Navigation – Plan du site

Detention, Alternatives to Detention

Typologies of Measures and Penal Sanctions
Pierre V. Tournier
Cet article est une traduction de :
Détention, alternatives à la détention


The member-states of the Council of Europe share the objectives of reducing the role of prison in their penal systems, resorting more to measures and penalties other than imprisonment.

It is therefore important to better understand current developments as well as the situations of the various states in this matter. This means having at your disposal certain concepts and methods likely to be used in various national contexts with their differences in laws, penal systems and statistical apparatus.

This is the task undertaken by the author, relying on long experience as an expert in the Council of Europe.

Haut de page

Texte intégral


1At a time when the European Union is attempting, albeit at a careful pace, to create a judicial space, it seems important that researchers develop new tools able to facilitate international comparisons1. The contribution that we are proposing to readers of Champ Penal -Penal Field involves the creation of penal systems related to penalties. We are clearly trying to go beyond purely legal considerations-which punishment (s) incurred for which offense (s)-in attempting, in order to quantify this production, to define typologies by taking into account not only the sentencing but also the application of sentences. Only they are able to show the complexity of things, and, for example, measure the significance of « deprivation of freedoms » (to be later defined) in relation to those that we label as « alternatives » to detention (to be later defined).

2In this task we rely on the experience acquired with numerous foreign colleagues, from diverse disciplines, encountered in the context of the work of the Council of Europe2 - the Europe of 46 states whose contributions remain fundamental for all those who want to advance Europe towards this « zone of liberty, justice and security ». We are not naive enough to think that these typologies, as complex as they are, can, in their current forms, be applicable to the 46 states of the whole of Europe, any more than they are to the 25 states of the European Union. More modestly, we hope that in an heuristic effort, colleagues from different countries might want to attempt the exercise on their own data, and to discuss any difficulties they might encounter.

1. The Rate of Detention

3When one attempts to compare the « production » of penal systems of two given countries, to understand, for example, whether one punishes penal infractions more severely than the other, the simplest method, if not the most effective, is to go to the prisons and count their prisoners. The presence of high walls naturally leads one to think that this census must not be very complicated, no one concerned, on the face of things, being able to escape. It is still necessary to relativize this number, based on the demographic importance of the countries concerned, without mentioning several problems of definition.

4The detention rate is an indicator of the current number (or stock) obtained through reporting, at a given date, the number of prisoners in relation to the total number of people in the country in question. It is, in fact, a ratio (a part compared to the whole).

5This rate must not be confused with the rate of admission into detention that we have labelled rate of incarcerations, which originated in 1983 as an Annual Penal Statistic of the Council of Europe (SPACE).

6An indicator of flux, this second rate is arrived at by counting the annual entries into detention of people previously free, and relating them to the average number of inhabitants during that period. It represents the frequency of entries into detention in relation to the whole of the population.

7The Council of Europe's statistics for the 1st of September, 2004, showed a rate of 91 per 100,000 in France (Council of Europe, 2004). This rate is quite close to that of Switzerland (82), of Germany, or Italy (97). We find, traditionally, the lowest rates in northern Europe: 40 in Iceland, 65 in Norway, 66 in Finland, 70 in Denmark, 76 in Northern Ireland, (but 82 in Sweden). On the other hand, many European countries now experience rates that are far higher than 100 prisoners per 100,000: 121 in Luxembourg, 124 in the Netherlands, 136 in Scotland, 140 in England, Wales or in Spain, and that is without considering the former communist countries of central or eastern Europe where the rates are generally far higher.

8This calculation of the detention rates, in spite of the apparent simplicity of the index, is not without several methodological problems. We will briefly describe a few of these problems. The rate in France appearing in the SPACE database on September 1, 2004, was obtained by dividing 56,271 by 62,177,000 inhabitants. On this date, the number of « jailed persons », that is to say, judicially detained3 – is in fact 56,971 (domestic and overseas). In excluding the 472 sentenced to electronic monitoring and the 228 sentenced but not confined, we obtain the 56,271 on the SPACE database. Referring to the total number of jailed persons, we would have arrived at the rate of 92 per 100,000. We could have also decided to not count prisoners having permission to leave (temporary absence)-they are legally but not actually physically detained. Thus, the adage that « jail makes the prisoner » is a great oversimplification. We will return to this subject.

