Review of continued detention of imprisoned terrorists

Karen Thornton

Executive Officer

Independent National

Security Legislation Monitor

3-5 National Circuit Barton

Karen.Thornton@ag.gov.au

Dear Ms Thornton

Review into Division 105A of the Criminal Code Act 1995 (Cth) (Criminal Code)

I refer to your email to the QCCL of 12 September 2021.

Subject to the following comments, we endorse the joint submission of the New South Wales Council for Civil Liberties and the Sydney Institute of Criminology (“the joint submission”)[1].

The argument is advanced in the joint submission that in certain circumstances preventive detention should be accepted because the right to life is said to impose an obligation on a State to protect the lives of its citizens by continuing to detain those assessed as being at risk of reoffending.

We respectfully dissent from that conclusion.

The presumption of innocence dictates that the State has no higher duty to protect its citizens from the risk a person released from prison might commit an offence than it must protect them from the risk other people walking the street might commit one. Whilst the presumption of harmlessness may be rebutted by past behaviour the presumption of innocence is not.

Moreover, the duty referred to in the joint submission can only come into existence when the State knows that a person is going to commit a crime.

As the joint submission makes clear there is great difficulty in saying that the State has or can have the requisite degree of knowledge that a person currently in prison is if released going to commit crime[2].

As was noted some years ago the concept of dangerousness which underlies this legislation has two core characteristics[3] :

“The first is that as a prediction dangerousness is an estimation of the potential that a person will do something that is defined as dangerous.  As such dangerousness is a perception of the evaluator and not a characteristic constant or otherwise of the evaluate.  Second, dangerousness is by its nature a prediction.  It means that because of certain characteristics or behavior a person is seen to have a high probability of performing certain acts in the future.  Thus, the essence of dangerousness is that it is a perception and a prediction.”

The joint submission points quite correctly to the many inadequacies of the basis for the prediction the Courts are required to make in relation to a person’s likelihood of reoffending.

The inadequacy of this predictive process was underlined by a “real-life experiment”. Kozol, Boucher & Garofalo[4] reported on the dangerousness of a population consisting mostly of sex offenders. Of the 226 accepted for treatment, staff subsequently discharged 82 with a recidivism rate of 6.1%. 49 patients were released by Court Order, against the advice of medical staff, with a subsequent recidivism rate of 34.7%. Clinical judgment to retain the 49 high risk individuals was wrong in 69% of cases. A decision to release all of the dangerous offenders would have been 65% correct.

It will no doubt be said that the science in this area has moved on since 1972. However, that proposition must be seriously questioned. The usual basis for that assertion is to refer to the findings of actuarial risk assessments which have been developed since that time. Again, in our respectful submission correctly, the joint submission points out that in cases before the Courts these risk assessments form a part only of the clinical judgement. As they must, because those tools assess the risk of the person reoffending by reference to the status of that person as a member of a demographic group or groups, not by reference to the particular characteristics of the person in question. The expert’s final opinion is still a matter of clinical judgement and ultimately a judgement is made by the Court about the individual before the Court.

Even the strongest claims in support of the reliability of these risk assessment models to predict future dangerousness claim a success rate of only 70-80%. Meaning that at least 20 to 30% of those detained are liable to be “false positives” ie people who would not have re offended

Amongst all this epistemological uncertainty one thing is certain- the State knows there is a probability that its actions may result in harm. Harm to a member of the community, ie the victim of a crime committed by the released person or the person who is detained but would never have offended. Is not the State more morally culpable for any harm it inflicts than for the harm inflicted by the third party? We would submit that on normal moral principles that is the case.

Furthermore, liberal principles limit the power of the State to imprison people, not because it values accused or convicted persons more than victims or potential victims, but to protect everyone from excessive government power.

Finally, there is practical consideration. Once we start permitting the extension of expired terms of imprisonment based on predictions of what the person might do, where does it end? We have already seen it now extended from sexual offenders to terrorism offenders.

As John Stuart Mill argued the preventive power of the State is, “far more liable to be abused, to the prejudice of liberty, then the punitory function; for there is hardly any part of the legitimate freedom of action of a human being that would not admit of being represented, and fairly too, as increasing the facilities for some form or other of delinquency.”[5] These strictures clearly apply to this type of legislation

We trust this is of assistance to you in your deliberations.

[1] https://www.inslm.gov.au/sites/default/files/2021-12/1.NSW-Council-for-Civil-Liberties-and-Sydney-Institute-of-Criminology.pdf

[2] I acknowledge my indebtedness for what follows to materials supplied by and discussions with one-time QCCL executive member and forensic psychologist Geoffrey Grantham.

[3] H J Steadman The Right Not to be a False Positive, Problems in the Application of the Dangerousness Standard Psychiatric Quarterly 1980 52(2), 84-99

[4]Kozoil A.L, Boucher R. J & Garofalo, R. F. The Diagnosis and Treatment of Dangerousness 1972 18 Crime and Delinquency 371-392

[5] Three Essays: On Liberty, Representative Government and the Subjection of Women Oxford University Press 1911 at page 117