Navigation – Plan du site

The Punishment of Mentally Ill Offenders in Belgium: Care as Legitimacy for Control.

Yves Cartuyvels et Gaëtan Cliquennois
Cet article est une traduction de :
La défense sociale pour les aliénés délinquants en Belgique : le soin comme légitimation d’un dispositif de contrôle ?


Since 1930, Belgium has a law for the protection of society aimed at guarding society against the dangerously insane. That law responded to one of the prevailing issues in late 19th century Europe: the question of “abnormal individuals”, a category that included insane offenders perceived as dangerous. The 1930 Social Defense Act prescribed a special regime for insane offenders and immediately demonstrated deep-seated ambivalence. We propose to analyze the origin of this law, its content and its evolutions over time by adopting an approach taking into account long duration. This approach allows us to highlight continuities and discontinuities in the treatment of criminal insanity and dangerousness over the last century. We also rely on several recent empirical inquiries dedicated to the regime of insane offenders oscillating between care and safety, the assessment of dangerousness and the role held by expertise, insane offender’s trajectory and their release increasingly characterized by risk management approach and techniques.

Haut de page

Entrées d’index

Géographique :

Haut de page


Haut de page

Texte intégral


1Since 1930, Belgium has a law for the protection of society (“défense sociale”) aimed at guarding society against the dangerously insane. That law responded to one of the prevailing issues in late 19th century Europe: the question of “abnormal individuals” (Foucault, 1999), a category that included insane offenders whose “condition makes them dangerous” (Thiry, 1910). In the context of a changing political project, new ways of managing such individuals endangering social order appeared necessary (Pratt, 1997; O’Malley, 2000, 24).

2At the time, recidivism was mainly associated with criminal madness and degeneracy (Maus, 1907; Sutherland, 1908), so that mentally disordered offenders, like habitual offenders, vagrants and alcoholics, were feared. Considered partially irresponsible, they often escaped severe punishment and contributed to figures on recidivism (Bodeux, 1910). In response to the need to deal with these “abnormals”, the new positivist criminal justice approach advanced by the Italian school became quite influential. The idea of approaching these irresponsible criminals scientifically and that of a penal strategy favoring the protection of society against such dangerous individuals penetrated the legal systems of many European countries, particularly through the conferences of the International Union of Penal Law (Jesckek, 1979). The influence of the International Union of Penal Law led to a dilution of the positivist approach within European criminal justice systems, and a logic of compromise prevailed. Scientific knowledge and the priority given to combating dangerous individuals were accepted, but as complementary to the prevailing criminal law, with the view to preserving the rule of law, but also to respect the framework of a welfare-oriented State project (Garland, 1996). In most cases, the criterion of “dangerousness” was to complete the notion of responsibility as the basis for repression, and safety measures for dangerous individuals were added to the sentences prescribed for “normal” offenders.

3This eclecticism was particularly present in Belgium. The debate between the neo-classical school of criminal philosophy and the positivist trend produced a compromise between criminal responsibility and dangerousness, sired by Adolphe Prins, one of the founding fathers of the International Union of Penal Law. The 1930 Social Defense Act prescribed a special regime, on the fringe of criminal law, for mentally disordered offenders (as well as for recidivists and habitual offenders), at the edges of the penal system. That law, reformed in 1964, immediately demonstrated deep-seated ambivalence, oscillating between care and safety, a criminalizing version of care and a medicalized version of punishment. The entire scheme was marked by this ambiguity, including the core measure, internment, the role of expertise and its evolution, and the path followed by the inmate until his release. In 2007, a law on internment dealing specifically with the regime for the mentally disordered was passed and replaced the 1930-1964 Act. Whereas it establishes distinct regimes for mentally disordered offenders and for recidivists and habitual offenders, such law tended to reinforce a penal logic oriented towards the protection of society. However, this law never came into force and has been replaced with the new law of 5 May 2014 that will come into force in 2016.

4The outlook adopted in the present article has three facets. First, it opts for a long-term perspective on the subject under study. We find this approach interesting in that it points up the continuities and discontinuities in the penal management of madness and dangerousness over approximately a century. Next, it concentrates on the issue of mentally ill (insane) offenders, to the exclusion of the regime for recidivists and habitual offenders, although the same law dealt with both of them. This is justified both because the 1930 law and discussion thereof were mostly concerned with the insane, and because the implications of chapter VII of the law, pertaining to recidivists and habitual offenders, have been more symbolic than practical throughout the 20th century (Cartuyvels, 2012). Third, the methodology adopted is interdisciplinary. On the one hand, it calls on the jurist’s methodology when analyzing legal texts and discussion thereof, but prefers to adopt a moderate outsider’s viewpoint (Ost, van de Kerchove, 2002, 454-466). What is intended here is the use of those social stakes which legal discourse is in charge of arbitrating to shed light on that discourse itself, in an attempt to achieve a dialectic articulation between the legal and social science viewpoints. But on the other hand, this approach also leans on the rare sociological researches conducted in the recent years on the Social Defense in Belgium. More precisely, we mobilize different empirical researches conducted by the authors since 2010, who have applied and combined in an original way many qualitative methods: semi-structured interviews (with prison governors, prison officers, psychiatrists, psychologists, medical staff and social counselors), “group analysis” method (gathering within 4 days judges, prison governors, barristers and psychiatrists serving in Boards of Social Defense) (Cartuyvels, 2012; Cartuyvels et al., 2010), observations conducted in psychiatric wings of prisons (five wings were observed during one month), in an Institute for the Protection of Society (the Paifve institute having been observed for 3 months) and in a mental health platform responsible for coordinating care and for improving communication between professionals (Cliquennois, 2012). The article is therefore based on the results of various empirical studies conducted in Belgium (including its French and Dutch parts), even though very little empirical research has been done on the system of Social Defense in Belgium (Casselman, 2011, 245).

5This approach combining historical, legal and empirical studies allow us to show the rise of a medical rationality dominated by safety considerations at the core of a penal logic. In this respect, we also underline the increasing colonization of this medical rationality by safety and control logics that tend to dominate the entire Belgian Social Defense and many other European countries, to the point to raise and nurture a wealth of case law of the European Court of Human Rights (Tulkens et al., 2010).

I - At the roots of the Social Defense Act from 1930: from fear to danger, guilt and punishment to dangerousness and preventive detention

1) The neo-classical school or individualization of “just punishment”

6In the late 19th and early 20th centuries Europe experienced conflicting schools with respect to criminal justice responses. Like some other countries, Belgium was dominated by neo-classical penology, as illustrated by the 1867 Penal Code. The latter was the work of J. J. Haus, and may be viewed as a monument of neo-classical thinking, like Italy’s Zanardelli Code in 1889. Neo-classical criminal law, designed by criminal justice experts, is pervaded with moralism and humanism. Closer to Kant’s retributivism than to Bentham’s utilitarianism, these neo-classical writers stressed the offender’s subjective responsibility. They felt that sentences should be individualized, for the sake of fairness, and made proportionate not only to the seriousness of the offense but to the offender’s moral responsibility as well (Haus, 1874, 53). The outcome of this consideration was mainly the introduction of the clause on mitigating circumstances, symbolized in continental Europe by a French law dated April 28, 1832, which brought it into general use. The widespread recourse to mitigating circumstances enhanced the effectiveness of punishment. In a number of cases the harshness of the sentences prescribed by the 1810 Napoleonic Code, in force in Belgium until it was reformed in 1867, led to a sort of impunity. Indeed, juries hesitated to pronounce sentences they deemed disproportionate (Donnedieu de Vabres, 1947, 32). But mitigating circumstances also contributed to making punishment less harsh, at a time of growing political and social tension sparked by class struggle in Belgium and elsewhere in Europe (Van Outrive et al., 1991).

7In the late 19th century, the humanism characterizing neo-classical criminal law gradually came to be viewed as sentimentalism (Simons, 1908, 543). In Europe, this period marked the height of the free-market economy and industrialization, producing such undesirable effects as extreme poverty and beggary, alcoholism and prostitution, housing and sanitation problems (Digneffe, 1995; Debuyst, 1988). As elsewhere in Europe, the laboring classes were perceived as dangerous classes (Chevalier, 2000), some elements of which were defined as an underclass of dangerous individuals (Pratt, 2000, 37-39). Composed mostly of petty recidivists, habitual offenders, professional criminals and juvenile delinquents, the category of dangerous individuals also includes mentally ill (insane) offenders. The latter were found irresponsible if insane, thus avoiding punishment. They were subjected to administrative caretaking which escaped the control of the authorities in charge of punishment. Insane offenders who were not labeled demented were found responsible for their acts and sentenced, but with mitigating circumstances (Ruttiens, 1911, 486-487; Passelecq, 1913, 354-355). They were then given short prison sentences and soon faded into the big cities’ anonymity, joining the horde of offenders. As Adolphe Prins pointed out, the neo-classical method is almost paradoxical here: the more its ideas on responsibility are respected, the less it is actually capable of protecting society against crime (Prins, 1910, 71). In effect, it leads to a system of short but always repeated sentences, and the granting of pardons and releases (Prins, 1912, 150).

2) The positivist reaction: individualization aimed at reducing the risk of offending

8The inability of the neo-classical approach to cope with the risk of offending is clearly evidenced by figures on recidivism, which became an obsession at the close of the century (Prins, 1905-1906; Robert, 2002). The criminal anthropology movement that emerged then in Italy served as a basis for the construction of criminology as a discipline. Well described by Michel Foucault, this new interpretation of offending backed by science and determinism changes the grounds for individualization of punishment rather than eliminating it. According to orthodox positivism, individualization of punishment is no longer based on the offender’s subjective moral responsibility, but clearly on his social responsibility or social dangerousness (Foucault, 1978, 1981).

9In the context of the new scientific positivism, the popularity of theories on degeneracy and heredity, and the fear of crime as a threat to the social order, criminological discourse offering an explanation of crime and criminals was increasingly popular. It was grounded in new postulates which were attractive to neo-classical thinkers but also elicited some distrust (Saleilles, 1898; de Lantsheere, 1910, 916). These postulates may be summarized as follows. First, the primary goal of criminal law is to serve the protection of society and order-maintenance, rather than the ideal of “fair” sanctioning of moral failings. Second, the metaphysical search for the offender’s moral responsibility must be abandoned and replaced by the investigation of his dangerousness as the basis and measurement of punishment (Cornil, 1946, 14). Third, in this context, with dangerousness linked to degeneracy and heredity (Da Agra, 1988, 89), abnormals represented a specific target for the penal reaction. The targets were recidivists, habitual offenders, professional criminals as well as the criminally insane, who constitute the army of crime and the epitome of dangerousness. They were perceived as “the dregs of society”, irremediably deficient, the problem being how to sort them out and eliminate them (Prins, 1912, 148). Fourth, judges are obliged to share their power with physicians. At a time when science is the only authority (Lombroso, 1871, 10), expertise became central to the identification of dangerous offenders. In compensation for the relative decriminalization implicit in the thinking of the Italian school, there was to be a medicalization of punishment (van de Kerchove, 1981b). This medicalization occurred within a backdrop of conflicting psychiatric approaches, between deterministic and dynamic schools, the former advocating interpretation of the individual’s discourse and the latter preferring probabilistic and actuarial quantitative approaches based primarily on risk scales (Debuyst et al., 2008, 53-271; Harcourt, 2011; Rafter, 1997, 168-173). The point is clearly interesting, inasmuch as the same conflicting approaches are at work today, as will be shown below.