9The choice of denominator in the calculation of the rate-the population of France-also presents a problem. Indeed, the youngest inhabitants-who are all included in the denominator-have zero probability of being prisoners, for legal reasons and, for the oldest members of the population, the likelihood is extremely weak. Inversely, some categories of people can actually be recognized as prisoners without being recognized by the census as being part of the French population: foreigners in irregular situations, foreigners in passage (seasonal workers, tourists).

10But more fundamental to our problem-the place of prison in the systems of punishment-these rates of detention clearly do not inform us with regard to alternatives to prison. To do this, as per tradition, we refer to the sentencing statistics issued by the national office of criminal records.

2. The Classic Approach: An Appraisal Using Multiple Perspectives

2.1. An Alternative for Two  Sentences

11In France, penal infractions are classified, according to their gravity, as crimes, délits4, and contraventions (article 111-1 of the French penal code)-this final category being subdivided into five classes. For a crime, the maximum incurrable term of reclusion or criminal detention-is life; for a délit, the maximium term is ten years. The sentences for contraventions are fines (to a maximum of 1500 euros for contraventions of the 5th class, 3000 euros for recidivism in certain cases) and sentences involving deprivation or restriction of rights, as defined in article 131.14 of the penal code. Since the coming into effect of the new penal code on March 1, 1994, sentences requiring deprivation of liberty cannot be incurred for contraventions, even of the 5th class.

12The sentences registered at the office of criminal records are for crimes, délits and contraventions of the 5th class. For 20035, they amounted to 525,053.

13Let us define a first dichotomy (A) in this array:

14A1. The « sentences of deprivation of liberty, broadly defined », punishments that, in one way or another, refer to « prison »:

151.Imprisonment for a minimum of 10 years,

162.Strict Confinement Without Reprieve of Execution,

173.Strict Confinement Combined With a Partial Reprieve of Execution (mixed sentence),

184.Confinement Combined With a Complete Reprieve, Simple (that is, without a specific condition),

195.Confinement Combined With a Complete Reprieve, With Conditions not Accompanied by Community Service (SME),

206.Detention Combined With a Complete Reprieve, With Conditions, Accompanied by Community Service (SME-TIG).

21A2. The other sentences, described as « alternative sentences-narrowly defined » :


232.Suspension of Driving Permit,


254.Day-Fines (to be paid within a certain time frame),

265.Community Service as the Main Sentence,

276.Withdrawal of Access to Driving Permit,


29Complete reprieve with or without conditions does indeed refer to prison, since a quantum is effectively specified (six months, one year of imprisonment...) but not executed (except for a later revocation of the reprieve). It is the difference that exists between « French probation » and « English probation » (without reference to prison).

30On the basis of this first dichotomy, we find that for the year 2003, there were 283,202 sentences of imprisonment in the wider sense, this representing 54% of the total number of sentences registered at the office of criminal records. The percentage of alternative sentences is, therefore, 46% of the three categories of penal infractions considered. This amounts to just under one alternative sentence for every two sentences passed.

2.2. One Alternative for Three Sentences ( délits )

31The 525,053 sentences registered in 2003 are divided as follows: 3,174 sentences for crimes (0.6 %), 411,373 sentences for délits (78.3 %) and 110,506 sentences for contraventions of the 5th class (21 %).

32We have counted 3,145 sentences of imprisonment (in its broad sense) among the 3174 sentences for crimes (99%), 280,057 among the 411,373 for delits (68%)-these sentences not being available in law for contraventions of the 5th class.

33If, as before, we describe an alternative sentence as a punishment that is not an imprisonment in its broad sense, the proportion consisting of alternative sentences becomes 1% of the sentences for crimes, 32% of the sentences for délits and, obviously, 100% of the contraventions of the 5th class. Thus, in this formulation, one third of the sentences for délits are alternatives.

34Among the alternative sentences for délits, first comes the fine (52%), next, suspension of the driving permit (13%), re-education measures (12%), day fines (9%), community service as principal sentence, (7%), withdrawal of access to driving permit (2%).

