THE debate on bail has been topical since at least 1978 when the then Minister for Justice first promised to promote a referendum on the topic. By now many people are familiar with the broad thrust of the Supreme Court's decision in the O'Callaghan case (1966) which affirmed that bail could only be refused where it was likely that the accused would abscond or interfere with witnesses.

The court trenchantly rejected the suggestion that the possibility that the accused would commit further offences if released on bail saying it would be contrary to the guarantee of personal liberty enshrined in Article 40.4.1 of the Constitution that "any person should be punished in respect of any matter upon which he has not been convicted or that in any circumstances he should be deprived of his liberty only upon the belief that he will commit offences if left at liberty . . ."

While the result of this decision may well have proved controversial, it appears to have been regarded by most lawyers as unquestionably correct (although Mr Justice Keane of the Supreme Court, as a barrister, wrote an incisive critique of that decision in an article published in 1967).

It is, perhaps, less well known that the Supreme Court has subsequently refused to overturn this decision, most recently in the Ryan case in 1989. There the court was urged to adopt a test whereby bail could be refused if there was a real likelihood the accused would commit fresh offences on bail, but the court declined to do so, observing that "the criminalising of mere intention has been usually a badge of an oppressive or unjust legal system".

The demands for reform have grown apace over the last decade. We now recognise that classes of habitual offenders have emerged which were practically unknown at the time of the O'Callaghan judgment, among them professional criminals, child sex abusers and drug addicts. Whatever the niceties of the presumption of innocence, it is a fair guess (even if not readily susceptible of proof in a court of law) that a heroin addict with a long criminal record is likely to offend again if released on bail, if only to secure money to feed his addiction through robbery.

There is evidence that the High Court has on occasion misapplied the O'Callaghan ruling when granting bail. Thus, for example, there have been instances of where persons were granted bail who were either caught red handed or who had even previously escaped from lawful custody and who, in the event, either absconded or subsequently committed serious offences while on bail.

Perhaps in some cases the courts overlooked the fact that the O'Callaghan rules permitted the courts to take account of the gravity of the charge, the strength of evidence and the likelihood of a lengthy sentence and that, had they done so, bail would not have been so readily granted in some of the more recent and controversial bail decisions.

The Government's proposal is likely to affect only a small number of cases and, even then, is likely to be fraught with considerable practical difficulties. The proposed amendment would only apply to a "serious offence". While this term is properly left undefined (since any attempt to define it with greater specificity would be out of place in a Constitution), it will give rise to litigation in marginal cases. Murder, manslaughter and rape clearly qualify as "serious offences" for this purposes, but, would, for example, shoplifting or other forms of petty theft?

While bail may be refused "where it is reasonably considered necessary to prevent the commission of a serious offence by that person", if it is accepted that the accused will re-offend if granted bail, it will be therefore necessary to prevent the grant of bail to that person.

Perhaps a better wording would read "if the court is of opinion that such person will commit a serious offence if granted bail".

But the key question is how the prosecuting authorities can satisfy a court that the accused is likely to commit another serious offence if released on bail. In most cases, this would be difficult, if not impossible.

IT IS true that one could readily envisage strong cases such as serious sex offenders where the likelihood of repeat offences is very real. But even here the prosecution will encounter difficulties of the kind anticipated in the Ryan case.

Chief Justice Finlay highlighted some of these problems:

"How can such an intention [to commit a future crime] be proved, and by what standard of proof could it be established? Could there be any grounds on which an accused person suspected of such an intention would be afforded less comprehensive notice of the evidence to be offered against him of the grounds for such suspicion and less opportunity to prepare and to be represented to contest such allegations than he is afforded in relation to the presenting of a criminal charge against him? Would every application for bail accordingly, in which this ground was advanced as the substantial ground of opposition, take on the nature and necessary requisites of a criminal trial?"

This passage gives a clear indication that even under the Government's proposals, prosecuting authorities who want the court deny bail on the ground that the accused may offend in future will find such bail motions may take on the character of a mini-criminal trial. As a matter of ordinary fair procedures the accused will be entitled to advance notice and particulars of the allegation that he will reoffend.

This will, in all probability, present the prosecution with a well nigh impossible standard of proof, especially where, as will probably be the case in the majority of such applications, the accused has not got around even to contemplating what future offences he might commit.

It is difficult to see how the Government's proposals will have any real effect, save, perhaps, in marginal cases. While the public has been understandably frustrated by what (in my judgment) has been the excessively liberal manner in which the O'Callaghan principles have, on occasion, been applied, those who seek in this measure a panacea for an unacceptably high crime rate are likely to be sorely disappointed.