  • 1 Bulletin de l’Union Internationale de droit pénal, 1889, 4.

10In the late 19th century the thinking of the Italian positivist school pervaded the Western world. The role of the International Union of Penal Law, created in 1889 by Prins, von Hammel and von Lisz, respectively Belgian, Dutch and German, was decisive in this respect. With its conferences and debates, the International Association of Penal Law also gave birth to comparative criminal law, the outcome being the promotion and development of a sort of shared criminal law in a protection-of-society-oriented Europe (Tulkens et al., 2010, 110). But as the debates within that International Association clearly showed, the limits of a criminal justice project on dangerousness entirely articulated around the protection of society were soon criticized. And whereas the original statutes of the Association called for “the unreserved recognition of the doctrine of positivism”,1 it soon adopted a far more moderate position. Concretely, there was a new awareness of the possibility that the positivist project might endanger the basic principles of classical criminal law. There were allusions to the risk of the demise of criminal law (Saleilles, 1898, 95), while other people stressed the need to reconcile the scientific study of offending with respect for republican legality (Donnedieu de Vabres, 1951, 197). In point of fact, it was mainly the idea of offending as law-breaking that was threatened, inasmuch as the idea of dangerousness implied acting against a dangerous individual representing potential acts (Foucault, 1981, 421). Secondly, it touches the idea of lawful and proportional punishment, if the concern with reducing the risk of offending leads to the adoption of safety measures or other forms of preventive detention for unlimited duration. The risk then is that of arbitrary criminal law (Garçon, 1909, 395), or even a totalitarian abuse of criminal law incompatible with the principles of a democratic state governed by the rule of law (Tulkens, 1988, 46). This danger was to lead advocates of the protection of society to seek arbitration or a compromise. In Belgium, it was Adolphe Prins who embodied that middle path. His writings also served as a basis for the legislation on dangerousness that saw the light in Belgium at the turn of the 20th century, including the 1930 Social Defense Act Regarding Abnormal and Habitual Offenders.

3) Adolphe Prins and the protection of society in Belgium: towards a conciliation of views, or legislation on dangerousness on the fringes of criminal law

  • 2 Prins, 1886, 1899, 1910.

11In Belgium, Adolphe Prins (1845-1919) is the main representative of the movement for the protection of society (défense sociale). This specialist of criminal law, author of several books,2 introduced a number of “laws and bills for the protection of society”, which complete the Belgian neo-classical criminal code of 1867 (Tulkens et al., 2010, 115-116).

12Quite soon, Prins was to opt for a conciliatory attitude towards the neo-classical school, taking the same path as the International Association of Penal Law. The compromise revolved around two points. First, the new criminal law advocated by Prins would be marked, primarily, by the basic principles of classical criminal law. The punitive logic would not be relinquished for “normal offenders”, and indeed, any other attitude would not be acceptable to the general public (Prins, 1912, 141). Individualization of sentence-serving is in fact conceivable for those offenders, who are susceptible of adjustment and rehabilitation (Prins, 1905-1906). Conversely, for abnormals, those “degenerate offenders”, or “social scum”, against which society must be protected, a regime of safety measures on the fringes of criminal law must be considered (Prins, 1912, 142). What Prins actually suggests is a principle of bifurcation of penal policies (Bottoms, 1997), based on dualism and complementarity between guilt and dangerousness (van de Kerchove, 1981a). This point is essential. For neo-classical thinkers, it is the fact that the security legislation is located outside of criminal law (Cornil, 1930, 14) that makes it acceptable. Secondly, security measures must be given a legal framework. They can only apply in situations defined by the law and cannot be transformed into purely preventive measures taken ante delictum (Garçon, 1909, 398).

13In his desire to restrict the security regime to some specific types of dangerousness, Prins heralded Belgium’s future Social Defense Act of 1930. He even foreshadowed one point of discussion as to the nature of security measures. Prins was quite clear on this point: he did not consider security measures as punishment. The goal is different, and the places in which measures are served are not the same (Prins, 1912, 50).

II - The 1930 Social Defense Act Regarding Abnormal and Habitual Offenders: the legal framework

1) Mentally ill (insane) offenders, the main (but not the exclusive) target of the 1930 Act

  • 3 The source of the 1930 Social Defense Act is found in a bill dated April 15, 1880 regarding the org (...)

14Compared with the other dangerousness laws passed elsewhere in Europe at the time, the 1930 Social Defense Act Regarding Abnormal and Habitual Offenders dated April 9, 1930 was a latecomer. The grounds had been laid, however, as early as 1880,3 and unsurprisingly, that first law looked a great deal like Prins’s protection of society project. This law bore one striking feature: whereas most of its content was devoted to “abnormals”, which is to say the criminally insane, its chapter VII nonetheless discusses the case of recidivists and habitual offenders, included in the same category of dangerousness and abnormality as the mentally ill. The present article focuses exclusively on the question of the criminally insane, for the reasons mentioned above.

  • 4 The deaf and dumb were also mentioned (art. 76 CP), but did not represent a political problem.

15The only abnormals referred to in the 1867 Belgian Criminal Code are the insane (art. 71 CP).4 Insane (demented) offenders were classed with mentally ill non-offenders and excluded from criminal prosecution. They were subjected to collocation measures and sent to a mental hospital, in which they were exempt from any control by the criminal justice system. A second problem group was composed of the half-mad, those abnormals located somewhere between insanity and responsibility. They were granted mitigating circumstances and given short prison sentences after which they contributed to figures for recidivism.

  • 5 M. Janson, Minister of Justice, Annales de la Chambre, session 1927-1928, 81.

16The two groups of abnormals were the main target of the 1930 Social Defense Act. As Chief Prosecutor L. Cornil pointed out, this new law does not affect the foundations of the right to punish and is definitely a complement to the classical criminal regime aimed at protecting society against some specific categories of dangerous individuals (Cornil, 1930, 14-15). Those abnormals targeted by the new law included the insane and individuals “in a state of severe mental imbalance or mental deficiency such that the defendant is unable to control his acts” (art. 1 of the 1930 Social Defense Act). In these two cases, the mentally ill offender leaves the criminal justice system and takes the “protection of society” path, provided he meets three requisites: he has committed a felony or misdemeanor making him liable to a prison sentence of at least three months; he is declared to be in a serious state of insanity or mental imbalance; he represents a danger for society. The law has a twofold effect, then: first, insane offenders are no longer subject to administrative management but are subjected to the more security-oriented social defense regime. Second, the law extends the regime of criminal irresponsibility to people who are not quite mad, and not quite sane, but who represent a danger for society, to some extent.5 However, this apparent decriminalization is balanced by the shifting of these “half-mad” individuals to the security regime for the protection of society, whereas they were previously given only lenient sentences.

2) The protection of society regime: an observation period and internment for an unlimited period

17The regime set up as of 1930 to protect society against mentally ill offenders was based on a twofold measure, both medical and legal. The first aspect involved placing the offender in observation within the psychiatric wing of a prison (art. 1 LDS). Such wings existed in Belgium since 1920, where they were introduced by Dr. Vervaeck, head of the Correctional Anthropology department and strong supporter of Lombroso’s theories (Vervaeck, 1926). These wards already contained prisoners and mentally ill criminals awaiting forensic investigation. This first measure, instating an observation period, evidenced the new importance of forensic psychiatry in the protection of society system. It aimed at improving the practical conditions surrounding the expert assessment of responsibility and dangerousness used to guide the defendant’s future orientation.

18Which expert? The physician in the psychiatric wing is a scientifically apt candidate for the job, but the law says nothing on the subject, and the legal authorities are free to choose any physician. They can also request expert reports from outside specialists, independently of the observation procedure, as was the case prior to the 1930 law (Cornil, 1930, 41-43). Further, article 3 of the law states that the defendant can always receive care from a doctor of his choice. This clause means that in the absence of two initial reports (contradictory expertise) the defendant can take the initiative of producing a counter-expertise report and the latter may pertain to the appropriateness of both the procedure prescribing an observation period and the decision to intern him (Cornil, 1930, 46). Another point is that the psychiatric wing also houses sentenced prisoners whose mental state is judged problematic. When attached to a prison, then, the psychiatric wing is not merely a place for observation, any more than it is impermeable.

19The second measure prescribed by the law is an internment measure that may be ordered by investigating or court judges (art. 7). This measure is pronounced for a “relatively” indeterminate period of 5, 10 or 15 years depending on the seriousness of the act committed (art. 19). However, this period may be prolonged indefinitely, so that the goal pursued, that of preserving freedom and combating arbitrary confinement, is hardly achieved. The measure actually makes it possible to confine an individual who continues to represent a danger for society for his entire lifetime. The 1964 law, reforming its 1930 predecessor, eliminated those time limits, thus transforming “relatively” indeterminate periods into an “absolutely” indeterminate period (van de Kerchove, 2010, 491). Article 23 (which became art. 21 in 1964) prescribes the possibility of sending a sentenced prisoner into internment if he/she is shown to be insane or mentally imbalanced while serving his/her prison term. This clause provides a means of maintaining control of some prisoners viewed as dangerous once their prison term is ended (Cartuyvels et al., 2010, 273).

20Whereas internment is justified primarily by the internee’s alleged social dangerousness or by the risk of recidivism, his his/her release can logically be considered once that risk seems to be under control. In 1930, the law stated that release, on probation or final, was possible when “the internee’s mental state has improved to the point where he/she no longer represents a danger for society” (art. 19, 1930 Act). The 1964 law was to replace the notion of “social dangerousness” by an expression indicating that the internee meets “the conditions for social rehabilitation” (art. 18, 1964 Act). Release on probation is a possibility, but the 1930 law provides that the internee must be subjected to psychiatric supervision for at least one year (art. 21, 1930). The 1964 law replaced that psychiatric supervision by a “medico-social guardianship” (art. 20, 1964 Act). In 2000, further legislation subjects internees having committed an offense of a sexual nature to additional restraints (art. 20, 1964 Act). Release, probationary or final, is also subordinated to the reasonably founded consent of a bureau specialized in the guidance and treatment of sex offenders (art. 20bis, 1964 Act). Moreover, if the internee released on probation gives any signs of social dangerousness or does not respect the stipulations imposed on him, he may be sent back to the psychiatric wing (art. 20, par. 7, 1964 Act).

3) The nature of the internment measure and its implementation: the ambivalence of a half-care, half-safety measure

  • 6 Projet de loi, Exposé des motifs, Doc. Parl., Chambre, session 1922-1923, n° 151.

