Amnesty: Some Thoughts in Response to the Attorney General

We are delighted to welcome this guest post by Professor Bill Rolston, Director of the Transitional Justice Institute, University of Ulster. Bill can be reached at

Globally there tend to be two diametrically opposed positions in relation to amnesties. One, voiced by a number of lawyers and academics, is that amnesties are never acceptable. The state has a ‘duty to prosecute’ human rights violators and victims should be able to have their day in court where justice can be done and can be seen to be done. The other position is that amnesties are one mechanism that can be reached for when conflicts end; a line is drawn under the past and no one is made amenable to prosecution. This position is particularly favoured by leaders of previous authoritarian regimes, and it is hard to escape the conclusion that it is a position motivated by self-interest. That is certainly how it is seen by those in the first camp: amnesty is the way in which human rights violators ensure impunity for themselves and their mates.

In practice, there is often a pragmatic middle path between these two irreconcilable poles.

For example, in Northern Ireland the release of politically motivated prisoners was certainly not an amnesty – ‘residual criminalisation’ means that they are blocked from certain employment possibilities and benefits and can be legally discriminated against in certain circumstances; but with anyone now sentenced for a politically motivated offence that occurred between 1968 and 1998 facing a maximum two year prison sentence it does begin to look remarkably like an amnesty.

It is true that state actors were not given such a dispensation and would face serving a full sentence if found guilty now for an offence committed during the conflict. There is an argument which says that, in terms of equity, the benefits of the Good Friday Agreement should be equally shared across the board, although the state might have a huge problem with, in effect, equating the actions of their soldiers and police with those of ‘terrorists.

A form of ‘transactional amnesty’ was granted in relation to the issue of arms decommissioning and also in relation to the recovery of the bodies of the disappeared. Statements made by people assisting in the recovery of bodies were ‘protected’, that is, not open to being used as evidence for prosecution, and there was an agreement in both the decommissioning and recovery of bodies processes that there would be no forensic examinations involved and therefore no collection of evidence that could be used in a court of law. Further afield, there are other examples; for example, in South Africa’s Truth and Reconciliation Commission amnesty could be granted to individual perpetrators in return for genuine disclosure of truth.

And therein lies the clue to this ‘middle ground’ – that amnesty is not the blanket application of impunity but the result of a trade-off: amnesty in return for truth in South Africa; non-prosecution in return for the delivery of weapons or disappeared bodies in Northern Ireland. At the base of both approaches is a philosophical and practical judgement that moving forward requires some transitional arrangements which fall short of ideal type legal arrangements. If the price of peace, progress, stability is that some fudge is required in the duty to prosecute, then the society, often reluctantly, judges that that is a price worth paying. This is the same logic behind other such pragmatic solutions globally – for example, pursuing the prosecution of high-ranking military personnel while avoiding clogging up the criminal justice system with endless prosecutions of rank and file military personnel.  It is also, ultimately, the logic behind the legal position apparent in much international humanitarian literature, that certain crimes – torture, genocide, rape as a weapon of war – are non-negotiable in terms of amnesty; the corollary is that other offences are not automatically ruled out in discussions of amnesty.

The outcome is that, if societies want to seriously consider the possibility of amnesty, their deliberations must be in the context of this pragmatic middle ground. What is the trade off for amnesty? What are the costs and benefits which result and might the latter outweigh the former?

It would be nice to think that Attorney General John Larkin is trying to kick start this debate in Northern Ireland by suggesting an amnesty process. But the conclusion has to be much less sanguine than that. He did not suggest conditionality or trade off, nor present a balance sheet of possible costs and benefits. Instead, what he has suggested sounds remarkably like the sort of blanket amnesties which Latin American dictators conferred on themselves at the end of their regimes. If so, it is worth drawing a sober lesson from that experience. There has been in Latin America in recent years what has been termed a ‘justice cascade’, a wave of prosecutions for human rights offences of decades before. This is in the context where there had been in effect an ‘injustice cascade’ in those societies where dictators and military chiefs had written a ‘get out of jail’ card for themselves. Lawyers and NGOs have now found ingenious ways to challenge or bypass such blanket impunity; for example, recognising that an amnesty applies only to military personnel, they have gone after civilian leaders of repressive regimes. If Larkin’s suggestion in any way mirrors the Latin American experience, then it may seem to ‘solve’ a problem only to kick it down the road for the next generation.

Amnesties and inquiries in different ways subvert the ordinary operations of the law, one by avoiding prosecutions, the other by investigating issues that may not be adequately handled by the criminal justice system. The logic of each is that post-conflict transformation requires imaginative and unusual innovations beyond what regular law can offer. Larkin’s suggestion is a confused mixture of urging what could be represented as a key transitional mechanism – amnesty – while ruling out another – inquiries.

But more worrying is his suggestion about inquests. Inquests are not a transitional mechanism. They emerge from a long history of struggle to produce a mechanism which can protect citizens by holding the state to account when sudden or violent death occurs. There is no legal ground for abandoning inquests. The state cannot absolve itself of its Article 2 obligations. Moreover, the British state is also bound by the European Convention of Human Rights and the European Court is unlikely to uphold such a widespread derogation. The suggestion has an added frisson in our society where banning inquests was for decades a central plank in the discredited Special Powers Act and where more recently families have struggled in some cases for two decades and more to wrest information from British intelligence sources to ensure that the inquest system can in fact work. The way forward is surely for inquests to be given the space to operate to their best ability rather than curtail them even further.

An empowered system of inquests could combine with other mechanisms – inquiries, trials, truth recovery processes, storytelling, etc. – in a genuinely comprehensive attempt to deal with the past. This is the vision that Eames and Bradley sketched out in their report of the Consultative Group on the Past. And it should be remembered that they suggested the possibility of ‘protected statements’ in relation to the operations of their proposed Legacy Commission. That is the context in which a less than full application of the duty to prosecute can be imagined – as part of a joined up process where amnesty, if considered, is linked to the other elements of the process. John Larkin would have done us all a much bigger favour if he had resurrected that debate instead.