Slovenian npm's 10th anniversary conference report One approach would be to consider NPM impact as observable change triggered by the mechanism, such as a modification of practice or legislation, an improvement of living conditions or a judicial decision on a detention-related issue. However, this assumed that causality between the NPM’s work and the change observed could be established. Moreover, given that the overall objective is prevention (of illtreatment), an NPM’s impact could not necessarily be assessed in terms of change; the mechanism’s impact might consist of maintaining the status quo i.e. avoiding a worsening of the situation. The very presence of an NPM in a place of deprivation of liberty might in itself be regarded as a form of impact, as closed institutions were thereby rendered visible to the outside world. However, it was argued that the presence of an NPM was not enough and could even become a fig leaf if it led to no tangible benefits in terms of strengthening protection from ill-treatment. It was agreed that criteria on impact should address inter alia the «internal performance» of an NPM. Whether the mechanism concerned was able to have an impact would certainly depend to a considerable extent on how it went about the business of carrying out visits and drawing up reports and recommendations. Does the NPM have the necessary legal basis (independence, powers) and resources (financial and human); is there a well-defined visits strategy covering all forms of deprivation of liberty; are visits sufficient in number; are they properly prepared and carried out in a professional manner by a suitably-qualified team; is information sought from all relevant sources; are visit reports and related recommendations focussed on the key issues in terms of the prevention of ill-treatment and drafted in a «user-friendly» way; etc ? According to several participants, the focus should be less on the number of visits and recommendations and more on their quality combined with a strategic choice of establishments visited. In this connection, the advantages of thematic visits were emphasised. At the same time it was underlined that an NPM’s impact could not be assessed merely by reference to visits and reports. The involvement of the NPM in the process of transforming recommendations into positive change was equally important. The mechanism was not the decision maker, but it should seek to influence the outcome through dialogue with those possessing the power of implementation (the national authorities) and, when necessary, by enlisting the support of others. This implied the creation of procedures/structures enabling a meaningful dialogue with the administration to occur. And if that dialogue proved fruitless, it was incumbent on the NPM to mobilise other forces (parliament, the judiciary, the media, staff trade unions, NGOs andpublic opinion) in favour of change. In this context, reference was made to the «political» nature of the mechanism’s work. Some participants considered that criteria on impact could be devised by reference to the standards which the NPM seeks to have respected. The mechanisms should certainly be expected to strive to ensure compliance with the (minimum) standards which have been developed at regional and/or international level in relation to many types of places of deprivation of liberty. Of particular interest were the wellknown «fundamental safeguards» which should apply as from the outset of custody - notification of a third party as well as access to a lawyer and to a doctor. Recent research commissioned by the APT had confirmed that these safeguards, if both recognised by law and applied in practice, significantly reduce the risk of torture and other forms of deliberate ill-treatment. As suggested in the concept paper prepared for the September 2017 Paris meeting, the impact of an NPM might be assessed in part by the extent to which it ensures compliance with these three key safeguards in the country concerned. Similarly, was the NPM delving into the measures taken when there were indications of possible illtreatment; did it check whether investigations were carried out by the relevant authorities and examine the «effectiveness» of any such investigations ? It was widely acknowledged that firm action when illtreatment occurred had a powerful preventive effect. For several participants, whether an NPM engages in the non-visit activities listed in paragraph 9 of the SPT’s analytical assessment tool should also be taken into account when assessing impact. Particular reference was made to training for staff with responsibility for persons deprived of their liberty; if staff learned how to adopt the appropriate attitude in their relations with persons in their custody, this would pay dividends in terms of the prevention of ill-treatment. The importance of instruction for medical personnel on application of the Istanbul Protocol was also emphasised. Similarly, systematic training for the police, prison officers, etc on the NPM as such - its role, powers and working methods - could have a positive effect. IMPLEMENTATION OF THE DUTIES AND POWERS OF THE NPM IN 2018 NATIONAL PREVENTIVE MECHANISM UNDER THE OPTIONAL PROTOCOL TO THE UN CONVENTION AGAINST TORTURE AND OTHER CRUEL, INHUMAN OR DEGRADING TREATMENT OR PUNISHMENT 53