21The enactment of the Social Defense act in 1930 raised issues which are still relevant. The first involves the nature of the internment measure. In line with Adolphe Prins’ reasoning, the 1930 legislators felt that the internment measure was not punishment. Abnormals should be treated “not as offenders, but as sick”. Internment is thus presented as “a social, humane measure”, and the abnormal person is subjected to a “scientifically organized” treatment regime.6 Shortly thereafter, however, the distinction between punishment and human measure was criticized as fictitious. Several high-ranking judges pointed out that it is in fact an improved punishment (Cornil, 1930, 50-51). This type of measure indeed produces the same suffering and infamy as punishment, and would be experienced by the internee as punishment and perceived as such by the public. The fact that the punishment is called internment and that the prison in which it is served is called a “psychiatric wing of a corrections center” makes no difference (Leclerc, 1930, 215).

22The second issue concerns the place where the internment is to take place. Where should mentally ill (insane) criminals be put? This question arose long before the 1930 law. Throughout the 19th century the balance swung between prison/mental asylums (introduction of a psychiatric ward in prisons) and the creation of mental asylum/prisons (creation of a custodial unit within the mental asylum) (van de Kerchove, 1988). The 1930 law mentions “special facilities” without specifying their nature. In fact, psychiatric wings, initially designed as places for observation, were chosen for internment, which would symbolically and practically reinforce the punitive character of this treatment and security measure. The 1964 law also authorizes internment in a “private facility”, that is, a general psychiatric hospital, so as to reinforce the curative nature of the measure (art. 14).

  • 7 For the situation in Flanders, see Cosijns et al., 2008; Vandenbroucke, 2009.

23What is the situation today? We are confronted with a particularly complex situation, with a combination of places of internment and different practices in the Dutch-speaking north and the French-speaking south of the country.7 Generally, internees are first sent to psychiatric wings of prisons, for a large part yet. However, since these wings are overpopulated, some are sent to prison wards for sentenced prisoners (Goorden, Oei, 2007). In addition to the psychiatric wings of prisons, the French-speaking region has an Institute for Social Defense (Établissements de Défense Sociale – EDS) located in Paifve, with a capacity of 208, while the Flemish Region has a high-security section for treatment (“De Haven”) at the Merksplas prison, with a capacity of 60. The latter two facilities are subjected to the correctional authorities’ rules and correspond to the prison/mental asylum model. In French-speaking Belgium, two other EDSs have been created in Mons (30 places) and Tournai (376 places), but these “legal psychiatry centers” are attached to the Ministry of Health. They correspond to the mental asylum/prison model and are theoretically more medically oriented than punitive. There is no equivalent, in Flanders. Responding to critiques pointing out the lack of healthcare and specific institutions for internees, the construction of two top-security legal psychiatry centers for high-risk internees has been planned in Ghent and Antwerp (with a total capacity of 390) and the Gent’s institution opened its doors in May 2014. In addition, in conformity with article 14 of the 1964 Social Defense Act, some general psychiatric hospitals, both public and private, accept internees. These psychiatric hospitals tend to accept medium risk internees (De Vuysere; 2005, 265) or those on probation who are felt to represent a lesser risk. Last, there is also a network of psychiatric hospitals, homes, and sheltered living services offering ambulatory support for low-risk internees (Casselman, 2011, 238), or those on probation. The variety of care facilities ranging from prison structures to ambulatory care networks illustrates the ambiguity of the internment measure, caught between risk management and treatment.

  • 8 Figures provided by the General Directorate of Correctional Facilities, Powerpoint presented by the (...)

24In absolute figures, there were 1,131 internees in psychiatric wings of prisons and in the Paifve EDS as of August 27, 2013, for a prison population of 11,475. The average daily number of internees under the control of the Ministry of Justice and subjected to a custodial regime therefore represents 10% of the prison population. In addition to these internees in prison/asylums, there are (as of February 5, 2013) 413 inmates in the two Walloon Region EDSs attached to the ministry of Health and 135 placed in private psychiatric hospitals on the basis of article 14 of the law. There are also 2,165 probationary released internees8 for a total of 2,713 internees subjected to a regime theoretically focusing on treatment. The overall population of non-released and probationary-released internees is therefore approximately 3,800, representing a substantial increment with respect to the 2004 figures, where that same group represented 3,306 individuals (Cosyns, 2005, 5). The trend, then, is clearly upward, and is all the more significant when we look further back in time: between 1997 and 2011, the number of internees in prison doubled. This means, essentially, either more time spent in internment or a tendency for CDSs to grant releases later than in earlier days (Mary, 2013, 112; Mary et al., 2011).

25Last, the ambivalent nature of the protection of society system is further illustrated by the creation, in 1930, of a medico-legal agency in charge of the implementation and the follow-up of internment. Once the judge decides to send the defendant within the protection of society system, the Commission de Défense Sociale (CDS), or Board of Social Defense, takes over. This board, composed of a judge, a lawyer and a physician from the psychiatric wing, is in charge of assigning the internee to a facility, deciding any transfers to another establishment, and granting probationary or final release (art. 14 of the 1930 law). The 1964 law made no fundamental changes in the composition or prerogatives of this board, which is still active at present (art. 12 of the 1964 law). The composition of the CDS reflects the desire to retain a balance between the justice system’s concerns as keeper of the social order (with the judge), respect for individual liberties (the lawyer) and the need to produce scientifically grounded decisions (the specialist). Moreover, the creation of this board indicates the intention to develop collaboration between judges and physicians. In practice, the CDS functions very consensually, in contrast with the argumentative model characterizing legal culture. This reflects an implicit shift in decision-making from judges to medical practitioners (Cartuyvels, 2012, 156), along with a change in the role of lawyers, often closer allies of the Social Defense system than defenders of their clients’ rights. In accordance with the welfare law logic, legitimacy shifts from judgment based on moral values to judgment based on the authority of science.

III - Some practical issues: care and rehabilitation versus risk and control

  • 9 Cartuyvels et al., 2010.
  • 10 Cartuyvels, 2012 ; Cliquennois, 2012.

26There has been very little empirical investigation of the Belgian protection of society system (Casselman, 2011, 245). The present text is based primarily on two recent qualitative studies, one devoted to the dialectic between care and control,9 the other to the question of risk and its impact on the trajectories of internees,10. It offers a series of reflections on three key issues pertaining to the law and its enforcement. The first has to do with the purpose of internment, oscillating between care and control; the second is the question of power and truth, at stake in expertise; the third is the impact that concerns about risk, or dangerousness, have on the internee’s trajectory, from the internment decision up to release. We have added to the results of our research those resulting from other empirical inquiries.

1) Internment: an incapacitating measure between care and control

  • 11 ECHR, 30 July 1998, Aerts vs. Belgium.

27In Belgium, the dilemma of care versus control (Adams, Ferrandino, 2008) is as old as the law for the protection of society. Mary internees are still currently in psychiatric wings of prisons, especially in Flanders (Cosijns et al., 2008). This accounts for criticism primarily focused on the lack of care dispensed in these prison-asylums (Cosijns et al., 2008). And rightly so, since Belgium was indeed condemned a first time by the European Court of Human Rights in 1998, on the grounds that detaining a mentally ill offender in the psychiatric wing of a prison without providing real psychiatric treatment was a violation of article 1 of the European Convention for the Protection of Human Rights.11 Several years later, in 2003, the psychiatric wing of the Lantin prison was closed following harsh criticism by the European Committee for the Prevention of Torture and Inhumane or Degrading Treatment or Punishment (CPT).

  • 12 Administrative instruction n° 1800 dated June 7, 2007.

28The Belgian authorities reacted in 2007 to the condemnations pronounced by the European Court of Human Rights (Cliquennois et al., 2014): a “multidisciplinary treatment team” was introduced in each psychiatric wing as well as in the Lantin EDS, in order to improve the quality of care delivered to internees.12 The creation of treatment teams went hand in hand with the decision to separate the task of care (assigned to a treatment team) and that of expertise (assigned to a psycho-social team) in the psychiatric wings of prisons and in the Paifve EDS. The improvement in care since 2007 has proved quite insufficient as several researches have pointed out. In particular, an empirical research conducted both in psychiatric wings of prisons and in an Institute for the Protection of Society has shown the primacy of the prison structure and the priority given to a security-oriented model in psychiatric wings. This empirical research has also stated the insufficient involvement of psychiatrists (who are employed as independent workers and work part-time in several prisons), the difficulty in recruiting trained personnel for jobs (that are neither professionally nor financially stimulating) as well as the lack of specific training for caregivers. These wings are still perceived by those who work there as “dumps”, where internees are viewed more as “prisoners” than as “patients”, as evidenced by the vocabulary used (“prisoner”, “cell”) (Cartuyvels et al., 2010, 120; 264). The care provided in psychiatric wings is a far cry, then, from the standards found in psychiatric hospitals (Vandevelde et al., 2011, 76). The situation is hardly better at the Paifve EDS. This facility, which is attached to the ministry of Justice, retains an organization and regime modeled after the correctional system. The personnel is mainly composed of wardens from correctional administrations, who are sometimes sent there without having requested the position, as punishment or for practical reasons unconnected with any specific skill (Cliquennois, 2012, 42). As in the prison wings, a multidisciplinary treatment team has been set up and the separation between care and expertise tasks put into practice. But the Paifve EDS encounters the same difficulties observed in psychiatric wings of prisons. In reality, the priority given to security in the detention regime, the lack of hospital staff and the wardens’ lack of specific training leave little room for care (Cartuyvels et al., 2010, 124 and 147).

  • 13 See for instance Civ. Charleroi (ref.), 25 February 2005, Journal des Tribunaux, 2005, 308, note L. (...)
  • 14 Cass. 26 March, R. G. n°C.09.0330.F; C.C., 17 September 2009, n°142/2009.
  • 15 ECHR, 10 January 2013, Claes v. Belgium ; ECHR, 6 December 2011, De Donder and De Clippel vs. Belgi (...)
  • 16 ECHR, 9 January 2014, Saadouni vs. Belgium, §56 and 61. On this jurisprudence, see Cliquennois et a (...)

29These issues have been underlined by the Belgian judicial power. Over the last decade in particular, several condemnations have been pronounced by Belgian jurisdictions that have forced Belgian authorities to remove and transfer an internee from the psychiatric wing of a prison to an Institute for the Protection of Society13. Additionally, trial judges, the Belgian Civil Supreme Court as well as the Constitutional Court have raised the same issues14. However, the harshest reactions and the highest pressures have come from European authorities. Both the Committee for the Prevention of Torture and the European Court of Human Rights have been really concerned with the incarceration of internees in the psychiatric wings of Belgian prisons. Belgium has been condemned again repeatedly from 2011 to 2014 by the European Court of Human Rights for violation of articles 3 and 5 of the European Convention of Human Rights.15 The Strasbourg Court has clearly underlined several times the inadequacy of psychiatric wings for therapeutic and medical ends and the existence of a structural problem in Belgium regarding the management of offenders suffering from mental illness and disorders16. We can also mention as an additional evidence of the significance of this issue the recent request expressed by an internee to be either transferred into a psychiatric facility located in the Netherlands or to be granted an euthanasia (Herremans, 2015).

  • 17 ECHR, 3 February 2015, Smits vs. Belgium, §74.