2. 3. Three Alternatives to Four Sentences (délits)

35We can design the dichotomy of sentences (B) differently:

36B1. The « sentence of imprisonment in its narrow sense », a punishment that necessarily includes a strict element of imprisonment:

371.Imprisonment for a 10 year minimum,

382.Strict Confinement without Reprieve,

393.Strict Confinement Coupled with a Partial Reprieve (Mixed Sentence),

40B2. Other Sentences Described as « Alternative Punishments in a Broad Sense »,

411.Imprisonment Coupled with a Complete Reprieve, Simple (without a specific condition,

422.Imprisonment Coupled with a Complete Reprieve, Without Public Service (SME),

433.Imprisonment Coupled with a Complete Reprieve, With Conditions, Coupled With Community Service (SME-TIG),


455.Suspension of Driving Permit,


477.Day-Fines (to be paid within a certain time frame),

488.Community Service as Principal Sentence,

499.Withdrawal of Access to Driving Permit,


51Thus, when we exclude, from imprisonment in its wider sense, imprisonment with complete reprieve, things appear quite different. Indeed, they represent 2/3 of the sentences for délits, in the earlier sense.

52In now labelling as an alternative punishment (or sentence) that which is not a strict form of detention or accompanied by a partial reprieve, we then find that the percentage of alternatives for délits is 75%. Coming first is the basic reprieve (39% of the alternatives), then the fine (22%), reprieve with conditions (15%), suspension of the driving permit (5%), re-education (5%), day-fines (3.7%), community service as main sentence (TIG) (2.9%), reprieve coupled with community service (2.8%), suppression of driving permit (.07%).

2. 4. Punishment and the Qualification of Infractions

53Let us return to the rate of imprisonment (in either the wide or narrow sense). No matter what definition one uses, the global rate (T) is the weighted average of rates for crimes, délits, and contraventions of the 5th class, the coefficients used being the weights of each of the categories of infractions related to the total of sentences registered at the office of criminal records.

54Here, T1 is the rate of imprisonment for crimes, T2 the rate for délits and T3 the rate for contraventions of the 5th class.

55Also, p1 the proportion of crimes, p2 the proportion of delits and p3 the proportion of contraventions of the 5th class.

56We have T = p1 .T1 + p2 . T2 + p3 T3.

57If we consider as a first estimate that p1 = 0 % (in reality : 0.6 %), as we have T3 = 0 %, the preceding equation is simplified and we get :

58T = p2. T2, with p2 as the proportion of délits and T2 the rate of PPL for the délits.

59The global rate of sentences of imprisonment is approximately equal to the product of the rate among délits and of the proportion of délits among the totality of sentences.

60With the definition that includes full reprieve, we obtain:

61 T = p2 . T2 = 78.4 % x 68 = 53.3 %

62With the definition that excludes full reprieve, we obtain

63 T* = p2 . T2* = 78.4 % x 25 = 19.6 %,

64We clearly see here the importance of the qualification of facts.The same fact can, in reality, qualify as a contravention of the 5th class or as a délit, in terms of assault and battery, for example. The qualifier, « contravention of the 5th class », has automatically resulted, since 1994, in some alternative to detention. On the other hand, the qualifier « délit », leaves the door open to...prison.

65Thus, the fact that infractions a priori qualified as contraventions of the 5th class can be qualified as délits makes it possible to increase the global proportion of sentences of imprisonment. However, the consideration or non-consideration of sentences that allow for full reprieve and the distinctions among infractions based on legal categories are not the only problems posed by such calculations.

2.5. Other Domain-Related Questions

66Are we thus not creating a bias in reasoning, as we just did (without saying so) in relation to the totality of the jurisdictions of judgement, by not distinguishing the jurisdictions of the 1st degree from the courts of appeal? In order to avoid this problem, the solution might be to abandon the transversal approach used here (taking into account the sentences passed in any given year) in favour of a longitudinal analysis which consists of following the cases through the different levels of jurisdiction until we encounter the definitive sentence.

67Moreover, should we not take heed of the manner of judgement, in other words, must we only think about the punishments meted out in a contradictory fashion or should we also include the judgements that might have been called into question because of the absence of the condemned (judgement by default)?

68One can see that, in spite of an apparent simplicity, this classic approach to the question of « alternative sentences » is not without some problems of its own. As well, it inconveniently ignores the fact that the penal process, at the point culpability is established and a choice of punishment is required, is part of a complete process, that there is a « before » (pre-sentencing) and an « after » (post sentencing) where the question of alternatives to detention is posed as well. This is what we try to consider infra.

3. European Classification of Penal Measures and Sanctions

69A first approach to enlarging the field of analysis is by referring to the complex concept of penal measures and sanctions (MSP in French) introduced indirectly by the Council of Europe when it was discussed to elaborate rules for the « open milieu » after the adoption, for the « closed milieu », of European prison rules in 1987.

