Roberts, R (on the application of) v Secretary Of State For Home Department

A prisoner's denial of guilt cannot by itself prevent a reduction in security category. The authorities must nevertheless assume the conviction to be correct. Denial may severely restrict the available evidence that the risk of similar offending has diminished, particularly where it prevents participation in offence-focused work.

A categorisation review must assess present dangerousness. It need not anticipate the effect of its decision on a future parole assessment, although an actual Parole Board decision may become relevant. For a highly dangerous prisoner, the authorities may treat accommodation below category A as exceptional and ask whether anything unusual requires detailed examination of escape potential.

The claimant was a life prisoner who continued to deny the murders of which he had been convicted. He sought judicial review of the Category A Review Team's decision of 20 May 2003 to retain him in category A.

The challenge concerned whether the team relied irrationally on his denial and consequent non-participation in offence-focused courses, gave insufficient weight to his good custodial behaviour, failed to consider the consequences for future parole, and applied the wrong approach to his actual escape potential.

1. The application was dismissed. The review team lawfully assumed the correctness of the claimant's convictions and required cogent evidence that the danger inferred from the grave index offences had diminished. A denial of guilt was neither irrelevant nor necessarily conclusive. Its significance lay in its effect on the prisoner's ability to address his offending and demonstrate reduced risk. The principles concerning prisoners in denial stated in R v The Parole Board and Secretary of State for the Home Department, ex p Oyston and R v Secretary of State for the Home Department, ex p Hepworth and Others applied equally to security categorisation, although parole and categorisation assess public safety in different contexts: paras 33–42.

2. The weight assigned to good custodial behaviour, constructive conduct and participation in general rehabilitation work was for the review team. Judicial intervention would require a perverse conclusion. The team had considered those matters, and no reasonable team was compelled to find that they demonstrated a significant reduction in the risk of similar offending. Its statement concerning the reasoning and rehabilitation programme, read sensibly, meant that participation was insufficient by itself, not that it had no evidential value: paras 48–52.

3. The review team was required to determine whether the prisoner remained so dangerous that escape had to be made impossible. Before the Parole Board had considered the case, the team neither had to nor should attempt to anticipate the Board's future approach. Once an actual Parole Board decision indicated that category A status might defeat or undermine an opportunity for parole, the Board's assessment became a relevant factor, particularly on danger to the community. The categorisation function nevertheless remained with the prison authorities and could not be fettered by the Board: paras 53–61.

4. Following William Pate v Secretary of State for the Home Department [2002] EWHC 1018 (Admin), escape potential could not be excluded altogether. The Prison Service could, however, begin from the legitimate policy that category A safeguards were required for highly dangerous prisoners and ask whether anything unusual took the case outside the ordinary position. This was an exceptional threshold question, not a balancing exercise. Where potentially exceptional considerations existed, they required careful attention; otherwise no detailed escape-risk analysis was necessary. The team applied that approach lawfully, and any error could not have affected the result: paras 62–71.

not stated in the judgment.