30The situation is better, at least on the institutional level, in the two EDSs in French-speaking Belgium, which are attached to the ministry of Health. The caregiving staff is definitely larger and the team is headed by the psychiatrist. The wardens are gradually being replaced by specialized educators and nurses. The care model is more substantially enforced, which also translates into the use of a vocabulary more in step with that of psychiatric hospitals (“patient”, “seclusion room”) (Cartuyvels et al., 2010, 125 and 264). Last, the care model is a fortiori more developed in conventional hospital facilities and in out-patient care networks. The difficulty there, however, is in having internees transferred to intermediate facilities of this type, which are quite reluctant to receive patients with the “protection of society” label. With regard to this issue, the European Court of Human Rights has deplored that institutions in charge of mental disorders cannot force appropriate and external institutions to house internees17.

31How can care be delivered in closed units dedicated to the protection of society? The experience of psychiatric wings (and of EDSs in French-speaking Belgium) shows three things: first, that care is to a large extent reduced to medication, for order-maintenance and security reasons. The administration of drugs or a change of medication is often the first response when an internee poses any problems, medication is used to stabilize his condition and prevent him from making trouble, and not really for treatment purposes. As a result, it is not unusual to see drugs distributed systematically, with some forms of overmedication or even the forced distribution of drugs, although the August 22, 2002 law on patients’ rights no longer authorizes this (Cartuyvels et al., 2010, 153-155). Secondly, the position of the prison personnel within these wings produces a “new panoptic vision” exerted by wardens over the other professionals, including those in charge of care. These wardens are constantly present, and occupy a central room equipped with the implements of security and control (including cameras and keys) enabling them to monitor their colleagues. They occasionally replace the therapeutic personnel in some tasks such as distributing medicine, and are able, then, to shape the actual ‘therapeutic’ practices In a security-oriented context, order-maintenance and avoiding attacks are more important than treatment, as shown by the pressure in favor of overmedication (Cliquennois, 2012, 108). Third, the separation between care and expertise tasks as well as the considerable funds allotted to the latter tend to favor expertise and risk assessment to the detriment of care and clinical work (Cartuyvels et al., 2010, 167).

2) Expertise in the protection of society theory: power relations and the truths at stake

32Expertise plays a major role in the protection of society theory. It is brought to bear from the outset, as soon as a decision on the legal responsibility and dangerousness of the defendant is required. It is called upon all along the trajectory, with expert reports written by the psycho-social teams in charge of expert evaluation within the psychiatric wings of prisons and the EDSs. Expertise also plays an important role at the end of the trajectory, when the Board of Social Defense has to decide whether to release the internee on probation or definitively.

33Expertise raises many issues, including its depreciation and insufficient resources, its expeditious dispatching and the place and way it is done, and the status of the specialists involved (Pham et al., 2007, 53-54). Here, we have chosen to focus on two specific issues. The first has to do with the power relations that develop around expert evaluation. The second points up the conflict between two models of expertise, against a background of opposition between hermeneutical and probabilistic approaches.

2.1. Judges and psychiatrists: a conflict-laden collaboration

34The 1930 Social Defense Act clearly evidenced the rise of psychiatric power over the judicial, in comparison to the earlier situation (van de Kerchove, 1983, 381). The legal texts emphasize the importance of psychiatric monitoring and of expert reports at several stages of the course prescribed for the protection of society. These texts also sanction the presence of a psychiatrist within the key decision-making authority, the board of social defense (Commission de Défense Sociale). Joint decision-making became the rule, and is viewed by some as evidencing a shift in power from judges to psychiatrists owing to the authority afforded by the latter’s knowledge (Martens, 2004).

35In practice, the medico-legal continuum for the protection of society produced more ambiguous relations between these two protagonists. Medical rationality, based on truth of a scientific nature, unquestionably weighs on court verdicts, viewed as decisions based on value judgments. Over time, a “sovereign expertise” model developed (van de Kerchove, 1983, 390), reducing both the judge’s decision-making power and the role of lawyers. The latter are quite powerless when faced with the expert’s truth, and are reduced to playing the role of auxiliary of the law rather than defending their client (Matthijs, 1965, 167). Here, scientific discourse challenges the very essence of the judicial process, based on an exchange of arguments and the need for an adversarial hearing. But actually, judges both acknowledge and question this delegation of power. On the one hand, they frequently if not always demand expertise in socially sensitive cases such as those involving sex offenders. They tend to prefer an overconsumption of expertise, for fear of making the wrong decision (Cartuyvels et al., 2012, 89-90). If two expert reports are contradictory, judges tend to call in a third team of experts rather than making a decision on their own. Debate between experts is given precedence over legal debate here, with judges preferring to hide behind scientific truth rather than shouldering responsibility for deciding on the basis of the law. This shift of power and responsibility, constantly stressed in reports on cooperation between the various actors within the criminal justice system, is also indicative of the growing difficulty in accepting risk-taking in security-centered societies (de Coninck et al., 2005). But on the other hand, judges relativize the importance of expertise. They explain that expert reports are often contradictory, that their assertions are relative and that expertise is only one of a number of pieces of information on which to base a legal decision. Faced with the uncertainties of science, judges assert their decision-making role, reviving the legal tradition in which the court verdict is a wager with a doubt in the background (Garapon, 2008, 118).

36Nonetheless, in the present protection of society system the forensic psychiatrist’s role remains decisive for at least two reasons. First, the initial expert report, which commands the defendant’s entering the protection of society system, turns out to be primordial. This psychiatric report labels the internee and orients his entire subsequent trajectory, including transfers to one facility or another, the granting of leaves, or the decision on probationary or definitive release (Cartuyvels, 2012, 161). Second, a forensic psychiatrist sits on the board of social defense, the medico-legal organ in charge of monitoring internees through the protection of society circuit. In this consensus-based body, the psychiatrist is officially a co-decision-maker and the scientific knowledge he represents is quite systematically decisive (Cartuyvels et al., 2010, 260).

2.2. Expertise, between clinical diagnosis and probabilistic prognosis

37The second issue has to do with the status of expertise as truth, given the opposition between psychodynamic and organicist approaches to mental health (Strauss, 1992). The importance of expertise in criminal justice depends on the degree of truthfulness and the evaluative and predictive efficiency ascribed to it.

  • 18 The situation may be different in Flanders, where Anglo-Saxon approaches are more prevalent.

38In French-speaking Belgium, expert evaluation definitely gives priority to the clinical approach.18 This is mainly due to the prevalence of psychodynamic approaches throughout French-speaking Europe. In France as well as in French-speaking Belgium, psychoanalysis is still very important, in spite of repeated attempts by the cognitive behavioral school to destabilize it (Meyer, 2005). The emphasis placed on speech as clinical praxis, stressed by Foucault in opposition to the clinical gaze, is still dominant for psychiatrists in their training and practice, irrespective of their field of intervention. So, for those psychiatrists in charge of expertise for the protection of society, the anamnesis of the patient’s life and what the latter has to say about it remains essential. It is much more important for example than the “objective” facts with which the patient is charged, be they offenses, past problems with the law or other elements in the legal file. Some experts actually say they refuse to read the legal file or only read it after the clinical interview, to avoid having it disturb their impressions (Cartuyvels et al., 2010, 95-96). These experts, who remain clinicians first and foremost, are also quite satisfied with the questionnaire on responsibility/dangerousness that they must fill out for the justice department. The fact that the questionnaire is open enables them to submit a “personality report” and does not confine them to a binary diagnosis of responsibility versus dangerousness. This personality report allows them to use a clinical approach in step with their therapeutic function. It also enables them to cast a veil on the policing, social control function of the diagnosis of dangerousness, which many psychiatrists do not easily accept. As clinicians, psychiatrists resist this logic of diagnosis that makes them, on the grounds of their truths, take the responsibility for judging, something which society (and judges) are often reluctant to do.

39This clinical approach to expertise has fallen into disrepute in the English-speaking world since the 1960s (Cocozza, Steadman, 1976). It is viewed as subjective and of poor predictive value (Monahan, 1981), and is also accused of having consistently over-evaluated dangerousness (Steadman, Morrissey, 1981). It has been widely replaced by actuarial skills focusing on the identification of risk-related profiles (Hannah Moffat, 2006). In Belgium, the clinical approach to expertise was first called into question by the use of psychological testing. In recent years psychologists have appeared on the Belgian expertise scene. They are more competent than most psychiatrists in the use of batteries of tests and risk scales, and develop competing forms of knowledge, thus threatening the psychiatrists’ position and power in the field of expert evaluation. In psychiatric wings and EDSs, psychological testing has also been developed within the psycho-social teams in charge of evaluating internees. In a psycho-social work milieu where the psychodynamic model is still prevalent, although systemic and cognitive behavioral approaches are also present (Pham et al., 2010, 459), this trend elicits resistance among some members of the therapeutic staff. Depending on their theoretical leanings, the personnel either endorses the logic of “objectifying” psychological testing or denounces the “desubjectifying” use of tests to the detriment of relating to and dialoging with patients. Similarly, the poverty of a “photographic” diagnosis is opposed to the richness of “dynamic” expertise taking the internees own words and evolution over time into consideration (Cartuyvels et al., 2010, 133-134).

40More broadly speaking, behind the critiques addressed to the use of technical evaluation models we perceive the fear of a transformation of the nature of mental health professionals’ work, to the detriment of care (Rose, 1998, 177). From this viewpoint, psychological testing may be seen as the spearhead of more complete technologizing of the expertise process, linked with actuarial risk scales. Use of the latter, encouraged in some academic circles (Pham et al., 2007), is recommended to combat the uncertain, unequal character of clinical expertise and put an end to dissension among experts. Over the years, the Anglo-Saxon model of actuarial “criminological expertise” has had an impact on expert evaluation in Belgium, probably more so in Flanders than in French-speaking parts, the ambition being to replace “mental expertise”, considered archaic (Benezech et al., 2009, 43-44). The specialist’s fluid clinical sense is contrasted here with the need for solid, standardized diagnostic screening and assessment instruments (e.g. HCR-20.PCL-R or the Hare Scale) based on high international standards (Vandevelde et al., 2011, 74-75). In the latter expertise model, the expert’s task is overdetermined by the question of predicting the risk of offending. The construction of risk scales is of course still uncertain and the reliability of predictions as to the future behavior of internees is very relative. Nonetheless, trust is placed in systematic knowledge obtained by crossing various statistical variables and claiming greater objectivity than psychiatric expertise. Moreover, the expert’s social control mission is clearly assumed here, and no longer seems to cause any ethical quandaries.

  • 19 In France, the violent debate elicited by the Inserm’s collective investigation on “The screening, (...)