3.1. Sanctions and Measures Applied in the Community

70The concept of « sanctions and measures in the community » (SMC in French) derives from recommendation No.R (92), 16 (Council of Europe, 1994). The SMCs are those that maintain the offender in the community and involve a certain restriction of his freedom by the imposition of conditions and/or obligations, and which are implemented by organizations designated by current laws in force. This concept refers to the sanctions determined by a tribunal or a judge and to the measures taken before a sentencing decision taken in lieu of such a decision; it equally refers to those consisting of a modality of execution of a sentence of imprisonment outside of a prison institution. Thus, the modalities of application of the SMC must anticipate some form of assistance and monitoring by the community.

71While, in French law, a fine or simple reprieve (at the execution of a sentence of imprisonment) without assistance or monitoring, are not SMCs- in the sense of the Council of Europe-the reprieve with conditions is.

72The SMCs also included a socio-educational control (a pre-sentence measure), community service that can also be considered as a sanction if given as the main penalty or as a measure-it is one of the obligations of a reprieve with conditions, placement outside prison, electronic monitoring, conditional release (measures prescribed ab initio or post sentencing).

3.2. An Implicit Typology in Three Categories

73This approach defines, in an implicit fashion, a triptych (C) of sanctions and penal measures:

74C1. -Sanctions and measures executed within prison,

75C2. - Sanctions and measures applied in a community, that is, executed outside of prison, but with supervision,

76C3. - Sanctions and measures executed outside of prison, but without supervision.

77This triptych is compatible, for sanctions, with dichotomy B, defined supra. Indeed, B1 =C1 and B2 becomes in its turn, the object of a dichotomy (B2 = C2 +C3).

78We return to sentences for délits handed down in France during 2003. If we  consider the Council of Europe's concept of sanction applied in the community, which requires that the sanction be non-carceral but also accompanied by a follow-up (supervision), the sentences for délits divide up in the following way:

79C1. 25% of carceral sentences – strict or with partial reprieve,

80C2. 17% of sentences being applied within the community,

81C3. 58% of alternative sentences without follow-up.

82We thus find ourselves establishing a dichotomy among the measures and alternative sanctions according to whether they can be accompanied by « supervision » (17%) or not (58%). When we debate, in France, the issue of alternatives to detention, those who denounce the small number of alternatives generally have Category B in mind (for délits)-namely, « an alternative » to 5 sanctions Let us recall that we are only referring here to findings of guilt. But the essential idea of « MSP » is obviously larger.

4. The Prison Mosaic: A Static Approach

83Here we leave the area of given sanctions, or the abstract classification of sanctions and measures, in order to approach the « field » of the execution of sanctions and measures belonging to the categories C1 and C2 of the European typology, and this to a « t »moment (static approach).

84The concept of the « prison mosaic » (Tournier, 2004) allows for the inclusion, in the same approach, of measures and sanctions, closed and open milieus, but also the interface between the two, and all the while specifying who is who (penal status) and who is where (the concrete conditions of placement under the control of the justice system).

85The first dichotomy that is necessary in this collection of people placed by law under the control of the justice system (PPMJ in French) is the one that allows us to distinguish between people who are waiting for a definite judgement and thus still presumed innocent from those who have been declared guilty by a jurisdiction empowered to pass judgement, who have been subjected to a penal sanction and are in the process of executing this sentence, whatever its nature and the modalities of its execution6. We will refer to them as « defendants » in the broad sense of the term and of the definitively « sentenced » (table 1)7.

Prison Mosaic – Table 1



86The second essential distinction relates to the idea of « custody ». Some PPMJs are in custody, others are not. This second approach must, a priori, be combined with precedent. From this derives Table 2, in four categories, whose pertinence will be clarified later (the cells not being empty).

Prison Mosaic – Table 2.

87« Open Milieu »



88« Closed Milieu »



89The PPMJs of Category 1., « defendants not in custody », are people under judicial control. They can be supervised by the prison-provided services of insertion and of probation (SPIP in French) or by some authorized association. They may have been placed under judicial control ab initio, or during the course of provisional detention. The Category 2 PPMJs, « sentenced but not in custody » are followed by the SPIP8. Those are people found guilty, having the benefit of a reprieve (of execution) with conditions (SME) or a sentence of community work (TIG), but in addition, the people sentenced to a strict term of imprisonment or imprisonment for a ten year minimum with a lifting of the detention before the anticipated end of the sentence, in the context of a conditional release (LC). The SMEs and TIGs do not execute sentences of imprisonment, the LCs continue to execute sentences of imprisonment, but arranged within the « community ».

