Part I: Basis for an effective follow-up Where NPMs do prioritise recommendations this was seen as a useful tool for follow-up as it can provide the NPM with a specific focus for any subsequent follow-up visit. Prioritisation may be appropriate in certain but not all circumstances depending on the number of recommendations issued at any one stage and the ability to use different categories of recommendations. At the very least our research found that it would be appropriate for an NPM to have a policy on whether or not and in which circumstances it should prioritise some recommendations. Standards In formulating recommendations, a range of factors may be taken into account by the NPM. These include international and domestic law, regulations, jurisprudence, rules and codes of conduct. Article 19(b) OPCAT requires that the recommendations of the NPM ‘take into consideration the relevant norms of the United Nations’. The reference to regional or international standards gives further credibility to recommendations and is used by most NPMs interviewed in the Study. The extent to which they do so, however, appears to be inconsistent and some NPMs have recognised the need for this to be taken more seriously. Nevertheless, as stated by the UK NPM, ‘recommendations where their requirement is based in legislation or has emanated from good practice are more likely to be implemented’.96 Furthermore, as noted earlier in relation to measuring implementation, the UK’s HMIP uses published inspection criteria, ‘Expectations’, which are based on and referenced against international human rights standards.97 Most NPMs regularly refer to the CPT’s recommendations in their recommendations. At the same time, the Austrian NPM (Austrian Ombudsman Board) mentioned that, while international standards are important, they are only useful to a certain extent because they constitute minimum standards that are often not sufficient. The NPM thus aims to develop its own standards through interdisciplinary expertise and drawing upon external experts. Silvia Casale underlined in this respect, that the development of standards should be guided by the pro homine principle, meaning ‘the standards developed should be those most capable of benefitting individuals deprived of liberty.’98 According to her, there will naturally be a difference in standards of between international and national bodies. The need for universal applicability ‘tends to reduce the capacity for standards developed at the UN level to be precise in the sense of including operational specifics. […] As soon as operational details are included, the standards become less relevant to some settings.’ International regional standards, e.g. by the CPT, tend to be more specific, ‘reflecting the fundamental principles, but also elaborating to some degree on the technical operational details’. According to Casale, NPMs are in the best position to take the progression in standards further: ‘Ideally the process of improving the situation of persons deprived of liberty should benefit from the development of NPM standards, consistent with global and regional standards but adding another layer of operational specificity.’ The French NPM furthermore explained that its recommendations can be based on good or bad practices, e.g. if a positive practice is witnessed in one institution this is then used in the recommendations in order to be picked up in other establishments. Targeted recommendations Many NPMs considered the ability to identify a particular institution or body to which the recommendation could be addressed was an important tool as it engendered a greater degree of responsibility. Inevitably many NPMs address a significant proportion of their recommendations to the actual institution visited.99 Indeed, in Portugal, for example, article 38(1) of the Ombudsman Statute provides that ‘The Ombudsman’s recommendations shall be addressed to the body with the power to correct the illegal or unjust act or to UK, NPM, Making Recommendations (2014) para 5. See UK, HMIP, Inspection Framework (2014) para 2.21. 98 Comments provided by Silvia Casale during the peer review of this Study. 99 E.g. the UK, NAO, Inspection (2015) para 2.5, where it is noted that HMIP ‘directs over 80% of its recommendations at the prison’s governor’. 96 97 - 40 -