41This call for a change in the expertise paradigm is anything but trivial. For one thing, it may well evidence the revenge of the actuarial, probabilistic sciences aimed at prediction over the determinist, comprehensive approaches that dominated the late 19th century (Maurutto, Hannah-Moffat, 2006, 441). In addition, it produces a shift from the imaginary of dangerousness to one of risk. The idea is, indeed, not so much to deliver an individual diagnosis of dangerousness based on an understanding of psychopathological mechanisms as to establish a prognosis based on the individual’s consistency with a scientifically defined risk profile (Castel, 1983). This legal application of the precautionary principle, the child of an alliance between the neurosciences and cognitive behavioral theory (Inserm, 2005), results in a logic of risk detection that bespeaks the trend toward a security-oriented mentality encouraged by our victimization-obsessed society19. This expertise model is to the liking of the Board of Social Defense. Actuarial knowledge is viewed as a valuable additional source of knowledge, especially in cases of internees who are sex offenders. It is, moreover, sufficiently flexible to include social and clinical data. This type of knowledge also generates distrust, however, since its ability to categorize seems relevant at the group level but is ineffective at the individual level. For the internee as a unique individual, its predictive ability turns out to be relative and to be as much in the domain of a “wager” as a judgmental decision. Furthermore, on the ethical level, orienting the course of an individual’s life on the basis of a statistical risk profile is contrary to the mentality of judges, whose idea of criminal justice is entirely constructed on the notion of individual responsibility (Cartuyvels, 2012, 154-155).

3) The internee’s trajectory and release: the primacy of risk reduction

42The questions of dangerousness and risk deeply impact the protection of society system. This is naturally the case at admission, when the existence of a threat to society conditions the internment decision. But it is also true throughout the trajectory of the person confined for the protection of society, until a probationary or final release decision is handed down.

43The course followed by social defense internees points up three facts. First, the distribution of internees is at least partially conditioned by the danger they represent for others or for themselves. In Flanders, high-risk internees are presently held in psychiatric wings. For reminder, two EDSs for high-risk internees are now being built in Antwerp and Ghent, to relieve those overcrowded wings, and should open in 2015. In the French-speaking part of the country, the wings of the Forest prison in Brussels and the Lantin prison in Liège apparently act as top security wings for the country’s other wards (Cliquennois, 2012, 114 and 121). Second, dangerousness seems to act as an organizer of custodial and medical spaces within these psychiatric wings and EDSs. Internees are also assigned to floors and quarters on the basis of a logic aimed at reducing the risks of acting-out towards themselves or others. This assignment may be based on a judgment made from afar (using the criminal curriculum of the freshly arrived internee) or on a face-to-face evaluation depending on the internee’s attitude within the facility. This would condition assignment to a (closed) cell or to an (open) community regime (Cliquennois, 2012, 115-116). Third, an internee’s dangerousness may encourage or discourage transfer from one institution to another. For instance, a dangerous internee is occasionally transferred to give the wardens some breathing space or to avoid having a problem situation escalate. Conversely, a label of dangerousness hinders transfer of internees to hospitals or outpatient circuits, as shown by the lack of room for internees in hospital facilities or specialist health services (Vandevelde et al., 2011, 73).

44The impact of the criteria of dangerousness is just as great when the time for release arrives. Although release on probation is not a compulsory phase according to the 1930-1964 law, it is in fact the rule for the vast majority of internees. What are the criteria defined by Boards of Social Defense for granting or refusing probationary release? In the words of the law for the protection of society, the idea is to make sure that “the internee’s state has improved sufficiently and that the conditions for his rehabilitation are satisfied”. In practice, alluding to the internee’s improved mental state is a mere formality. It is mentioned in the grounds for the decision only to respect the legal requirement (Van Vlaenderen et al., 2010, 12). The main concern of the board of social defense is actually assessment of the risk of recidivism.

45In reality, four factors influence calculated risk-taking (Cartuyvels, 2012, 159). The first is the seriousness of the offense committed by the internee. Even at the end of his term, the act still seems to be the main indicator of the person’s dangerousness. The objective logic of criminal law, theoretically eliminated by the subjectivist protection of society dimension, surfaces here once again. This is all the more true since reversion to the act and its seriousness sometimes conceals retributive reckoning in a penal sense: even if the internee seems to be cured, (overly) rapid release may pose a problem because of the nature of the offense committed and the punishment it deserves. The second factor is the potential dangerousness of the internee as indicated by the first and last expertise reports. These two reports are the only ones systematically consulted by CDSs. The first report is all-important in that it illustrates the pathology leading to the offense, whereas the threat to be avoided at all costs is repetition of that offense. The last report is important in that it assesses the present mental state and dangerousness of the internee when it is time to take the risk of probationary release. The third factor is the existence (or absence) of a hospital or out-patient facility prepared to take charge of the person given his degree of dangerousness. This external relay is judged especially important, particularly since institutional treatment is viewed as ensuring monitoring of medication. It is not rare, then, for release on probation to be granted but postponed for years until there is room in a psychiatric facility (Van Vlaenderen et al., 2010, 12). The fourth is the potentially incurable nature of some pathologies. In the case of psychopaths, sexual perverts, pedophiles or rapists, extreme caution is exerted and often the internee is never released (Cartuyvels, 2012, 163-166). Here we touch on the extreme limit of the protection of society measure which, as both a care and security measure, is turned into an “elimination measure” (Cornil, 1930). Furthermore, caution is redoubled when the internee has committed sexual abuse. CDSs are exposed here to considerable changes in the social perception of damage and in how this exposition to a perception is linked to decision making, making it much more perilous to take risks with this type of interned offenders. In other words, the socio-political environment of the boards of social defense have changed significantly in recent years in regards to the ways in which victimization is portrayed and discussed. In all cases caution is exerted and the internee’s release generally takes the form of a “slowly but surely” policy relying on a series of phases (leaves, vacations, and so on) before any probationary release is granted (Cartuyvels et al., 2010, 271).

46Fear of recidivism also accounts for the conditions set for probationary release. The law for the protection of society calls for “medico-social supervision” of internees released on probation, the content of which is set by decision of the CDS (art. 20). The social part is entrusted to a judicial assistant and the medical one to a psychiatrist or specialized service provider. The purpose of social supervision is clearly “to avoid recidivism” (art. 20 par. 7) and this supervision may be accompanied by specific conditions imposed on sex abusers (art. 20 bis). The content of medical supervision is defined by the law when sex abusers are involved. The latter are required to “follow guidance or treatment in a service specialized in the guidance and treatment of sex offenders” (art. 20 par. 2). The purpose of medico-social supervision is clearly to subject internees to the closest possible monitoring during the probationary period. In practice, however, there are some holes in the monitoring system, for two reasons: first, the judicial assistant confines himself to monitoring the requisites imposed by the CDS and has very little access to the person’s real living conditions. Second, medical supervision replicates the above-mentioned splits between two categories of practitioners. There are those who endorse their role as expert in charge of social control over the internee and collaborate with judicial assistants, whereas others opt for their therapeutic function and invoke their professional ethic and medical confidentiality to keep their collaboration to a minimum (Cartuyvels, 2012, 169).

47The decision on final release is generally made after a 3 to 5-year probationary release period depending on the CDS in Flanders (Van Vlaenderen, Beyens, 2010, 13) and after somewhere around 5 years in French-speaking Belgium (Cartuyvels, 2012, 173). The law does not set any conditions other than those prescribed for release on probation, which is to say an improvement of the person’s mental health and proof of social rehabilitation (art. 18). The CDS is then faced with the same question of risk assessment as for the decision on probationary release. One would think that the satisfactory completion of the probationary period would be a decisive factor, but this is not necessarily the case. While having been in therapy is often viewed as a positive element, the CDS’ decision again leans heavily on the seriousness of the offense (Van Vlaenderen, Beyens, 2010, 16). Similarly, some CDSs systematically request a last expert report on dangerousness (Cartuyvels, 2012, 173). The seriousness of the acts and dangerousness remain the two decisive factors to the end, then, in assessing the risk of recidivism.

IV - The recent legal reform of the Social Defense Act: between change and continuity

1) The 2007 law on internment of mentally ill offenders: a virtual law?

  • 20 Exposé des motifs, Doc. Parl., Sénat, 2006-2007, 3-2054/1, 4.

48In 2007, two laws modified the protection of society regime in that they separated the regime for the mentally disturbed from that of recidivists and habitual offenders (to which some kinds of sex offenders were added in 1998). The initial confusion introduced by the social defense act is denounced here. There can be no confusing these two categories of individuals, those who suffer from a mental disorder and those who do not, even if both represent a danger to society.20 Regarding the regime of mentally disturbed offenders, the new law dated April 21, 2007 was supposed to replace the 1930-1964 law.

  • 21 Exposé des motifs, Doc. Parl., Chambre, 2006-2007, 51-2841/001, 24.
  • 22 Exposé des motifs, Doc. Parl., Sénat, 2006-2007, 9.
  • 23 Exposé des motifs, Doc. Parl., Sénat, 2006-2007, 47.

49The 2007 law retained most of the earlier philosophy of the Social Defense Act, while introducing a number of modifications. For a person to enter the protection of society system, the law no longer demanded insanity or “a state of mental imbalance”. It mentioned the existence of a “mental disorder”, essentially because that concept is “more in step with present psychiatric notions”.21 In practice, this imprecision seemed liable to extend the range of application of the law (Mary et al., 2011, 9). This law conditions internment on the existence of a “danger” that further offenses connected with the mental disorder might be committed (art. 8). Clearly, then, the security-related dimension of the law was retained, the main concern being to protect society against the risk of recidivism. Similarly, while psychiatric expertise was prescribed to establish the existence of a mental disorder, it also had to assess the risk of recidivism linked with that mental disturbance (art. 5). Such expertise, with no second assessment demanded (Vandermeersch, 2008, 119), is made compulsory prior to any internment. The person in charge of evaluation was always a psychiatrist, but other types of psychological, criminological or social expertise might have been called upon.22 The law, validating actual practice, rubber-stamped competition between the psychiatrist’s clinical knowledge and the psychologist’s technical knowledge, termed “equipped knowledge” since it required “proficiency in specific evaluation tools”, without actually taking sides.23

  • 24 Exposé des motifs, Doc. Parl., Chambre, 2006-2007, 51-2841/001, 12-13.

50Internment remained the main protection of society measure. In this respect, the 2007 law specified that existing facilities should be ranked according to their degree of security (top, medium, and minimum security).24 In addition, internment measures were henceforth be monitored by the court of sentence implementation, which replaced the Board of Social Defense (CDS). Juridicization of the procedure was an underlying trend in a law that took its distance from the administrative, welfare logic. From now on, then, it was a court agency that oriented and monitored the internee’s path until final release. This juridicization resulted in the absence of a specialist in psychiatry and a lawyer within the decision-making agency. While the purpose here was to avoid a confusion of roles and to restore the psychiatrist’s expert function and the lawyer’s counseling function (van de Kerchove, 2010, 497), the present members of the CDSs actually had some fears about this trend and view the elimination of psychiatrists as an indication of de-medicalization and a probable re-criminalization of the regime (Cartuyvels et al., 2010, 272).

