What happened
In July 1973 three men—J.W. Power, P. Selenski and J.A. Lyons—became involved in a brawl at a hotel in the Australian Capital Territory. The jury convicted Lyons and Selenski on two counts of assault occasioning actual bodily harm (Crimes Act 1900 (N.S.W.) s. 59, applied in the Territory) and convicted all three on one count of common assault (s. 61). Before passing sentence Blackburn J. received police antecedent reports and pre-sentence reports prepared by qualified social workers.
On the two counts of assault occasioning actual bodily harm Lyons received concurrent sentences of thirty months’ imprisonment with a non-parole period of eighteen months; Selenski received concurrent sentences of thirty months’ imprisonment with a non-parole period of twelve months. On the common assault count each received eighteen months’ imprisonment with a non-parole period of twelve months (Power received only this sentence). The non-parole periods were fixed under s. 4(1) of the Parole of Prisoners Ordinance 1971 (A.C.T.), which required the court to “specify a lesser term of imprisonment during which the person so sentenced is not to be eligible to be released on parole”.
Each applicant applied for leave to appeal to the High Court, complaining that both the head sentences and the non-parole periods were excessive. Counsel ultimately confined the argument to the non-parole periods, relying entirely on two recent New South Wales Court of Criminal Appeal decisions—Reg. v. Portolesi [1973] 1 N.S.W.L.R. 105 and Reg. v. Sloane [1973] 1 N.S.W.L.R. 202. Those decisions had held that, under parole legislation, a sentencing judge should not fix what he regarded as the minimum period of punishment justice required but should instead fix only the short period necessary to enable the paroling authority to form an opinion on the prisoner’s prospects of rehabilitation. Special leave was granted in each matter but, after hearing argument together, the High Court dismissed all three appeals.
Why the court decided this way
The joint judgment of Barwick C.J., Menzies, Stephen and Mason JJ. (with which McTiernan J. agreed in a separate judgment) rests on a straightforward reading of the Ordinance and a rejection of the policy premise underlying the New South Wales decisions. The court began by noting that the differences between the New South Wales Parole of Prisoners Act 1966 and the A.C.T. Ordinance were immaterial for present purposes. It then examined the chain of reasoning in Portolesi and Sloane.
That reasoning, the High Court said, proceeded from the view that after the introduction of parole legislation a sentence of imprisonment was no longer primarily punishment but “a provision for an opportunity for rehabilitation”. Consequently the judge’s task was said to be to set the shortest non-parole period that would allow the parole board to make an informed assessment. The High Court identified two passages that encapsulated the error. In Sloane the majority had stated that the proper approach was “not, as a matter of principle, to fix a minimum period of punishment in each case, but to consider in each case the question of possible rehabilitation”. The New South Wales court had also considered it incongruous for a judge to fix both a head sentence and a shorter non-parole period that he regarded as appropriate.
The High Court gave three decisive answers. First, the legislation does not convert a sentence of imprisonment into anything other than “a severe punishment”. Imprisonment, with or without hard labour, remains punishment even if the prison system is enlightened. The encouragement to reform provided by the possibility of parole “does not and obviously is not intended to take the sting out of imprisonment”. Confinement serves the same purposes—punishment directed towards reformation—both before and after the non-parole period expires. The only difference is that during the non-parole period the prisoner cannot be released on the ground that punishment has served its purpose.
Second, the judge’s statutory task under s. 4 is to determine “the minimum period for which in his judgment, according to accepted principles of sentencing, the prisoner should be imprisoned”. That task requires regard to all circumstances of the offence and the offender, including retribution, deterrence and rehabilitation. The judge is not required to make an “administrative guess” about how long the parole board needs to form a view. To read the Ordinance otherwise would impose a predetermined limitation on judicial discretion that the statutory language does not support.