90This category is still complicated by the application of law no. 98-468 of June 17, 1998, relating to the prevention and the punishment of sexual infractions as well as the protection of minors and the application of law no. 2005-1549 of December 12, 2005 relating to the treatment of penal recidivism.

91Thus, we also find in this category those sentenced who are considered dangerous and who will be made the object of a « socio-judicial supervision » (as the principal sanction or as a measure accompanying a sentence of imprisonment without reprieve) or of a « judicial surveillance » after release, with or without an order to undergo care, with or without satellite electronic monitoring.

92Categories 1 and 2 constitute the « open milieu » (or « probation » in the broad sense of the term) and categories 3 and 4 constitute the « closed milieu » (prison).

93But things become complicated when we understand, as we already noted supra, that some people, while being in custody, are not present at that moment, in a penal institution. This reference-not to the « legal presence » (custody)- but rather to the physical presence, is the third distinction which, obviously relates to only part of the « prison mosaic ». From this derives table 3 in six categories.  

94Inversely, some people can be considered not in custody while yet being in a certain sense « closed in ». It relates, by application of law no. 2002-1138 of September 9, 2002, to minors placed in closed education centres (CEF in French) by the application of a legal constraint or a reprieve with conditions or a conditional release.

Prison Mosaic -Table 3. « Open Milieu »



95Closed Education centres (CEF in French)




96« Closed milieu/in detention/outside the walls »

5. PPMJ  DEFENDANTS IN CUSTODY « outside the walls »



97« Closed milieu/in detention/within the walls »

7. PPMJ DEFENDANTS IN CUSTODY « within the walls »

8. PPMJ SENTENCED IN CUSTODY « within the walls »

98That which we describe as « closed milieu – ouside the walls » is itself a true mosaic: some people that are under detention are legally absent, others not (escapees). We will limit ourselves here to the first category and define three cases9.

99« Permanent and total absence from prison » : sentencees put under electronic monitoring (PSE in French), sentencees placed on release without being housed in a penal institution.

100Total, temporary absence » : sentencees on temporary absence, defendants or those sentenced and hospitalized outside the facility, including in units for difficult patients (UMD in French).

101The table then consists of 10 categories, showing a complex intersection, between the open and closed milieus.

Prison Mosaic – Table 4

102« Open Milieu »

1. PPMJ DEFENDANTS NOT IN CUSTODY legal control by SPSIP or associations...

2. PPMJ SENTENCED NOT IN CUSTODY SME, TIG, conditional release, socio-legal supervision, legal surveillance (with or without PSE mobile electronic monitoring)

103« Closed Education Centres (CEF) »

3. PPMJ DEFENDANTS NOT IN CUSTODY in CEFs (for minors) under legal control


In the CEFs (for minors)

SME, conditional release

104« Closed milieu/in custody/within the walls »

9. PPMJ DEFENDANTS IN CUSTODY  « inside the walls »


« within the walls »

105The population under legal control is thus made up of 9 categories :

106P = [P1 + P2] + [P3 + P4] + [P5 +P6 + P7 + P8] + [P9 + P10]

107If we return to the European classification of penal measures and sanctions, limiting ourselves to category C1 (measures and sanctions executed in a penal institution) and C2 (measures and sanctions applied in the community, that is to say, executed outside of a penal institution but under supervision), we arrive at the following classification :

108C1 = P9 + P10

109C2 = P1 + P2 + P3 + P4 + P5 +P6 + P7 + P8

110We might note, nevertheless, that within this dichotomy, one can be subjected to a sanction or a measure  within the community while having a greatly restricted freedom of movement (people detained in a UMD, minors placed in a CEF, sentencees on day parole).

111We can, therefore, define the degree of deprivation of freedom, in terms of the execution of measures and sanctions by the indicator of stock at the moment t C1 / (C1 + C2).

112At this point, it is not possible to calculate this index in light of the absence of certain data. In a future study made with the cooperation of the management of the prison administration, we will attempt to fill these gaps.

5. Typology of Alternatives to Detention : A Dynamic Approach

113The last approach that we are proposing involves elaborating a typology of the totality of measures and sanctions that are part of category C2 (measures and sanctions that are applied within the community) or of C3 (measures and sanctions executed outside of a penal institution but without supervision), and this based on the analysis of the renewal of the prison population : analysing the stock based on the admissions in detention and the length of those terms.