51On the whole, the security-oriented nature of the law was criticized (Casselman, 2011, 241-243). A form of re-criminalization seemed to be written into the texts, inasmuch as the modes of enforcement of the law are modeled after the regime of sentenced offenders whenever possible. At the enforcement stage, there was little difference between a measure and a sentence of punishment. Four facts corroborated this: 1) in both cases the same penal agency, the Court of Sentence Enforcement, was competent; 2) whereas release on probation (for a two-year period) was now compulsory prior to final release, from now on it can only be visualized following a period of internment. This eliminated the possibility of releasing a person immediately. Moreover, internees, like ordinary prisoners, can have only be given probationary release following other measures (leaves, vacations, limited detention, electronic monitoring) (art. 25). The law made official the “slowly but surely” practice that we have observed bringing the internee’s path to release closer to that of prisoners. The law also called for at least two years of post-release monitoring for internees, whereas the previous law did not prescribe any specific period; 3) as in criminal offenses, the victim was given the right to be informed or heard regarding the granting of any mode of implementation of internment (art. 4); 4) the law introduced the same system of negative selection as in the regime of sentenced offenders, with respect to the internee’s request for release, which cannot be refused unless the Court of Sentence Enforcement notes some counter-indications (Mary et al., 2011, 10). Last, final release can be granted only if the internee’s mental state has improved to the point that there is no reason to believe he is dangerous. Here again, the law ratified the previous practice of CDSs. Dangerousness, which is to say the risk of recidivism, remained the main factor conditioning both entering and leaving internment.

2) The 2014 Act on the internment of mentally ill persons: change in continuity

52Adopted unanimously, the Act of 5 May 2014 on the internment of persons is intended and supposed to replace the Act of April 21, 2007, even before the latter has really come into force. Transitional arrangements have been made in this regard up to the entry into force of the Act of 2014 (planned on July 1st, 2016), while the current regime of 1930 and 1964 legislation is still in force.

  • 25 See Colette-Basecqz, 2015.

53Broadly speaking, this new legislation, which like the 2007 Act could be postponed due to a lack of resources, is characterized by the overall philosophy of the 1930-1964 Social Defense Act as a system oscillating between security and care. In this respect, the 2014 Act adopts several innovations introduced by the 2007 law while trying to address some of the criticisms expressed against the 2007 law25.

54The new law also adopts the new terminology of "mental disorder", a notion deemed to be more in adequacy with the evolution of contemporary psychiatry (art. 9). The law makes psychiatric assessment compulsory prior to any decision on internment and sets the minimum content. The law also sets up a panel of experts (already regularly used in practice), and provides the assistance with other experts in behavioral science (also commonly used) while making the expertise potentially contradictory (art. 7). In this regard, non-legal observers have suggested for a long time that the expertise must be potentially contradictory (with respect to the same critiques addressed to the use of technical evaluation models) and now the law is accepting this claim by codifying legal norms allowing for dissensus what some have claimed for a long time (van de Kerchove, 1983). In response to the condemnations pronounced by the European Court of Human Rights (Cliquennois, Cartuyvels, Champetier, 2014), internment remains the basic measure of the regime but can no longer, in principle, be served in the psychiatric wing of ordinary prisons. Internment should be only served in an institution for the protection of society, in a social defense section, in a forensic psychiatric center for internees at "high risk", or in an institution recognized by the competent authority organized by a private institution, a Community or a Region for internees at "low or moderate risk" (Art. 3). The management and control over internment is attributed to the Social Protection Chamber of Sentence of the Sentence Implementation Court, made up of a judge (the sentence implementation judge) who will sit as president, an assessor specialized in social rehabilitation and a specialist assessor in clinical psychology (art. 93, 2). This new competence contributes to the juridicization of the internment (kept in the criminal orbit) introduced by the 2007 Act under the pressure exerted by the Council of Europe (Cliquennois, Cartuyvels, Champetier, 2014), while reintroducing at the same time in this organ a clinician actor that the 2007 law had removed, raising criticism among medical professions

  • 26 National Medical Association, Opinion of 11 December 2010 on the law of 21 April 2007 related to in (...)

55Finally, while the 2007 law had already introduced various ways of execution of internment measures (largely inspired by those for persons convicted by the law of May 17, 2006 on the external status of prisoners), the Act of May 5, 2014 maintains these various modalities but considerably softens them. Specifically, more leeway is given to the Social Protection Chamber to select a particular modality (transfer, furlough, leave, electronic surveillance, conditional release...) in order to design a flexible “care pathˮ which, in contradiction to a sentence implementation model based on successive steps, remains consistent ) with the state and evolution of the internee. The rigidity of the system established by the 2007 Act which prevented from granting release without first going through the required stages of a closed internment, furloughs, leaves…was heavily criticized, especially by the National Medical Association, which estimated that “the rigidity of the legal system was in contradiction with the flexibility required by therapeutic actionˮ26. The new legislator has clearly taken into account this critique Furthermore, we can note that the final release of the internee could only occur after a period of release test of two years renewable (art. 66) and as far as the mental disorder has sufficiently improved and is no longer dangerous (art. 66).


56The 1930 social defense law is definitely a first expression of individualization of punishment, the goal of which is to protect society against mentally disordered offenders viewed as dangerous. This law evidences the priorities of the welfare compromise (Garland, 2006) and relies on both control and care logics. Medicalization comes to the forefront with the growing power of psychiatry and the goal of treating the mentally disturbed so as to return them to society. Concern with control remains prevalent, however, and justifies taking some liberties with the rules of classical criminal law. Internment, a measure taken for an indefinite period, as well as preventive detention, are obstacles to the principles of legality and proportionality of punishment, whereas the vagueness of concepts such as “severe state of mental imbalance” or “mental disorder” challenge the idea that criminalization must have a legal basis.

57Clearly, then, criminal law based on dangerousness was constructed to combat the dangerously insane, as a complement to, and on the fringes of, classical criminal law. The former was made possible by concern with controlling the group of abnormal individuals who escape penal repression and feed statistics on recidivism. This occurred with the development, over time, of a specific linguistic mystification (van de Kerchove, 1977). Exceptions to the principles of criminal law were made possible and acceptable through alleged official de-criminalization involving treatment of the mentally disordered. Officially, the protection of society measure is not punishment but a measure providing care and ensuring security.

58The fiction was very rapidly criticized in those terms. Actual practices merely corroborated the analysis, as soon as it became evident that concern with care took the back seat to the ideal of reducing security risks. Maintenance of a majority of internees in psychiatric wings of prisons, the absence of adequate care in these custodial establishments, Belgium’s repeated condemnation by international agencies (the CPT and ECHR), all illustrate the difficulty in escaping the security focus. The 2007 law, which in many respects re-criminalizes the internment regime by modeling the internees’ regime on that of sentenced offenders, merely reinforced this trend. In this regard it was clearly a law of continuity (Vandemeulebroeke, 2008, 361) and one can wonder whether the 2014 Act - assuming that it will come into force shortly - will fundamentally change the situation. Changes in the balance between security and care would imply a significant shift in culture within Social Protection Chambers, a larger access to the intermediate structures for internees and probably increased resources to enable these actors responsible for intermediate housing to set up and apply appropriate care paths. Will these resources be increased whereas the leitmotif of the new Minister of Justice is “efficiencyˮ and the desire to do better with less?

59 Over and beyond these institutional changes, we can conclude that internees are from the beginning those in the legal system without a voice and through cracks. Oscillating between a status of patient and offender, their management still seems to be dominated by a managerial, risk-reduction logic that threats their reintegration into society. As if for many of them, had settled the idea that they were idle individuals condemned to dump and wander the halls of internment, even out for reentry. This managerial, risk-reduction logic translates into an increase in the internee population, the adoption of long-term internment, the fear of taking risks concerning release and the gradual introduction of actuarial tools for expertise purposes.

Haut de page


ADAMS K., FERRANDINO J., 2008, Managing mentally ill inmates in prisons, Criminal Justice & Behavior, 8, 913-927.

ANCEL M., 1981, La défense sociale nouvelle, Paris, Cujas.

von BAR L., BELL T.S., WIGMORE J.H., 1916, An history of continental criminal law, 1836-1913, Boston, Little brown and Co.

BENEZECH M., PHAM T.-H., LE BIHAN P., 2009, Les nouvelles dispositions concernant les criminels malades mentaux dans la loi du 25 février 2008 relative à la rétention de sûreté et à la déclaration d’irresponsabilité pénale pour cause de trouble mental : une nécessaire évaluation du risque criminel, Annales médico-psychologique, 167, 39-50.

BERNARDI A., 2010, Renaissance et mutations de la défense sociale en Italie, Déviance et Société, 4, 547-570.

BODEUX M., 1910, L’augmentation de la criminalité et ses causes, Revue de Droit Pénal et de Criminologie, 1, 628-647, et 749-788.

BOTTOMS A., 1997, Reflections on the Renaissance of Dangerousness, Howard Journal of Penology and Crime Prevention, 2, 70-96.

CARTUYVELS Y., 2009, Troubles de conduite et déviance : des amours en eaux troubles, Chronique de criminologie, Revue de Droit pénal et de Criminologie, 4, 32-61.

CARTUYVELS Y., 2012, La sortie du circuit de défense sociale : évaluer le risque et encadrer la dangerosité, in LANCELEVÉE C., CLIQUENNOIS G., DUGUÉ F., BESSIN M., CARTUYVELS Y., Ce que la dangerosité fait aux pratiques. Entre soin et peine, une comparaison France-Belgique, Paris, Mission Droit et Justice, 124-174.

CARTUYVELS Y., 2013, De Terbeschikkingstelling van de Regering (TBR) van recidivisten, gewoontemisdadigers en “seksuele delinquenten” : van “veiligheidsmaatregel” naar dubbele straf, Orde van de Dag, 61, 46-55.

CARTUYVELS Y., CHAMPETIER B., WYVEKENS A., Soigner ou punir ? Un regard critique sur la défense sociale en Belgique, Bruxelles, FUSL, 2010.

CARTUYVELS Y., CHAMPETIER B., 2011, L’expert psychiatre et le juge face à l’expertise en défense social. Entre collaboration et rapports de pouvoir, in TULKENS F., CARTUYVELS Y., GUILLAIN C. (dir.), La peine dans tous ses états. Hommage à Michel van de Kerchove, Bruxelles, Larcier, 275-293.

CASSELMAN P., 2011, Internering in Vlaanderen anno 2010 : ondanks het goede nieuws nog steeds mensonwaardig, in AERTSEN I., DAEMS T., HUAREN A., HUTSEBAUT F., MAESSCHAELK J. (red.), Deviante wetenschap. Het domeine van de criminologie. Liber amicorum Johan Goethaels, Leuven-Den Haag, Acco, 237-247.

CASTEL R., 1983, De la dangerosité au risque, Actes de la recherche en sciences sociales, 47-48, 119-127.

CHEVALIER L., 1978, Classes laborieuses et classes dangereuses à Paris, pendant la première moitié du XIXe siècle, Paris, Hachette.

CLIQUENNOIS G., 2012, Défense sociale : entrer, circuler, sortir, in LANCELEVÉE C., CLIQUENNOIS G., DUGUÉ F., BESSIN M., CARTUYVELS Y., Ce que la dangerosité fait aux pratiques. Entre soin et peine, une comparaison France-Belgique, Paris, Mission Droit et Justice, 94-123.

CLIQUENNOIS G., CARTUYVELS Y., CHAMPETIER B., 2014, Le contrôle judiciaire européen de la prison : les droits de l’homme au fondement d’un panoptisme inversé ?, Déviance et Société, 38, 4, 491-519.

