of people doesn't always rhyme the same way for another. These beat meetings are also the occasion for the police to report on their actions to the local residents, proof that their requests have been taken seriously. Actions are deployed in line with the philosophy of problem solving. Criminal acts are not considered a posteriori, case by case, but in their repetitiveness that shows a "problem" exists. Each problem is seen as a triangle where each angle (author, victim, place) can be acted on, working more proactively than in reaction. In other words, it's a form of prevention that relies both on law enforcement authorities and on the involvement of the residents.

7The French Local Security Contract (CLS) model springs from another type of partnership: not between the police and the residents, but among various institutions (schools, the police, justice, public housing, public transport, social workers) who pool their efforts to conduct a vertical form of action (instead of Chicago's horizontal thrust) towards the residents. The action primarily focuses on social prevention, revolving around solicitude (community initiatives, schooling assistance, legal consultancy...) or teaching lessons (legal reparations, the rappel à la loi procedure...). They aim to both renew the people's trust in the institutions (by facilitating access to their rights) and provide a minimum of references (by reminding people of their duties) in order to keep them from falling into crime. In other words, it consists in prevention, a renewed form certainly, but nevertheless always preventive and always competing with law enforcement measures that are established in parallel with firm or weak backing depending on the government.

8And what about the corresponding theories? On the American side, Chicago-style community policing was inspired by the "broken windows" theory. While some, especially in France, only see an updated version of the old adage "qui vole un œuf, vole un bœuf " (who steals an egg will steal an ox), the thoughts of James Q. Wilson and George L. Kelling draw from a true theoretical source: the intention to repair broken windows arises from a reversal of the causal link between crime and fear of crime. Although traditionally it was held that it was crime that frightened people, these authors showed that the fear of crime prevalent in a run-down neighbourhood (because it is run-down: broken windows or disturbing behaviour) was itself the cause of crime: For a deteriorated neighbourhood is one that has been abandoned by its residents, leaving it open to criminal activities that flourish in the absence of informal control. To adopt a medical metaphor, the Chicago model could be analysed as an immunological model: the disease (crime) exists, to fight it one must reinforce the defences of the organism (in this case the community) with the help of its institutions.

9 In France, social prevention and education on rights - the primary thrusts of the CLS - are based on socio-economic explanations for crime. Crime is rooted in social misery, family dysfunctions, the lack of authority and the anomy this context leads to. If young people become delinquents it is because they feel abandoned by the institutions, at the same time deprived of the references essential for life in society and/or full of aggressiveness towards this society that has abandoned them. It's "Society's fault", Society that has wronged them and which is responsible for a solution. To continue with the medical metaphor, the French model is prophylactic rather than immunological: one must ward off the disease, forestall it. And this is the State's role.

10So, does it work? In the USA the Sherman report (which classifies crime control strategies in three categories: those that "work", that "don't work" and "promising" programmes) provides us with several indications. No community initiative appears in the list of programmes that work. Moreover, community initiatives in underprivileged areas with a high crime rate are among programmes shown to be not effective. On the other hand, multiplying police patrols in these same areas reduces crime. And lastly, problem solving is on the list of promising programmes. So, exactly what does "work" signify in this meta-evaluation conducted by Sherman and his team? That the programmes that work have achieved a drop in crime in the area in which they were implemented. As for the Chicago programme, it is the subject of a long-term evaluation, under the direction of Wesley Skogan who has followed the programme since its very inception. More than a simple impact evaluation, this is a true follow-up, monitoring practices, seeking to analyse their multiple facets, effects, strengths and weakness, in order to adjust the programme's evolution in the most useful way.

11In France, the absence of an "evaluation culture" (and the lack of evaluations) does not keep people from answering the question "what works"... It is hardly a caricature to say that for people on the left repression doesn't work . Only prevention ­ obviously good because it's generous, understanding and based on solicitude ­ finds favour in their eyes. On the right, fed up with what they call the "culture of excuses", prevention is condemned. On one side and the other, the condemnation is by principle.