5.1. P = E x d

114The existence of the population in detention is regulated by the following mechanism: people placed in custody-previously free–enter the prison population; people freed and leaving detention. There is a lapse of time between the entry into and the release from detention, this period of time, different for each individual, assures the coexistence at each moment of the variable number of people that constitute the population in detention. A research instrument for the study of these modes of population renewal, demographic analysis have basic models, providing reference populations, easy to define, that allow us to consider by comparison some concrete situations. The simplest model is that of the stationary population. A population is said to be « stationary » if the annual entries into the population (E) are constant and if the exits of each generation–meaning the individuals entered in the same year–proceed according to the same rhythm, and the same calendar. We can thus demonstrate that the size of the population, at a given time (P) is equal to the product of the annual entries (E) and the average duration of detention in the population (d, expressed in years) : P = E x d (Tournier, 2005). This equation implies that a stationary population has a constant size. The typology which we will present (Tournier, 2003) rests on the foundation of this theoretical formula.

5.2. Three Types of Alternatives

115Considered as an alternative to the first category (D1), any penal measure and sanction (MSP in French) results in a reduction in the number of admissions into detention.. It is thus the case of ab initio judicial control or of a simple reprieve when sentence is pronounced upon a free defendant. These alternatives can be described as radical. By avoiding the admission into detention, these alternatives allow the sentencee to completely avoid detention.

116Naturally, among these alternatives of the 1st category, we find measures and sanctions that belong to category C2 (for example, judicial control) as well as C3 (for example, simple reprieve).

117The alternatives of the 2nd category (D2) allow for reductions in the duration of detention, or more precisely, the time spent in custody. Therefore, as a measure, it is the lesser of two evils, it is partial or relative : the recourse to prison could not be avoided, but we ensure that the time spent in detention is reduced. In this scenario, the credits given for the reductions of sentence for « good behaviour », the pardons, individual or collective, relating to detainees, are alternatives of the 2nd category. It is the same with conditional release, a measure used in the management of sentences that is associated with a release from prison.

118Certainly, this dichotomy does not allow us to classify the totality of MSPs into two distinct categories because many of them belong to both, depending on the conditions of application. Thus, judicial control is of the 1st category if it is declared ab initio. However, if it is decided upon while the person is in provisional detention, the measure is of the 2nd category : it reduces the duration of detention while waiting for the judgement. The conditional release belongs to the 2nd category : not reducing the duration of executed sentence, it allows for an anticipated release-with release from detention–the remainder of the sentence being carried out in an « open milieu ».

119Thus, the issue of the management of sentences seems very much an integral part of the issue of alternatives.

120This being so, the dichotomy shows its limits. What about, indeed, the day parole, the release into the community, measures that are also alternatives to prison, partial or relative, but which do not avoid detention – they are not of the first category -and do not reduce the duration of detention – are they not of the 2nd category?

121Thus, we describe as alternatives of the 3rd category (D3) those measures that reduce the time really spent behind the walls of penal institutions, without a full release from detention, and therefore without a reduction in the time spent in detention. Such is the case with day parole, release to the community as well as temporary absence. It is also the case with fixed electronic monitoring (PSE in French). The source of interest in such measures is obvious: reduction in demographic pressure, reduction in the negative effects of detention, tools for re-insertion, help in the development of conditional release.

5.3. Real or Virtual Alternatives ? The Issue of Net-Widening

122The typology we have just presented has the advantage of simplicity. Above all, it shows that by developing these three types of alternatives, we can suppress carceral inflation, as the Council of Europe was able to reaffirm in its recommendation of September 30, 1999 (Council of Europe 2000) regarding the overpopulation of prisons and carceral inflation. On a more theoretical level, this classification allows us to discuss the well known theory of net-widening in a new light.

123When a person who has not been in provisional detention benefits from judicial control, and is later given a sentence with a complete reprieve, it's possible to think that this individual measure of control actually allowed him to avoid prison. But we can also affirm that the judge would not have used provisional detention, if judicial control was not available in law. The judge used a supplementary guaranty that was offered to him. If this is the case, this judicial control does not play its role as an alternative to prison (it's a virtual alternative) but allows the net of social control to expand. In fact this same issue can more or less be considered for all the alternatives of the 1st category. Would a person sentenced to community work have been sentenced to a strict imprisonment if the TIG had not existed in the law? Would he not instead have benefitted from a simple reprieve or even a fine?