COCOZZA J., STEADMAN H., 1976, The failure of psychatric predictions of dangerousness: clear and convincing evidence, Rutgers Law Review, 29, 1084-1101.

COLETTE-BASECQZ N., 2015, La loi du 5 mai 2014 : un meilleur cadre légal pour l’internement des personnes atteintes d’un trouble mental, in BEERNAERT M.A. (dir.), Actualités de droit pénal, Bruxelles, Anthémis, 153- 212.

CONRAD P., SCHEIDER J.W., 1980, Deviance and medicalisation: from badness to sickness, Saint Louis, Mosby.

CORNIL P., (1930), La loi de défense sociale à l’égard des anormaux et des délinquants d’habitude du 9 avril 1930. Mercuriale du 15 septembre 1930, Cour d’Appel de Bruxelles (s. ed.)

CORNIL L., 1946, Propos sur le droit criminel. Discours prononcé par M. Léon Cornil, Procureur Général à l’Audience solennelle de rentrée de la Cour de Cassation du 16 septembre 1946, Bruylant, Bruxelles.

COSIJNS P. (président), 2005, Werkgroep Forensisch Psychiatrisch Zorgcircuit, Syntheseverslag mai 2005, Brussel, Ministère de la Justice.

COSIJNS P., D’HONDT C., JANSSEN D., MAES E., VERELLEN R., 2007, Geïnterneerden in België. De cijfers, Panopticon, 1, 46-61.

COSIJNS P., D’HONDT C., JANSSEN D., MAES E., VERELLEN R., 2008, Les internés en Belgique : les chiffres, Revue de Droit Pénal et de Criminologie, 4, 364-380.

COSIJNS P., KOZCK S., VERELLEN R., 2008, De justiciabele met een psychiatrische stoornis in Vlaanderen, Tijdschrift voor psychiatrie, 50, 63-68.

DA AGRA C., 1988, Dangerosité et dégénérescence. la médecine mentale en Belgique à la fin du XIXe siècle et au début du XXe siècle, in TULKENS F. (dir.), Généalogie de la défense sociale en Belgique (1880-1914), Bruxelles, ed. Story-Scientia, 91-111.

DE CLERCK M., 2006, Uitdagingen in de forensische psychiatrie vanuit een medium-riskbehandelingsperspectief, De Orde van de Dag, 34, 29-32.

de CONINCK Fr., CARTUYVELS Y. et al., 2005, Aux frontières de la Justice, aux marges du social, Gent, Academia Press.

De LANTSHEERE L., 1910, Discours à l’Assemblée générale de l’Union Internationale de droit pénal, Bruxelles, les 2, 3 et 4 août 1910, Revue de droit pénal et de criminologie, 911-920.

DE VUYSERE S., 2005, Veertig jaar wet ter bescherming van de maatschappij : reden tot feesten ?, in GOETHALS J., HUTSEBAUT F., VERVAEKE G. (red.), Gerechtelijke geestelijke gezondheidszorg : wetenschap, beleid en praktijk. Liber amicorum Joris Casselman, Leuven, Universitaire Pers, 259-269.

DEBUYST C., 1988, La théorie de la défense sociale et la mise en place de sa logique au XIXe siècle. Note sur Ducpétiaux et le problème des taudis, in TULKENS F. (dir.), Généalogie de la défense sociale, 1880-1914, Bruxelles, Story-Scientia, 231-241.

DEBUYST C., DIGNEFFE F., PIRES A.P., 2008, Histoire des savoirs sur le crime et la peine. T. 3. Expliquer et comprendre la délinquance (1920-1960), Bruxelles, Larcier.

DESSECKER A., 2009, Dangerousness, long prison terms, and preventive measures in Germany ; Champ pénal/Penal Field, Séminaire du GERN "Longues peines et peines indéfinies. Punir la dangerosité" (2008-2009), mis en ligne le 24 octobre 2009.

DIGNEFFE F., 1995, Problèmes sociaux et représentations du crime et du criminel. De Howard (1777) à Engels (1845), in DEBUYST C., DIGNEFFE F., LABADIE J.-M., PIRES A.P., Histoire des savoirs sur le crime & la peine, vol.1., Bruxelles, Montréal, Ottawa, de Boeck Université, Les presses de l’Université de Montréal, Les Presses de l’Université d’Ottawa, 139-208.

DONNEDIEU de VABRES H., 1947, Traité de droit criminel et de législation comparée, Paris, Sirey.

DONNEDIEU de VABRES H., 1951, Le professeur E. Garçon (1851-1922) et l’Union internationale de droit pénal, Revue internationale de droit pénal. Bulletin de l’Association Internationale de Droit Pénal, 2-3, 191-208.

DOURNEAU-JOSETTE P., TULKENS F., 2010, La défense sociale au regard de la Convention européenne des droits de l’homme, Déviance et Société, 4, 691-706.

DUPONT L., 1988, Jules Lejeune et la défense sociale, in TULKENS F. (dir.), Généalogie de la défense sociale, Bruxelles, Story-Scientia, 77-86.

FEELEY M., SIMON J., 1992, The new penology: notes on the emerging strategy of corrections and its implications, Criminology, 30, 449-474.

FEELEY M., SIMON J., 1994, Actuarial Justice : the Emerging New Criminal Law, in NELKEN D. (ed.), The Futures of Criminology, London, Thousand Oaks, New Dlehi, Sage, 173-201.

FOUCAULT M., 1975, Surveiller et punir. Naissance de la prison, Paris, Gallimard.

FOUCAULT M., 1978, About the Concept of the « Dangerous Individual » in the 19th-Century Legal Psychiatry, International Journal of Law an Psychiatry, 1, 1-18.

FOUCAULT M., 1981, L’évolution de la notion d’ « individu dangereux » dans la psychiatrie légale, Déviance et Société, 4, 403-422.

FOUCAULT M., 1999, Les anormaux. Cours au collège de France, 1974-1975, Paris, Gallimard.

GARAPON A., 2008, Un nouveau modèle de justice : efficacité, acteur stratégique, sécurité, Esprit, 349, 98-122.

GARÇON E., 1909, Rapport à la session d’Amsterdam, Bulletin de l’Union Internationale de droit pénal, Revue de droit pénal et de criminologie, 393-400.

GARLAN D., 1981, The Birth of the Welfare Sanction, British Journal of Law & Society, 1, 29-45.

GOORDEN J.-C., OEI T.I., 2007, Tbs en internering. Actuele ontwikkelingen in Belgïe, Proces, 1, 17-25.

HANNAH MOFFAT K. (2006), Pandora’s Box: Risk-need and gender responsive corrections, Criminology and Public Policy, 1, 1301-1311.

HARCOURT B., 2011, Surveiller et punir à l’âge actuariel. Généalogie et critique, Déviance et Société, 1, 5-33.

HAUS J.J., 1874, Les principes généraux du droit pénal belge, T.2, Gand-Paris, Hoste-Thorain.

HEDAYATI M. A., 1939, Les mesures de sûreté et la réforme moderne du droit pénal, Genève, Impr. du Journal de Genève.

HERREMANS J., Un détenu belge demande l’euthanasie. Quelle réponse lui donner ?, Justice en lignes, 30 novembre 2015 [].

INSERM, 2005, Les troubles de conduite chez l’enfant et l’adolescent, Paris, Éd. de l’Inserm.

JESCHECK H.H., 1979, L’influence de l’Union internationale de droit pénal et de l’Association internationale de droit pénal sur l’évolution de la politique criminelle, Actes du XIIe congrès de l’Association internationale de droit pénal, Hambourg, 34-46.

JUNG A., 2010, De Carl Stoob à l’internement à vie. La dérive sécuritaire des mesures en Suisse, Déviance et Société, 4, 571-596.

KALUSZYNSKI M., 2008, Le retour de l’homme dangereux. Réflexions sur la notion de dangerosité et ses usages, Champ Pénal/Penal Field, vol. V.

KHOLRAUSCH E., 1924, Sicherungshaft. Ein Besinnug auf den Streitstand, Zeutschrift für die gesamte Strafrechtwissenschaft, 44, 21-34.

LECLERC P., 1930, Note sous Cassation 12 mai 1930, Pasicrisie Belge Cassation, 212-218.

LIGA VOOR MENSENRECHTEN, 2012, Geinterneerd : cel of Zorg ?, Antwerpen, EPO.

LOMBROSO C., 1871, L’Uomo bianco e l’uomo di colore, Padua, Sachetto.

LOPEZ-REY M., 1981, Quelques réflexions sur le danger de la notion de dangerosité, in DEBUYST C., TULKENS F. (dir.), Dangerosité et justice pénale. Ambiguïté d’une pratique, Genève, Médecine et Hygiène, 327-373.

MARY Ph., 2013, Enjeux contemporains de la prison, Bruxelles, FUSL.

MARY Ph., KAMINSKI D., MAES E., VAN HAMME F., 2011, The Treatment of « dangerousness » in Belgium: internment and placing at the government’s disposal, Champ Pénal/Penal field, vol. VIII [http:/].

MARTENS F., 2004, Abus sexuels : enjeux cliniques et sociétaux. Colloque du 22 octobre, « Délinquants sexuels : comment protéger la société ?, Groupe MR, Maison des Parlementaires.

MATTHIJS J., 1965, La loi de défense sociale à l’égard des anormaux. Évolution des conceptions, Journal des Tribunaux, 165-172.

MAURUTTO P., HANNAH-MOFFAT K., 2006, Assembling risk and the restructuring of penal control, British Journal of Criminology, 46, 438-454.

MAUS I., 1907, La récidive, ses causes et ses remèdes, Revue de Droit Pénal et de Criminologie, 689-719.

MELOSSI D., SOZZO M., SPARKS R. (eds), 2011, Travels of the Criminal Question: Cultural Embeddedness and Diffusion, London, Hart Publishing.

MEYER C. (dir.), 2005, Le livre noir de la psychanalyse. Vivre, penser et aller mieux sans Freud, Paris, les arènes.

MICHAELSEN C. (ed.), (forthcoming) Comparative Perspectives on Theory and Practice of Preventive Detention, London, Hart Publishing.

MONAHAN J., 1981, The Clinical Prediction of Violent Behavior, Washington DE, Government printing office.

OST F., van de KERCHOVE M., 2002, De la pyramide au réseau ? Pour une théorie dialectique du droit, Bruxelles, FUSL.

O’MALLEY P., 2000, Risk Societies and the Government of crime, in BROWN M., PRATT J. (eds), Dangerous offenders. Punishment and social order, Routledge, 17-39.

PASSELECQ F., 1913, De la position pratique du problème de la ‘responsabilité atténuée’ dans le droit pénal belge, Revue de droit pénal et de criminologie, 353-371.

PHAM T.-H., SALOPPE X., BONGAERTS X., HOEBANX J., 2007, L’expertise dans le cadre de la loi de Défense Sociale en Belgique : repères diagnostiques et recommandations, Annales Médico Psychologiques, 165, 49-55.

PHAM H.T., DUCRO C., MARTIN M., PIHET B. (2010), Projet d’évaluation en continu des caractéristiques délictueuses, des aspects diagnostiques, de l’environnement social et de la prise en charge des auteurs d’infractions à caractère sexuel (AICS) au sein d’équipes de santé spécialisées en Région wallonne, Annales Médico-Psychologiques, 168, 6, 458-461.

