Behind Bars in Brazil
III. Overcrowding, Alternative
Severe overcrowding is perhaps the most basic, and most chronic, problem plaguing Brazil's penal system. Over a decade ago, national prison authorities estimated that the country's prisons needed an additional 50,934 spaces to accommodate the existing inmate population.58 Since then, despite some recent efforts to handle the problem, the disparity between available capacity and actual inmate numbers has only worsened. By 1997, with the growth in inmate numbers, the deficit in prison capacity was officially estimated at 96,010.59 In other words, for every space that exists in the prisons, there are 2.3 prisoners to lay claim to it.
True prison capacity is difficult to estimate objectively and, as a result, capacity figures are notoriously easy to manipulate,60 but there is no doubt that almost all of Brazil's penal facilities are woefully overpopulated. As every prison administrator is well aware, crowded prisons are dangerous ones: heightened inmate tensions lead to prisoner-on-prisoner violence, escape attempts, and attacks on guards. Unsurprisingly, a substantial proportion of the incidents of rioting,hunger striking and other forms of protest occurring in the country's penal facilities is directly attributable to overcrowding. In many instances,
particularly in the state of São Paulo, inmates have rioted simply to demand that they be transferred to a less crowded facility, typically wanting to leave a cramped police lockup for a more spacious prison.
If numbers from recent years are any indication, Brazil's inmate population will continue to increase and, most probably, to outstrip the expansion in prison capacity. The deficit in available capacity grew 27 percent between 1995 and 1997, as the total capacity of the prisons increased by only 8.1 percent during that period.61
The shortage of prison space is particularly dramatic when examined in light of the enormous number of criminal defendants who have evaded serving their prison sentences, leaving these sentences pending. The federal Ministry of Justice estimated in 1994 that there were 275,000 such unserved sentences (mandados incumpridos), significantly more than the number of prisoners in confinement.62 In Brasília alone, the public prosecutor's office announced this year that of the 15,077 prison sentences handed down in his jurisdiction over the past three years, only one-third of them have actually been served; defendants in the remaining cases are fugitives.63 Obviously, were these missing convicts suddenly to be found and confined, the prisons would burst.
The real number of fugitives from prison is difficult to estimate, however, as state and federal figures include multiple sentences for a single defendant, defendants who have died, and cases in which the statute of limitations has expired. One prisons expert advises that, at minimum, "the existing numbers should be divided by five" in order to take these factors in account.64 Even so, the number of additional inmates these sentences represent could place a significant burden on an already overwhelmed penal system.
Detention Before Trial
An important contributor to overcrowding in Brazil's prisons is the confinement of unsentenced prisoners, who constitute roughly one-third of the inmate population.65 Because such persons have not been convicted of any crime, they are presumed innocent under the law, and some proportion of them will indeed be acquitted of the crime for which they are held despite having already served time in confinement.
Under international human rights standards, defendants should normally be released pending trial. Articulating this principle, Article 9(3) of the International Covenant on Civil and Political Rights (ICCPR) provides in relevant part that: "It shall not be the general rule that persons awaiting trial shall be detained in custody, but release may be subject to guarantees to appear for trial."66 In interpreting this provision, the U.N. Human Rights Committee has ruled that detention before trial should be used only to the extent it is lawful, reasonable, and necessary. Necessity is defined narrowly: "to prevent flight, interference with evidence or the recurrence of crime" or "where the person concerned constitutes a clear and serious threat to society which cannot be contained in any other manner."67 The weighing of the relevant criteria for a finding of necessity requires an individualized determination.