124The questioned is posed in very different terms in relation to the alternatives of the 2nd category. A prisoner with three years of imprisonment remaining who is granted a conditional release benefits from a very real alternative. He will complete his remaining three years in an open milieu, under the supervision of a judge assigned by the SPIP (sentences, prison service, insertion and probation). And yet...

125In France, conditional release has become more and more rare.

126The Jospin government (1997-2002) came to a realization and finally, in the context of the law of June 15, 2000, started making an important reform of the procedures relating to the granting of conditional release. Let us imagine that the desired revival of conditional release will, in several years, be effective -pure hypothesis, since that road has not been taken10. Might it not eventually lead to a compensatory increase in the length of sentences passed by the jurisdictions, frustrated at seeing « their » sanctions too « eroded ». Thus, an alternative of the 2nd category, very real, at the individual level–the beneficiary will have no doubt about this-can become quite virtual at the global level , in the long run.

127For measures of the 3rd category, it's necessary to distinguish those measures taken at the beginning of detention from those not. Let us consider placement under fixed electronic monitoring (PSE in French) corresponding to the execution of a sentence of one year or less11 : in the absence of PSE, in law, would the person have been sentenced to the equivalent of imprisonment, or would he have been given a reprieve with conditions, or even a simple reprieve? Will jurisdictions not, from now on, be encouraged to pass sentences of strict imprisonment of at least one year in cases where formerly, they would have granted a reprieve, knowing that thanks to the bracelet, the sentencee could avoid detention? The reasoning is uncertain since the placement decision is not taken by the jurisdiction of judgement. We can imagine the « catastrophic » following scenario: a correctional tribunal passes a sentence of strict imprisonment of six months, for example, rather than a reprieve with proof, thinking that this will be executed within the context of a placement under electronic monitoring (without deciding on the measure ab initio), but the judge applying the sentence  and who has to manage the former decision-refuses. Far from being an alternative to imprisonment, the PSE encourages its uses as a recourse. Without empirical research on which to base it, this reasoning remains purely theoretical.

128On the other hand, the PSE at the end of the sentence does not pose the same type of question. The case of probationary measures at the conditional release is characteristic. Their existence can encourage the granting of conditional release while increasing the conditions upon which the judge managing the sentence can base his decision. Without these measures, conditional release might have been effective from a date t ; with these measures, the prisoner released from detention only at the date t + t ; a virtual alternative, since it results in an increase in the period of detention.

6. Reconciliation of the Points of View

129In the typology A, it is only about the sentences-measures not being pertinent here-passed in relation to crimes, délits and contraventions of the 5th class –the flow of sanctions imposed in a given year.

130A = A1 + A2, A1 representing all the sanctions that refer in one way or another to prison (imprisonment with complete reprieve included), A2 being made up of the other sanctions (without reference to prison).

131On this basis, we derive a first equation of the weight of the « alternatives » :

132a = A2/ (A1 + A2).

133In typology B, the basis is identical to typology A but we classify differently the sentences of imprisonment with full reprieve.

134B = B1 + B2, with B1 representing the sentences of imprisonment including a set minimum amount, with B2 consisting of all the other sanctions (which include full reprieve).

135We thus obtain a new equation of the weight of the « alternatives »: b = B2/ (B1 + B2).

136In typologyy C, we have enlarged the area to the totality of the measures and penal sanctions, without otherwise specifying the nature of the infractions concerned.

137C = C1 +C2 + C3, with C1 representing the carceral penal measures and sanctions (MSP in French), C2 representing the MSPs executed in the community (with the possibility of supervision) and C3 representing the other MSPs (non carceral but without supervision).

138This nomenclature is not compatible with typology A, but is with typology B. For sanctions, we have B2 = C2 + C3.

139Two equations might be worth considering:

140c1 = (C2 + C3) / (C1 +C2+C3) and c2 = C2 / (C2 + C3), the second equation showing the weight of alternatives with supervision.

141In the « PPMJ » typology, the approach is completelydifferent. We attempt to describe, at a given moment, the populations executing the measures and penal sanctions that belong to categories C1 and C2 of the preceding typology. Taking into account the distinction between defendants and sentencees, between people in detention and those not, while considering the concrete conditions of placement, we arrive at a classification consisting of ten compatible groups; we have seen this supra, with typology C:

142C1 = P9 + P10 et C2 = P1 + P2 + P3 + P4 + P5 +P6 + P7 + P8 .

