Crystal Balls

It has been shown, in comments relating to IPP sentences, that some prisoners have committed awful crimes after their release. This is as undeniable as it is deplorable.

IPP sentences are based on a simplistic idea - that some (many thousands) of criminals should be detained until they can demonstrate that they are safe to release. This is more than sentencing with the aim of incapacitation. This is explicitly rehabilitative sentencing, detaining people not for the crime they have demonstrably committed but for the crimes they may commit in future.

There is an obvious objection to this, though it can be dressed up in an infinite variety of sophisticated clothing. The objection is the basis for the sentence, which is a fundamental shift in sentencing philosophy. Sending a person to prison for their past acts is one thing; continuing to detain them on the basis that they MAY commit future crimes is, to me, fundamentally objectionable.

Not only is it repulsive on a philosophical level, it is also nonsense on the actuarial level. If this sentencing philosophy is to be clothed in the most meagre scraps of sanity, it must be possible to demonstrate that future offending is predictable at the individual level.

Obviously, on the aggregate level this can be done. A careful analysis of the histories of prisoners highlights factors which suggest a higher chance of future offending. More prisoners have been in local authority care and have poor educational attainments than non prisoners, for example. And so, on this actuarial level, it can be honestly said that people with those biographical attributes are more likely to end up in prison.

Yet this is a far cry from being able to pick any individual with those attributes and predict their chances of committing crime. This is an impossible prediction to make, though the prison and probation services refuse to face that mathematical truth.

IPP sentences, keeping people in prison for what they may do in future, is based on a statistical lie that attempts to mask a philosophically repugnant shift in sentencing policy.

Of course, some prisoners will commit crimes after release. This is a price individual victims pay in order to protect society from near-arbitrary sentencing. These victims bear a heavy cost, a reminder that that freedom itself must be paid for. A price also paid by thousands of prisoners, kept in prison for what they may possibly do in future.

And yet, I would argue that sentencing should be based upon the crime that a person has committed and never on what he may do in future. At the end of that punishment they should be released. Their future conduct must then be judged on its own merits.