Third, the supposed incongruity between a minimum and a maximum sentence disappears once it is recognised that there is only one sentence—the term fixed by the trial judge. The parole authority does not alter that sentence; it merely releases the prisoner conditionally from confinement. The non-parole period is therefore not a “minimum sentence” in the sense that converts the head sentence into an indeterminate one. Rather it is the judicially fixed minimum time the offender must serve before any question of conditional release can arise.
The joint judgment noted that legislation of the same kind had existed for years in Victoria and Western Australia and had not been given the narrow construction adopted in New South Wales. By contrast, South Australia and Queensland had given parole boards almost plenary power, with courts not obliged to fix any non-parole period. The A.C.T. Ordinance sat between these models: it required the court to fix a non-parole period but left the ultimate release decision to the paroling authority once that period had expired. That deliberate separation of functions confirmed that the sentencing judge’s role remained a sentencing role, not an administrative one.
McTiernan J. added that the crimes were crimes of violence and that Blackburn J. had plainly considered a considerable period of punishment by way of incarceration was required. No error in the exercise of discretion had been shown. Both judgments emphasised that appellate interference with sentence requires demonstration of error of principle or manifest excess (House v The King (1936) 55 C.L.R. 499). No such error appeared.
Before and after state of the law
Before Power the law on non-parole periods was unsettled. The New South Wales Court of Criminal Appeal had, in Portolesi and Sloane, introduced a novel limitation: the non-parole period should generally be fixed at the shortest time necessary for the parole board to assess rehabilitation prospects, with a punitive element reserved for “especial community need for the expression of revulsion”. That approach treated the legislation as having fundamentally altered the nature of a sentence of imprisonment.
After Power the law returned to orthodoxy. A sentencing judge must fix the non-parole period by reference to all conventional sentencing considerations—punishment (retribution and deterrence), rehabilitation, denunciation and protection of the community. The period represents the minimum time the offender must spend in custody having regard to the seriousness of the crime and the offender’s circumstances. Rehabilitation remains relevant but does not subsume the other purposes. The High Court’s insistence that there is only one sentence, and that the parole authority does not alter it, removed any conceptual difficulty about “minimum” and “maximum” sentences.
The decision also clarified the relationship between judicial and administrative functions. The judge sets the minimum custodial period; the parole board decides, after that period, whether conditional release is appropriate. That division, the court held, was a “clearly expressed policy of the legislation”.
Key passages with plain-English translation
The joint judgment contains several passages that have become canonical. One reads:
“It is our opinion that the Act as a whole does not convert a sentence of imprisonment from a punishment into an opportunity for rehabilitation. We cannot understand how a sentence of imprisonment, either with or without hard labour, can, however enlightened the prison system is, be regarded as otherwise than a severe punishment for a crime which has been committed…”
Plain-English translation: Prison is still meant to hurt. The chance of parole for good behaviour does not turn the whole exercise into a mere rehabilitation program.
Another key statement:
“The judge, in fixing a non-parole period, must, we believe, have regard not to the time within which the paroling authority must consider the prisoner’s case but to the time for which the prisoner must remain in confinement. The legislature in clear terms … provided that the trial judge should determine that minimum period for which in his judgment, according to accepted principles of sentencing, the prisoner should be imprisoned.”
Plain-English translation: The judge’s job is to decide how long the offender deserves to be locked up, not how long the parole board needs to fill out its paperwork.
The court also disposed of the supposed incongruity:
“We think nothing is to be gained by regarding the power of the paroling authority to release from prison as converting the full sentence into an indeterminate sentence. … In truth there is but one sentence, that imposed by the trial judge, which cannot be altered by the paroling authority.”
Plain-English translation: Parole does not rewrite the judge’s sentence; it simply lets the prisoner out early under conditions. The sentence itself stands unchanged.
Finally the court summarised the legislative purpose:
“To read the legislation in the way we have suggested fulfills the legislative intention … to provide for mitigation of the punishment of the prisoner in favour of his rehabilitation through conditional freedom, when appropriate, once the prisoner has served the minimum time that a judge determines justice requires that he must serve having regard to all the circumstances of his offence.”