Pretrial detention in Brazil is not always ordered in conformity with these exacting standards; indeed, many criminal defendants do not even have the right to request provisional liberty. The Law of Heinous Crimes, passed in 1990, barsjudges from granting pretrial release to defendants charged with any of a number of crimes, including homicide, rape, and armed robbery.68
Compounding the effects of excessive pretrial detention are lengthy criminal proceedings, during which the defendant remains incarcerated. The average length of trial proceedings seems to vary greatly from state to state in Brazil, and we were unable to obtain national statistics on this topic, but our visits to the prisons convinced us that some unsentenced prisoners are confined for years. The problem seems to be particularly severe in the north and northeast of the country. At the Campina Grande prison in the state of Paraíba, we met one unsentenced prisoner who had been incarcerated for three years and nine months. In Natal, Rio Grande do Sul, we spoke to another who had been preventively detained for two years and four months without a decision in his case. In Manaus, Amazonas, a number of prisoners reportedly languished for several years in pretrial detention until a prison rebellion drew attention to the problem. Even without concrete numbers on the average length of trial proceedings it is clear, as one member of the National Council on Criminal and Penitentiary Policy explained, that the substantial numbers of unsentenced prisoners in Brazil are "proof of the slowness of justice."69
Besides favoring pretrial release, international human right norms specifically bar long periods of pretrial detention.70 Trial proceedings lasting several years have been found to be excessive by the U.N. Human Rights Committee and other authoritative bodies.71
Failure of Sentence Progression
A fundamental concept undergirding numerous provisions of Brazil's national prison law is that of the individualization and progression of the prisoner's term of confinement. What this means in concrete terms is that, first, the sentencing judge should consider the defendant's individual circumstances in imposing the sentence. Thus, for example, the question of whether a prisoner is a recidivist or a first-time offender is relevant in deciding whether he is sentenced to a closed prison, to an open facility, or to community service. Second, the judge of penal execution should continually monitor the prisoner's case while he is imprisoned, adjusting the terms of sentence according to the prisoner's conduct. Normally, a prisoner who begins his sentence in a closed prison should, after he has served part of his sentence, be transferred to a semi-open facility and from there, after further time has passed, to an open facility, and finally to release into society.72 In sum, this view of incarceration sees it as a dynamic process, not simply a fixed term of years.
The prison law's prescriptions regarding sentence progression have not been translated into practice, however. Most prisoners never see an open- or semi-open facility; instead they serve their entire sentence in a closed prison, or even a police lockup. Human Rights Watch interviewed many such prisoners, who although eligible for transfer to less restrictive forms of incarceration remained in prison. In May 1998, Brazil's Ministry of Justice estimated that 11.2 percent of convicted prisoners in Brazil-or more than 11,000 prisoners-were eligible for sentence progression, though few were benefiting from it.73
The failure of sentence progression has several causes, including prisoners' lack of legal assistance, the shortage of judges to process their cases, and the small number of open- and semi-open spaces in the penal system. But keeping prisoners who are eligible for sentence progression confined in closed facilities not only contributes to overcrowding, it is also makes such prisoners frustrated andangry, resulting in frequent rebellions.74 Such prisoners have "literally been forgotten by the judicial system," as the legislative inquiry of the São Paulo prison system pointed out;75 prisoners' strong sense of the unfairness of this chronic inattention is obvious to any visitor.
Remedying the Overcrowding Crisis
There are only two basic ways to handle prison overcrowding: through new construction or through the release of excess inmates. Both of these strategies are to some extent currently in use in Brazil. Neither of them, however, has been sufficient to handle the extreme levels of overcrowding plaguing the country's penal system.
Reducing the inmate population
By reducing the inmate population the government not only remedies overcrowding, it lessens prison costs. With this in mind, public authorities-particularly in a developing country like Brazil-must consider carefully whether public monies are most effectively spent on prisons or on other methods of controlling crime.
At the intellectual and theoretical level, the notion of employing alternative sanctions to prison-and of confining people only as a last resort-is well established in Brazil. Numerous academic articles have been written lauding the use of alternatives to prison such as fines, community service, limitations on rights, and suspended sentences; and high government officials have endorsed the concept.76 Proponents of alternative sanctions cite the negative impact thatincarceration has on the inmate, the overwhelming failure of the rehabilitative ideal, the high costs of operating prisons, and the crisis of overcrowding in the penal system to support the argument that society must look for new methods of dealing with criminality. They point to studies indicating lower rates of recidivism among offenders sentenced to alternative sanctions than those given traditional terms of incarceration.77 According to this view, prisons should be reserved for the most violent offenders, those persons who represent a clear danger to the community.
The national prison law embraces these ideas to an extent, as demonstrated by its array of provisions relating to punishments other than prison.78 Still, the use of alternative sanctions remains relatively rare in practice; certainly, such sanctions have not yet had much of an impact in reducing prison overcrowding. According to the 1995 prison census, only some 2,098 offenders were serving alternative penalties, the majority (64 percent) of them having been assigned community service.79 In the state of São Paulo, as of January 1998, only 410 persons were serving such sentences.80
Several factors combine to limit reliance on such sanctions. To begin with, they can only be used with prisoners sentenced to less than one year of incarceration, and thus are not available with regard to a wide variety of crimes. In addition, few localities have set up the organizational structures needed to implement sanctions such as community service, lacking supervisory bodies to ensure that such penalties are indeed carried out. Judges, too, are said to be generally resistent to the idea of allowing offenders to serve non-custodial terms; they tend to regard such sanctions as overly lenient.81
Two states are said to be national leaders in the use of alternative sanctions: Mato Grosso do Sul and Rio Grande do Sul.82 Because of the paucity of reliable statistics, it is not known how frequently such sanctions are employed there. In São Paulo, as well, the government recently established a community service program to serve as a structure for the greater use of alternatives to imprisonment.
