E/CN.4/2006/6/Add.2 page 56 arrest by judges is demonstrated in a review of SIZO population dynamics in 2003, a year when the judicial arrest procedure produced its first tangible effects. In early 2003, the remand prison population was 145,000, while in June 2003 it reached 156,000. A drop in the number of detainees in 2004 was due to the adoption in December 2003 of the Federal Law amending the Criminal Code (Law No 162-FZ). The law enacted 257 amendments, most of them liberalising the criminal legislation, decriminalising certain acts, mitigating criminal sanctions, and limiting the use of incarceration. As an immediate result of the amendments, many criminal investigations were closed, and the accused individuals released, while the inflow of new detainees to SIZOs decreased. 259. The Government informed that in accordance with the federal special programme on reform of the penal correction system of the Ministry of Justice for the period 2002-2006, in the first nine months of 2005 ten new pre-trial detention facilities with the capacity to hold 3200 persons were created. Overall since 1995, more than 17,000 places have been made available. Until the end of the year, 10,000 more places for holding suspects and accused will be made available. In addition, in accordance with the law “On pre-trial detention of persons suspected or accused of having committed a crime,” within 134 prisons, 8,734 places for holding suspects and accused persons have been created. 260. Recommendation (e) stated: The draft Code of Criminal Procedure, giving effect to article 22 of the Constitution which places all deprivations of freedom under judicial authority, should be speedily adopted by the State Duma. According to information received from NGOs, a new Criminal Procedure Code (CPC) was adopted in December 2001, and parts of it came into force in 2002. Under the new CPC, the courts, rather than the procuracy, have the responsibility for determining whether or not a suspect will be held in detention during an investigation. Previously, the procuracy was responsible for determining whether a person would be detained pending investigation as well as for investigating the case. The new CPC also places more emphasis than the old code on alternatives to pre-trial detention, such as bail or release on recognizance. However, subjective factors can jeopardize the effective implementation of the new CPC. For example, research indicates that very few people within the Russian Federation’s criminal justice system genuinely subscribe to the presumption of innocence. This is true not only of officials from the procuracy and the Ministries of Justice and Internal Affairs, but also of judges and even defence lawyers. There is a perception that any defendant acquitted by a jury, or any suspect released by a court, is a criminal who has been delivered back into society unpunished. Up to now suspects of crimes are held in custody for extremely long terms. Though illegal, administrative punishment of criminal suspects is widely used. A detainee can be sentenced to administrative arrest for a period for up to two weeks. During this period a detainee is deprived of procedural guarantees and is often forced to give testimony. Federal Law No. 18-FZ of 22 April 2004, which amended article 99 of the Criminal Procedure Code, provides that in “exceptional circumstances” suspects of “terrorism” may be arrested for up to 30 days without receiving an official indictment, which increases the risk of an individual being forced to confess or testify. Cases of torture are reported in North Caucasus in which the new safeguards provided by the CPC have been shown not to be effective. For example, the requirement to bring a detainee before a court within 48 hours is sidestepped through practices, such as arbitrary detention by law enforcement officials who fail to identify themselves, are masked and drive in vehicles with identification plates covered; falsification or failure to properly register detentions; and detention of persons in unofficial