143Finally, in typology D, we take into account the measures and penal sanctions as in typology C, but limiting ourselves to categories C2 and C3, which we will distinguish according to their effects upon the population in custody :

144D = D1 + D2 + D2, with D1 MSP, which facilitate the avoidance of detention, D2 MSP, which reduces the duration of detention and D3 MSP which reduces the time spent behind the walls.  Here again, the classification is compatible with typology C.

Table of Concordances












P1, P3, P5


P2, P4, P6, P7, P8

D1, D2,D3

D1, D2,D3

D1, D2,D3

D1, D2, D3

145To conclude provisionally, we would like to invite our foreign colleagues to survey their own systems of measures and penal sanctions through the different proposed grids, in describing what they can find–in law–in the totality of cells and sub-cells and in... measuring the weights of different categories.

Haut de page


Conseil de l’Europe, 1994, Règles européennes sur les sanctions et mesures appliquées dans la communauté, Recommandation n°R (92) 16 et exposé des motifs, Référenc-es juridiques, 1994.

Conseil de l’Europe, 2000, Le surpeuplement des prisons et l’inflation carcérale, recommandation N°R (99) 22, adoptée par le Comité des ministres du Conseil de l’Europe le 30 septembre 1999 et rapport, coll. Références juridiques.

Conseil de l’Europe 2004, Statistique pénale annuelle du Conseil de l’Europe (SPACE I),

Lavielle (B.), Lameyre (X.), 2002, Le guide des peines. Prononcé. Application, Paris, Dalloz, collection des guides.

Tournier (P.V.), 2003, Real Alternatives versus Virtual Alternatives : On the Theory of Net-Widening Applied to Electronic Monitoring in France, in Mayer M., Haverkamp R. Lévy R. (Eds.) Will Electronic Monitoring Have a Future in Europe? Contributions from a European Workshop, June 2002, Kriminologische Forschungsberichte aus dem Max-Planck-Institut für ausländisches und internationales Strafrecht, Freiburg i. Br, 177-186.

Tournier (P.V.), 2004, Mosaïque pénitentiaire : une topologie mouvante, Actualité juridique. Pénal, Dalloz, 9, 333-334.

Tournier (P.V.), 2005, L’équation des prisons in Lehning (H.), dir., Les équations algébriques. Aborder les inconnues, bibliothèque Tangente, 22, 150-152, hors série.

Haut de page


1 This article was translated by Uri Ben-Gal.
2 Creation and development of the annual penal statistical manual of the Council of Europe (SPACE), a project of the Sourcebook based on the initiative of Martin Killias (University of Lausanne), elaboration of the recommendation on the overpopulation of prisons and carceral inflation (1999) and of that relating to conditional release (2003).

3 Detention is a judicial act, established by officials of penal institutions, declaring that a person has been remanded to the warden and describing his civil status, the date of admission into detention, the nature and title of his detention, the infractions prosecuted or sanctioned, etc.

4 Translator's Note: since there is no precise equivalent in English-speaking legal systems, the translator has retained the original term « délit ». The common-law « misdemeanor » is not equivalent.
5 The most recent data available.
6 We take into account here the situation of each at a given moment. A provisional detention followed by a sentence of strict imprisonment for a longer period will a posteriori be a time of execution. But this time will have been spent as a "defendant". It is the same for a person whose sentencing is not definitive, and yet impemented ( it is enough that the sentence is "enforceable").
7 The category of "imprisoned debtors", imprisonment by default, is often considered as being part of the latter.
8 Or are on the verge of (waiting for summons).
9 The terminology that follows is not ideal. It can only be understood through the definition "in extension" given infra.
10 Right-wing governments, in power since 2002, have not done anything to support the return of conditional release and this in spite of the unanimous adoption on September 24, 2003, of the Council of Europe's recommendation to proceed in this direction.
11 Whether the decision is taken by the jurisdiction of judgement (in applying the Perben 2 law), or by the sentencing judge.
Haut de page

Pour citer cet article

Référence électronique

Pierre V. Tournier, « Detention, Alternatives to Detention », Champ pénal/Penal field [En ligne], Vol. III | 2006, mis en ligne le 19 juin 2009, consulté le 26 mars 2017. URL : ; DOI : 10.4000/champpenal.7188

Haut de page


Pierre V. Tournier

Directeur de recherches au CNRS, Centre d’histoire sociale du XXe siècle (Université Paris 1. Panthéon Sorbonne) 43, rue Guy Môquet, F-75017 PARIS).

Articles du même auteur

Haut de page

Droits d’auteur

© Champ pénal

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