The unending series of violent incidents in the prisons in 1997, with their extensive media coverage, helped pave the way for greater acceptance of alternatives to incarceration. At present, a law is pending in the National Assembly to allow judges to substitute alternative sanctions for prison sentences of up to four years, rather than one.83 Minister of Justice Iris Rezende reportedly claimed that the proposed measure could result in the release of some 30,000 inmates.84 Supporters of such efforts are optimistic that the use of alternative sanctions will continue to grow.
Sentence reduction and early release programs offer additional options to reduce overcrowding. Under the terms of the national prison law, prisoners may reduce their sentences by one day for every three days that they work (with each working day being between six and eight hours long).85 Prison authorities are supposed to maintain a log that records which prisoners are working, a copy of which should be submitted to the appropriate judge of penal execution on a monthly basis. Unfortunately, because work opportunities are scant in many prisons-and hardly exist at all in police lockups and jails-inmates are often unable to take advantage of any of the benefits of this provision.
A federal commission reviewing the national prison law recently proposed to broaden the terms of the sentence reduction rules to cover prisoners who studywhile they are incarcerated.86 The revised rules would permit reductions for everything from basic literacy courses to higher education. It is thought that extending the law in this way will encourage inmates to study and improve themselves, while shortening the prison terms of those prisoners who are least likely to reoffend.
Finally, the national prison law also provides for early release or parole (livramento condicional) of inmates who can show that they satisfy a number of requirements, including having already served some minimum portion of the sentence imposed on them (at least one-third to one-half of their sentence, depending on the inmates' prior record), and having "demonstrated satisfactory conduct during the term of the sentence," a more subjective measure.87 The latter requirement is meant to discourage inmates from misbehaving while incarcerated; however, because it requires a qualitative not a quantitative analysis, it can greatly impede the speedy processing of early release applications.
Indeed, inmates' most frequent complaint about both sentence reduction and early release programs is the slowness with which applications for these benefits are processed. At the São Paulo Women's Penitentiary, for example, where the large majority of inmates work, we found women who were eligible for release but were still waiting a year or more for their applications to be heard, a problem that the prison director confirmed.88 Due to the lack of legal assistance in the prisons and the insufficient numbers of judges of penal execution, many prisoners who qualify for early release never obtain it. In recognition of this problem, a bill has been introduced in the National Congress to establish a summary procedure that would facilite the granting of such benefits.89
Finally, any discussion of early release is not complete without a mention of the Brazilian phenomenon of late release. Lacking legal assistance, some prisoners not only fail to benefit from early release programs, they even remain incarcerated beyond the term of their sentences. At the thirty-fifth police precinct in São Paulo, for example, the commander told us that his lockup held several people in this situation when he arrived.90 The head of the Depatri police facility said that one prisoner there had served four months more than his one-year sentence, but that this man had to be held until judicial authorization for his release was received.91 Until a 1997 prison rebellion forced authorities to take steps to resolve the problem, the central prison in Manaus, Amazonas, reportedly held a few dozen prisoners whose sentences had already expired.92
Under the Brazilian constitution and articles 188-93 of the prison law, the president may pardon any crime and release the convicted criminal from prison.93 In the interests of reducing prison overcrowding, the president has on occasion granted collective pardons and sentence commutations to large numbers of inmates. Such pardons are most commonly granted shortly before Christmas, allowing released prisoners to spend the Christmas holidays back with their families. In 1995, for example, 1,748 prisoners received Christmas pardons.94 The following year, in April 1996, the president issued the largest pardon in the country's history, under which some 15-18,000 inmates reportedly became eligible for early release.95 Nearly 3,000 more prisoners, including a group of paraplegic inmates, were pardoned in March 1998 in commemoration of the fiftieth anniversary of the Universal Declaration of Human Rights.96
Expanding prison capacity
As of late, the primary method whereby the Brazilian government proposes to remedy the prisons crisis is by building new prisons and expanding existing ones. After many years of relative inattention to the need to expand the prison infrastructure,97 the federal government began in 1994 to focus on prison construction and was prodded into more decisive action in 1997 by a rash of prison rebellions that drew media attention and shone a spotlight on the horrendously overcrowded conditions of Brazil's penal facilities. Over the course of the year, high federal officials including the president made a series of prominent announcements of new programs of prison construction.98
In all, the government claims that 105 new prisons will open by the end of 1999, fifty-two of which will be paid for by the federal government. The state receiving the largest single contribution of federal prison funds is São Paulo.99 As of late October 1998, São Paulo had inaugurated eleven new prisons whose total capacity was 9,420 inmates. Ten additional prisons, with a total capacity of 8,100 inmates, were scheduled to open by the end of the year.100 One of the stated goals of the construction effort is to replace the decaying and notorious Casa de Detenção, transferring the thousands of inmates held there to new facilities. Yetthe twenty-one new prisons-while allowing prison authorities to reduce the number of inmates held at the Casa de Detenção-are insufficient to make up the existing deficit in prison capacity. Moreover, if past trends are any indicator, the continuing growth of the state's inmate population may offset these gains.
