In a criminal trial, facts in issue need to be proved to the relevant standard. For example, the elements of the offence charged must be proved beyond reasonable doubt. In civil cases, the facts that have to be proved to make out a claim must be proved on the balance of probabilities.
Parole hearings are judicial (but inquisitorial) and civil. Where a prisoner who is subject to an indeterminate sentence, such as imprisonment for life, applies for parole, the governing legislation may provide, for example, that the Parole Board may not direct release on parole unless it is “satisfied that it is no longer necessary for the protection of the public that the prisoner should be confined.”
In that legislative enviroment, [1] one might ask whether this absence of a need to protect the public is a fact in issue that must be proved on the balance of probabilities before release may be ordered.
This was a point of contention in R (on the application of Pearce) v Parole Board [2023] UKSC 13 (5 April 2023). The proceedings in the Supreme Court did not involve the applicant for parole, as the release issue had been settled, but the appropriateness of guidance given administratively to the Board was under review.
Nevertheless, the judgment is of general interest because of its treatment of information received by the Board which amounted to accusations about the applicant's dangerous conduct, although those accusations were never proved. In what circumstances, if any, could the Board use unproven allegations to reach a conclusion that release on parole should be declined?
If an allegation has not been proved, that may still mean that it was nearly proved and that it raises sufficient concern to justify its being taken into account. But some accusations may never have reached the courtroom, or even been the subject of a formal complaint. Plainly, in those circumstances, the Board would have to be careful to comply with the requirements of fairness to the applicant.
The premiss of the legislation is that, for this category of prisoner, release will not be ordered unless the prisoner satisfies the Board that confinement is no longer necessary for the protection of the public. The prisoner’s absence of dangerousness is in issue in the proceedings, and taking unproven allegations into account on that issue is to do what in a trial would be unacceptable.
But parole proceedings, although they must be conducted in compliance with the rules of natural justice, are not trials, and the Board is not required to confine its deliberations to admissible evidence. The prisoner is not in jeopardy of conviction in a parole application, and indeed the prisoner can only gain an advantage without risking further penalty. So the decision process can be described as evaluation of risk based on an holistic assessment of all the circumstances. [2]
“[A] decision-maker, whether a member of the executive branch of government or a judicial body, when assessing future risk, is not as a matter of law compelled to have regard only to those facts which individually have been established on the balance of probabilities; the decision-maker, from the assessment of the evidence as a whole, can take into account, alongside the facts which have been so established, the possibility that allegations, which have not been so established, may be true” (at [44], and see the summary at [65] and the conclusions at [72]-[87]).
This case applies where a tribunal acts judicially and inquisitorially to assess a risk but without being bound by the rules of evidence. There may be a tendency for more issues in criminal law to be approached in that way, for example as a result of advocacy of the introduction of special courts to try allegations of a sexual nature. Would a move away from jury trials permit the abolition of the law of evidence? Would trial by expert assessors of information be any better than trial by jury? [3]
__________________________________
[1] Section 28(6)(b) of the Crime (Sentences) Act 1997 [UK].
[2] See [83] and [87(vii)].
[3] The use of unproven allegations as evidence may remain highly controversial, and would not necessarily be endorsed by logicians. Truth is a matter of veracity and reliability, and, as those are not logical matters, a logician would have to treat an allegation as if it were true, and ask what it then would say about the acceptability of the risk. The current potential for replacing judges and judicial tribunals with some sort of AI (which need not be confined to logical processes) is unclear, although the tools used by officials such as probation officers to assess an offender’s dangerousness or likelihood of recidivism could be seen as major steps in that direction.