12So what can be said about the relation between theory and practice? The keynote of Chicago's community policing programme is it pragmatism. It primarily consists looking at a problem in a way that enables one to solve it. Why? Because they're going to ask you to account for what you've done, they'll evaluate you. "They" can be politicians, Congress, who commissioned the Sherman report to find out how the Federal and State money has been spent. "They" are also the citizens and the local participants in beat meetings where, under the item Old Business, the police have to report on how they solved the issues raised at the last meeting.

13In France, the lack of evaluation goes hand in hand with remaining in ideological realm, and precisely for this reason, maintaining an ­ often sterile - opposition between prevention and repression. Each side recriminates the other, without looking into the concrete situations. One side unqualifiedly praises (social) prevention which is "naturally" good because it's generous. But they fail to see that people are asking for results: victims in particular. The other side, proponents of repression, barely give more thought to the victims. Their overarching concern is rather to keep the criminal or delinquent out of sight ­ no matter what. And neither do they give account for their actions. Evaluation, or at least endeavouring to evaluate, the search for practical effects, provides the possibility to shake this ideological absolutism. One cannot enforce the law with prevention, nor prevent without enforcing. Not to vindicate one approach over another, but because one is nonsense without the other. Prevention is meaningless unless it is coupled with demonstrations of reproach and endorsement of the law. And likewise, punishment is meaningless unless the means to respect the law are accessible.

14Conversely, limiting oneself to "what works" (measuring the "effects" of programmes) entails the risk of measuring results divorced from their context, results in the strictest sense of the term: lower crime figures. We thus lose sight of why and how something works or not. We deprive ourselves of the possibility to factor in a certain degree of complexity. One example: One of the shortcomings of community policing is the fact that when crime drops in one neighbourhood following the programme's successful implementation, this is often at the price of the criminal activities shifting to another neighbourhood. And it takes a theory, especially the notion of immunology, to understand why. Once one community has been strengthened, the "disease" moves to a nearby organism that is weaker. Another example: Community policing has not managed to reduce crime in some neighbourhoods, precisely the ones most seriously affected. It's a fact. Nevertheless, even in these areas people feel better and have the impression that crime has dropped. Once again, it's theory that helps us understand. Call it immunology or empowerment, it is just as important for people to feel they have control over their environment or institutions, as to see a statistical drop.

15In conclusion, we've seen that it's not a question of preferring theory over practice or vice versa, but rather to combine both. What underlies the Frenchman's apparently paradoxical question is that behind a practice that "works" there's a theory that doesn't always fit or the explanation of why a practice doesn't always work. Questioning the theoretical suppositions of a practice enables us to show that there is no univocal explanation. On the other hand, the empirical evaluation of practices teaches us, through requirements for accountability, that strong feelings, even when they're "good", are not enough to give meaning to a policy. A policy is both results and meaning.

16And I am sure that all, French and "Anglo-Saxons" alike, have found this exercise de style to be.... very French.

Haut de page


1 J. Donzelot, avec C. Mével et A. Wyvekens, Faire société. La politique de la ville aux Etats-Unis et en France, Paris, Seuil, 2003.

J. Donzelot et A. Wyvekens, La magistrature sociale. Enquêtes sur les politiques locales de sécurité, Paris, La Documentation française, coll. La sécurité aujourd'hui, 2004.

Haut de page

Pour citer cet article

Référence électronique

Anne Wyvekens, « “What Works in Reducing Crime” (November 2004) », Champ pénal/Penal field [En ligne], Archives, mis en ligne le 25 novembre 2005, consulté le 24 février 2018. URL :

Haut de page


Anne Wyvekens

Juriste et criminologue. Chargée de recherches au CNRS, responsable du département de la recherche et de la valorisation à l'INHES, ex-IHESI. Rédactrice en chef des Cahiers de la sécurité intérieure.

Articles du même auteur

Haut de page

Droits d’auteur

© Champ pénal

Haut de page
  • cnrs
  • Logo DOAJ - Directory of Open Access Journals
  • OpenEdition Journals