An extensive prison building program is a costly enterprise, and Brazil's current construction plans have been estimated at $440 million reais (approximately US $391 million).101 Whether this amount of money will indeed be allocated remains to be seen. The example of Rio Grande do Sul, Brazil's southernmost state, is a cautionary one: in 1995, the state government promised to build five new prisons by late 1996; not only were none built, but the overall prison capacity actually shrank during this period because of lack of maintenance.102 Moreover, even if this new crop of prisons is built, it seems far less certain that the government is ready to invest the money necessary to operate the facilities in a humane fashion, with all of the associated staff, food, medical and other expenses.
60 Although some countries have established national standards on the topic, there are no universally accepted objective standards setting out the amount of space necessary per inmate. Existing international standards, such the Standard Minimum Rules, simply mandate that prisoners be provided sufficient space to meet the requirements of health and human dignity. Complicating any effort to calculate minimum space requirements is the fact that whether a given amount of living space is adequate will vary according to a number of factors, including the amount of time prisoners spend outside of their cells, the cells' air circulation, etc. Cultural norms regarding privacy may also be relevant. Without strict numerical guidelines, however, prison capacity estimates are extremely malleable. See, for example, "Ohio `Eases' Prison Overcrowding," Prison Legal News, Vol. 7, No. 11 (November 1996) (describing how the prison system of the U.S. state of Ohio changed its rules on minimum space allowances, thereby inflating its rated capacity and "easing" overcrowding). Indeed, Human Rights Watch inspected a few facilities in Brazil whose stated capacities, as provided by the wardens, were clearly exaggerated. The national prison law does mandate that individual cells be at least six square meters (approximately sixty-five square feet) in size. In violation of the prison law, however, most Brazilian prisoners live in group dormitories; even those who live in so-called individual cells nearly always share them with one or more cellmates. 61 "Perfil dos presos . . . " Folha de S. Paulo, March 20, 1998.
65 According to the latest census, nearly 40 percent of prisoners have not received a definitive sentence, but this figure includes some prisoners who were convicted at the trial court level and are appealing the conviction. "Presos sem condenação somam quase 40%," Folha de S. Paulo, March 20, 1998. Again, the proportions vary greatly from state to state. In Amazonas, a northern state where the justice system is particularly overwhelmed, more than two-thirds of the inmates confined at the principal men's prison in late 1996 were unsentenced. "Situação da penitenciária masculina," Em Tempo (Manaus), November 10, 1996.
66 See also General Comment No. 8 of the Human Rights Committee on the International Covenant on Civil and Political Rights, Art. 9 (Sixth Sess. 1982), U.N. Doc. A/40/40 (stating "[p]re-trial detention should be an exception and as short as possible").
67 Hugo van Alphen v. the Netherlands (No. 305/1988) (July 23, 1990), Official Records of the General Assembly, Forty-fifth Session, Supplement No.40 (A/45/40), vol. II., annex IX, sect. M., para. 5.8.
71 In Dermit v. Uruguay, 71 I.L.R. 354 (1982), the Human Rights Committee found a delay of two years between arrest and judgment to violate the relevant provisions of the ICCPR. In Fillastre v. Bolivia (No. 336/1988) (November 6, 1991), U.N. Doc. CCPR/C/43/D/336/1988 (1991), the committee found violations because trial proceedings lasting over four years had not yet resulted in a verdict. See also Giménez v. Argentina (No. 11.245) (March 1, 1996) (OEA/Ser.L/V/II.91) (Inter-American Commission finds a violation of right to trial within a reasonable time where the defendant was detained more than five years pending a judgment).