PIN X., 2010, l’Internement de sûreté en Allemagne : une mesure de défense sociale à la dérive, Déviance et Société, 4, 527-546.

PRATT J., 1997, Governing the dangerous: dangerousness, law and social change, Sydney, Federation Press.

PRATT J., 2000, Dangerousness and Modern Society, in BROWN M., PRATT J. (eds), Dangerous offenders, Punishment and Social Order, Routledge, 35-48.

PRINS A., 1886, Criminalité et répression. Essai de science pénale, Bruxelles-Leipzig, Librairie européenne C. Muquardt, Merzbach et Falk.

PRINS A., 1905-1906, Le péril moral et social de la récidive d’après les dernières données statistiques, Revue de l’Université de Bruxelles, 545-566.

PRINS A., 1910, La défense sociale et les transformations du droit pénal, Bruxelles, Misch et Thron.

PRINS A., 1912, L’esprit nouveau dans le droit criminel étranger, Revue de droit pénal et de criminologie, 133-151.

PRINS A., 1899, Science pénale et droit positif, Bruxelles Paros, Bruylant-Christophe et librairie Marescq aîné.

RABINOWICZ L., 1928, Les mesures de sûreté, Paris, Les écrivains réunis.

RAFTER N., 1997, Creating Born Criminals, University of Illinois Press.

RAYNOR P., 2010, Usages et abus du risque dans la justice pénale britannique, Déviance et Société, 4, 671-690.

ROBERT Ph., 2002, L’évolution des politiques de sécurité, in MUCCHIELLI L., ROBERT Ph. (dir.), Crime et sécurité. L’État des savoirs, Paris, La Découverte, 42-52.

ROLIN A., 1907, La lutte contre le crime, Revue de droit pénal et de criminologie, 209-228.

ROSE N. (1998), Governing Risky Individuals: The role of Psychiatry in New Regimes of Control, Psychiatry, Psychology and Law, 2, 177-195.

RUTTIENS R., 1911, L’École positive et ses adversaires, Revue de droit pénal et de criminologie, 486-489.

SALEILLES R., 2001, [1898, reprint], L’individualisation de la peine, in OTTENHOF R. (dir.), L’individualisation de la peine. De Saleilles à aujourd’hui, Toulouse, Érès.

SIMONS R., 1908, Le crime et la défense sociale, Revue de droit pénal et de criminologie, 541-560.

STEADMAN H., MORISSEY J.P., 1981, The statistical prediction of violent behavior: Measuring the costs of a public protectionist versus a civil libertarian model, Law and Human Behavior, 5, 263-274.

STRAUSS A. L., 1992, La trame de la négociation : sociologie qualitative et interactionnisme, Paris, l’Harmattan.

SUTHERLAND J.F., 1908, Recidivism regarded from the Environmental and Psych-Pathological Standpoints, The British Journal of Psychiatry, 54, 68-93.

THIRY F., 1910, Les êtres dangereux par leur état, Revue de Droit Pénal et de Criminologie, 5-15.

TULKENS F., 1988, Un chapitre de l’histoire des réformateurs. Adolphe Prins et la défense sociale, in TULKENS F. (dir.), Généalogie de la défense sociale en Belgique (1880-1914), Bruxelles, ed. Story-Scientia, 19-46.

TULKENS F., van de KERCHOVE M., CARTUYVELS Y., GUILLAIN C., 2010, Introduction au droit pénal. Aspects juridiques et criminologiques, 9e édition, Bruxelles, Kluwer.

VAN CAMPENHOUDT L., CHAUMONT J.M., FRANSSEN A., 2005, La méthode d’analyse en groupe. Applications aux phénomènes sociaux, Paris, Dunod.

van de KERCHOVE M., 1976-1977, Des mesures répressives aux mesures de sûreté et de protection. Réflexions sur le pouvoir mystificateur du langage, Revue de droit pénal et de criminologie, 245-279.

van de KERCHOVE M., 1981a, Culpabilité et dangerosité : réflexions sur la clôture des théories relatives à la criminalité, in DEBUYST C., TULKENS F. (dir.), Dangerosité et justice pénale. Ambiguïté d’une pratique, Paris-Genève, Masson-Médecine et Hygiène, 291-309.

van de KERCHOVE M., 1981b, « Médicalisation » et « fiscalisation » du droit pénal. Deux versions asymétriques de la dépénalisation, Déviance et Société, 1, 1-25.

van de KERCHOVE M., 1983, Le juge et le psychiatre. Évolution de leurs pouvoirs respectifs, in GERARD Ph., OST F., van de KERCHOVE M., Fonction de juger et pouvoir judiciaire. Transformations et déplacements, Bruxelles, FUSL, 311-390.

van de KERCHOVE M., 1988, L’organisation d’asiles spéciaux pour aliénés criminels et aliénés dangereux. Aux sources de la loi de défense sociale, in TULKENS F. (dir.), Généalogie de la défense sociale en Belgique (1880-1914), Bruxelles, ed. Story-Scientia, 113-140.

van de KERCHOVE M., 2010, Les avatars de la loi de défense sociale belge : le changement dans la continuité, Déviance et Société, 4, 485-502.

VANDEMEULEBROECKE O., 2008, Un autre régime d’internement des délinquants atteints d’un trouble mental. La loi du 21 avril 2007, Revue de Droit Pénal et de Criminologie, 4, 308-363.

VANDEMEULEBROEKE O., 2008, Un autre régime d’internement des délinquants atteints d’un trouble mental, Journal des Tribunaux, 2, 308-363.

VANDENBROUCKE M., 2009, De wereld van de internering en van de geïnterneerden : het functioneren van de commissies te bescherming van de maatschappij, Panopticon, 6, 92-97.

VANDERMEERSCH D., 2008, La loi du 21 avril 2007 relative à l’internement des personnes atteintes d’un trouble mental, Journal des Tribunaux, 6299, 117-125.

VANDEVELDE S., VEERLE S., VANDER BEKEN T., DE SMET S., BOERS A., BROECKAERT E. (2011), Mentally Ill Offenders in Prison : the Belgian Case, International Journal of Law and Psychiatry, 1, 71-78.

VAN OUTRIVE L., CARTUYVELS Y., PONSAERS P., 1991, Les polices en Belgique. Histoire socio-politique du système policier de 1794 à nos jours, Bruxelles, Vie Ouvrière.

VAN VLAENDEREN S., BEYENS K., 2010, De definitieve invrijheidstelling van geïnterneerden... a never ending story? Fatik, 128, 9-20.

VERVAECK L., 1926, La stérilisation des anormaux et des criminels dangereux, Revue de droit pénal et de criminologie, 5, 449-450.

WYVEKENS A., 2010, La rétention de sûreté en France : une défense sociale en trompe-l’œil, Déviance et Société, 4, 503-526.

Haut de page


1 Bulletin de l’Union Internationale de droit pénal, 1889, 4.

2 Prins, 1886, 1899, 1910.

3 The source of the 1930 Social Defense Act is found in a bill dated April 15, 1880 regarding the organization of special asylums for the confinement of the insane following sentencing or if they are dangerous.

4 The deaf and dumb were also mentioned (art. 76 CP), but did not represent a political problem.

5 M. Janson, Minister of Justice, Annales de la Chambre, session 1927-1928, 81.

6 Projet de loi, Exposé des motifs, Doc. Parl., Chambre, session 1922-1923, n° 151.

7 For the situation in Flanders, see Cosijns et al., 2008; Vandenbroucke, 2009.

8 Figures provided by the General Directorate of Correctional Facilities, Powerpoint presented by the Health Coordinator of the Brussels internment network at the Congrès National de Défense Sociale, Brussels, September 20-21, 2013.

9 Cartuyvels et al., 2010.

10 Cartuyvels, 2012 ; Cliquennois, 2012.

11 ECHR, 30 July 1998, Aerts vs. Belgium.

12 Administrative instruction n° 1800 dated June 7, 2007.

13 See for instance Civ. Charleroi (ref.), 25 February 2005, Journal des Tribunaux, 2005, 308, note L. Brackman, « De la décision d’internement à son exécution » ; Civ. Liège (ref.), 17 September 2007, 1409 ; Civ. Namur, 18 March 2011, J. L.M.B., 2013, 450.

14 Cass. 26 March, R. G. n°C.09.0330.F; C.C., 17 September 2009, n°142/2009.

15 ECHR, 10 January 2013, Claes v. Belgium ; ECHR, 6 December 2011, De Donder and De Clippel vs. Belgium ; ECHR, 10 January 2013, Duffort vs. Belgium ; ECHR, 10 April 2013, Sweenen v. Belgium ; ECHR, 9 January 2014, Saadouni vs. Belgium ; ECHR, 9 January 2014, Gelaude vs. C. Belgium ; ECHR, 9 January 2014, Lankaster vs. Belgium ; ECHR, 9 January 2014, Van Meroye vs. Belgium, ; ECHR, 9 January 2014, Plaisier vs. Belgium ; ECHR, 9 January 2014, Oukili v. Belgium ; ECHR, 9 January 2014, Moreels vs. Belgium ; ECHR, 9 January 2014, Caryn vs. Belgium.

16 ECHR, 9 January 2014, Saadouni vs. Belgium, §56 and 61. On this jurisprudence, see Cliquennois et al., 2014, 508-514.

17 ECHR, 3 February 2015, Smits vs. Belgium, §74.

18 The situation may be different in Flanders, where Anglo-Saxon approaches are more prevalent.

19 In France, the violent debate elicited by the Inserm’s collective investigation on “The screening, treatment and prevention of behavioral disorders in children and adolescentsˮ (Cartuyvels, 2008) is a good illustration of this fear felt by some mental health actors.

20 Exposé des motifs, Doc. Parl., Sénat, 2006-2007, 3-2054/1, 4.

21 Exposé des motifs, Doc. Parl., Chambre, 2006-2007, 51-2841/001, 24.

22 Exposé des motifs, Doc. Parl., Sénat, 2006-2007, 9.

23 Exposé des motifs, Doc. Parl., Sénat, 2006-2007, 47.

24 Exposé des motifs, Doc. Parl., Chambre, 2006-2007, 51-2841/001, 12-13.

25 See Colette-Basecqz, 2015.

26 National Medical Association, Opinion of 11 December 2010 on the law of 21 April 2007 related to internment of mentally ill persons, 83.

Haut de page

Pour citer cet article

Référence électronique

Yves Cartuyvels et Gaëtan Cliquennois, « The Punishment of Mentally Ill Offenders in Belgium: Care as Legitimacy for Control. », Champ pénal/Penal field [En ligne], Vol. XII | 2015, mis en ligne le 12 février 2016, consulté le 21 janvier 2017. URL : ; DOI : 10.4000/champpenal.9307

Haut de page


Yves Cartuyvels

University of Saint-Louis. Contact :

Gaëtan Cliquennois

CNRS, SAGE, University of Strasbourg, law department of the European University Institute. Contact :

Haut de page

Droits d’auteur

© Champ pénal

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