In light of the likely agreement between the DUP and the Conservative Party, there has been speculation on whether a side-deal might exist on a statute of limitations for legacy offences committed by the security forces. The idea was raised by the DUP in a Commons debate and appeared in its manifesto. It was also proposed in a recent Defence Select Committee report. Such a move would present very serious legal and practical difficulties.
The United Kingdom Parliament is sovereign, and so introducing some form of amnesty or statute of limitations (the terms are interchangeable in this instance) is legally possible under UK law. However, as the Defence Committee correctly acknowledged, such legislation would have to be accompanied by some form of truth recovery process.
An amnesty would be outside the terms of the Stormont House Agreement (SHA). None of the parties argued in favour of an amnesty during the negotiations.
With a consultation on the SHA legacy legislation likely to start soon, arguing for the legislation to provide an amnesty would represent quite a departure for the DUP.
Equally, it's hardly credible to claim that amnesty discussions in Westminster are somehow separate from the local legacy negotiations.
The Defence Committee also stated that the decision on whether the statute of limitations should include paramilitaries was for the next government to consider. I cannot see how one can be introduced without extending it to them.
A statute of limitations for security forces only is what the United Nations has referred to as a "self-amnesty", wherein an amnesty "is adopted by those responsible for human rights violations to shield themselves from accountability". Other jurisdictions which introduced such self-amnesties have included Argentina, Chile and Brazil. Such amnesties are often introduced by outgoing military dictatorships to protect those loyal to the previous regime. For the UK to join such a list would be quite extraordinary.
Moreover, as the UN and a number of international courts have noted, self-amnesties breach international law as "by their nature, (they) epitomise impunity".
A further legal challenge for a 'State actors only' amnesty is what would happen in cases involving collusion between the State and loyalists or republicans. For example, we know that a small number of UDR and RUC personnel were members of loyalist paramilitary organisations or acted in tandem with such groups. Would an amnesty apply to such activities?
Another illustration of the messiness of such a process is illustrated by the Stakeknife case. Chief Constable Jon Boutcher is currently heading up the investigation into the activities of the alleged former agent and head of IRA internal security. Boutcher's investigation explicitly includes members of the IRA, British Army, security services or other Government agencies.
If such an amnesty is introduced, any evidence amassed by Mr Boutcher could not be used in prosecutions against security force or security services (e.g. M15) personnel. Would that guarantee extend to IRA members who were also British agents? Moreover, if he produces evidence against former IRA members, their lawyers are likely to argue that the statute of limitations (which would inevitably obscure the involvement of agent handlers) would represent an abuse of process against their clients.
In short, a statute of limitations for security forces would mean that any collusive element to cases such as Kingsmill, Claudy or Loughinisland would make prosecutions of the paramilitaries involved very difficult.
Arguments have been made that the rationale for the security forces amnesties was to redress a perceived imbalance between State actors and the treatment of paramilitary prisoners or suspects post-Belfast Agreement. Certainly, equality of treatment and non-discrimination are long established principles of both human rights and British common law. These arguments will be tested in court.
One argument has been a perceived imbalance in conflict-related prosecutions, particularly since Barra McGrory became Director of Public Prosecutions in 2011. However, as The Telegraph reported earlier this year, since 2011 seven republicans have been prosecuted, three loyalists, three soldiers and one police officer.
Another argument raised was that the early release provisions of the Good Friday Agreement did not apply to soldiers or police officers. That legislation, the Northern Ireland Sentences Act 1998, means that no one convicted of a pre-1998 scheduled offence can serve more than two years.
It is true that anyone convicted of an offence committed before 1973 - when 'scheduling' was introduced - is still liable to serve his or her full sentence. That anomaly needs to be addressed. However, there is nothing in the Sentences Act to suggest that soldiers or police officers would be ineligible for the two-year cap.
The final important argument to justify the statute of limitations proposals was that republican suspects had benefited from the on-the-run (OTR) letters that amounted to a de facto amnesty.
However, as the judge made clear in the failed Hyde Park bombing case against John Downey, and was confirmed categorically by Lady Hallett in her review of the OTR scheme, these letters did not amount to an amnesty.
The effect of OTR letters was to tell their recipients that there were no current charges or evidence against them. However, unlike an amnesty, they did not rule out a future prosecution if evidence emerged.
The reason why the shallowness of the arguments justifying a statute of limitations for State actors is important is that the principles of equality and non-discrimination work both ways.
In any case involving the post-conflict prosecution of a paramilitary suspect, even one not involving collusion, the fairness of the amnesty for State actors would inevitably be raised. Again, the legal effect of the amnesty for State actors could be to make any prosecutions extremely difficult.
Senior members of the Conservative Party appear to be aware of the difficulties associated with a statute of limitations 'for security forces only'.
In an interview with a Belfast newspaper after the Defence Committee report was released, its Tory chairman Dr Julian Lewis MP suggested that victims might have to be "big hearted" and accept that sacrificing the possibility of paramilitary prosecutions might be the price for protecting ex-military personnel from prosecution.
To my mind, that is a very big ask. Dr Lewis was at least being honest.
The alternative, of course, is to simply drop the talk of a statute of limitations and implement the SHA, that contains no such provisions.
Kieran McEvoy is Professor of Law and Transitional Justice at Queen's University Belfast