72 Lei de Execução Penal, art. 112 (stating that sentences of confinement "will be executed in a progressive form, including transfer to a less restrictive regime, to be determined by the judge, when the prisoner has completed at least one-sixth of his sentence in the previous regime and his behavior warrants such progression") (translation by Human Rights Watch).
74 As the vice-president of the National Council on Criminal and Penitentiary Policy asserted, "Delay in granting [sentence progression] benefits is one of the principal motives of prison rebellions, together with overcrowding." Ibid.
76 See, for example, Julita Lemgruber, "A Necessidade da Aplicação e Ampliação das Alternativas à Pena Privativa de Liberdade," Revista do Conselho Nacional de Política Criminal e Penitenciária, vol. 1, no. 5, January/June 1995; Lemgruber, "Os Riscos do Uso Indiscriminado da Pena Privativa de Liberdade," Revista do Conselho Nacional de Política Criminal e Penitenciária, vol. 1, no. 7, January/June 1996; Nelson A. Jobim, "Penas Alternativas: Pontos para Reflexão," Revista do Conselho Nacional de Política Criminal e Penitenciária, vol. 1, no. 7, January/June 1996; Heitor Piedade Júnior, "Reflexões sobre o Fracasso da Pena de Prisão," Revista do Conselho Nacional de Política Criminal e Penitenciária, vol. 1, no. 6, July/December 1995; Hans-Dieter Schwind, Ivette SeniseFerreira, and João Benedicto de Azevedo Marques, Penas Alternativas (São Paulo: Fundação Konrad-Adenauer Stiftung, 1996).
82 Human Rights Watch interview, Julita Lemgruber, Rio de Janeiro, December 30, 1997. The 1995 prison census, however, ranks the southern state of Santa Catarina first in the use of such penalties, with Minas Gerais second. 1995 Prison Census, table XXIV, p. 50.
87 Lei de Execução Penal, art. 131; Penal Code, art. 83 (listing requirements). Foreign prisoners do not qualify for early release because they cannot serve out the remainder of their Brazilian sentence (which still runs even after release) in a foreign country. We heard numerous complaints from foreign prisoners about this rule, which can keep them in prison twice as long as their Brazilian counterparts who commit the same crime.
95 Decreto no. 1.860, de 11 de abril de 1996, que concede indulto especial e dá outras providências; "Indultan en Brasil a entre 15 mil y 18 mil reos del órden común," La Jornada (Mexico City), April 13, 1996.
97 In 1994, for example, the federal government allocated only four million reais (approximately U.S. $3.6 million) toward prison construction, while by 1997 the corresponding amount had risen to 110 million (approximately U.S. $97.9 million). "País deve ganhar 44 novos presídios," Folha de S. Paulo, May 25, 1997.
98 See Daniela Falção, "Ministério promete 10 presídios para SP," Folha de S. Paulo, February 21, 1997; "País deve ganhar 44 novos presídios," Folha de S. Paulo, May 25, 1997; "Brazil to build new prisons to ease overcrowding," CNN World News, June 3, 1997.
99 In two other states Human Rights Watch visited, Amazonas and Rio Grande do Norte, federal funding was being put toward new prison construction, although the new facility in Amazonas will not be large enough to remedy overcrowding at the state's existing men's prison.
100 Fax from Cláudio Tucci, Adjuct Secretary of Prison Administration of the State of São Paulo, to Human Rights Watch, October 30, 1998. Of these twenty-one new prisons, nine were financed using 80 percent federal money and 20 percent state money, while the remaining twelve were fully state-financed. Human Rights Watch interview, João Benedicto de Azevedo Marques, Secretary of Prison Administration of the State of São Paulo, São Paulo, November 26, 1997.
101 "Ministério vê injustiça," Folha de S. Paulo, March 17, 1998. Federal funding for prison construction is collected via the National Penitentiary Fund (Funpen), established for this purpose in 1994. Administered by the National Prisons Department of the Ministry of Justice, Funpen is funded through court fees and lottery profits. Jobim, "Penas Alternativas . . . ," pp. 17-18.
102 Human Rights Watch interview, Judge Marco Antônio Scapini, Porto Alegre, Rio Grande do Sul, December 1, 1997. In August 1995, Secretary of Justice and Security José Fernando Eichenberg had said that he guaranteed that Porto Alegre's decrepit Central Prison in Porto Alegre would be closed the following year and that new prison construction was being undertaken. "Presídio está parcialmente liberado," Correio do Povo (Porto Alegre), August 7, 1995. It was still open in December 1997, when Human Rights Watch visited, but one wing had been ordered closed as unfit for habitation.