Plain-English translation: The law lets offenders earn early release, but only after they have served the least time a judge thinks is fair punishment for what they did.
What fact patterns trigger this precedent
Power is engaged whenever a court in a jurisdiction with comparable parole legislation must fix a non-parole (or “minimum” or “ineligibility”) period. It is directly applicable in the Australian Capital Territory under the 1971 Ordinance and remains relevant in any State where statute requires the sentencing court to nominate a period during which the offender is ineligible for parole.
The precedent is triggered most acutely in cases where counsel invites the sentencer to adopt a short non-parole period solely to allow rapid parole-board assessment, or where it is argued that rehabilitation should predominate over punishment, deterrence or denunciation. It applies with particular force to offences of violence, breach of trust, or where there is “especial community need for the expression of revulsion”—precisely the category the New South Wales court had reserved for punitive non-parole periods but which the High Court treats as ordinary sentencing territory.
Because the decision affirms that ordinary sentencing principles govern the non-parole period, it is routinely cited in appeals alleging that a non-parole period is manifestly excessive or that the sentencer failed to give sufficient weight to rehabilitation, or conversely gave too much weight to punishment. It is also engaged whenever a sentencer is asked to treat the non-parole period as an administrative rather than a judicial act.
How later courts have treated it
Although the source judgment itself does not cite subsequent authority, it expressly records that the New South Wales limitation had not been adopted elsewhere and that the Victorian and Western Australian legislation had been administered for years without that gloss. The joint judgment’s careful analysis of the South Australian and Queensland schemes—where courts were not obliged to fix any non-parole period—has served as a legislative comparator in later cases considering amendments to parole statutes.
The decision has been treated as settling the proposition that the non-parole period is a sentencing exercise, not an administrative one. Courts have repeatedly cited the passage that there is “but one sentence” and that the parole authority does not alter it. The emphasis on the judicial obligation to fix a minimum period “according to accepted principles of sentencing” has been used to reject arguments that rehabilitation or parole prospects should be the sole or dominant consideration. Later benches have also drawn on the judgment’s observation that deterrence operates more powerfully when the prisoner knows he must serve a substantial non-parole period rather than being released almost immediately.
McTiernan J.’s short judgment reinforcing that crimes of violence will ordinarily justify a substantial non-parole period has been relied upon in appeals involving assaults, sexual offences and other crimes against the person. The High Court’s application of the House v The King principles to non-parole periods has become the standard of appellate review in this field.
Still-open questions
The judgment leaves several practical questions unresolved. First, precisely how the various sentencing purposes—retribution, deterrence, rehabilitation and protection—are to be balanced when fixing the non-parole period remains a matter of judicial discretion. The court rejected any “predetermined limitation” but gave no formula for weighting the factors.
Second, the precise relationship between the non-parole period and the head sentence is not exhaustively defined. While the court said the non-parole period cannot exceed the head sentence and must be shorter, it did not address whether there is any presumptive proportion or range that should ordinarily be observed.
Third, the judgment notes that in “particular cases” a sentencing judge may legitimately include a punitive or retributive element, yet it does not attempt to catalogue those cases beyond giving examples of violence and breach of trust. Subsequent courts have therefore continued to debate the boundaries of when denunciation or community revulsion may properly lengthen the non-parole period.
Fourth, although the court approved Blackburn J.’s approach on the facts, it did not discuss whether, or in what circumstances, a non-parole period might be fixed at or close to the head sentence. The theoretical possibility remains open, subject only to the statutory requirement that the non-parole period be “a lesser term”.
Finally, the interaction between this sentencing principle and later statutory reforms—such as standard non-parole periods, serious offender regimes or mandatory minimums—raises questions the 1974 court could not foresee. While Power supplies the default common-law approach, the extent to which it survives particular legislative overlays continues to require careful analysis in each jurisdiction.