The Cadogan Group


The St Andrews Act



by Bill Smith

Any assessment of the St. Andrews' Agreement Act 2006 must take account of the 1998 Agreement which underpins it.


The 1998 Agreement represented a step forward in that it consolidated the move away from violence and called for the decommissioning of paramilitary weapons. It also

provided an outlet for nationalist aspirations through the NSMC, whileensuring that cross-border cooperation would not become creeping unification (since the NSMC would be accountable to the Assembly)

  • created incentives in the form of shared political power and officewhich have encouraged the extremes on both sides to moderate their positions
  • formalised the Irish government’s endorsement of the consent principle
  • created the British-Irish Council as a vehicle with some potential to strengthen east-west links

These gains persist and the two governments' proposals at St. Andrews do not detract from them.

But the 1998 Agreement also had flaws and vulnerabilities:

  • the provision that Northern Ireland would simply transfer to asovereign united Ireland if 50% + 1 of the electorate so voted is (a) based on an optimistic illusion; (b) unbalanced, in that the process is irreversible; (c) counter-productive to the extent that it stimulates unrealistic republican hopes and Unionist anxieties; (d) unsatisfactory, in that it leaves the issue that is central to the conflict unresolved
  • the arrangements for IRA decommissioning were insufficiently precise,failing in particular to link progress with the release of prisoners and British "demilitarisation"
  • it created arrangements for governance within Northern Ireland which seem equitable in theory but are unsustainable in practice, as experience both here and in other divided societies shows
  • it kicked the core issues of policing and justice into touch, allowingSinn Fein to gain the benefits of the rest of the package (including Ministerial office) without properly endorsing Northern Ireland's constitutional status or policing and criminal justice systems.

The first flaw remains, but the consent principle which it embodies is better for unionists than any alternative which they might reasonably expect to negotiate at present. Although based on an illusion, the provision for referenda at least reinforces the democratic legitimacy of the Union. The issue of what might happen if and when a majority votes for a united Ireland is unlikely to have to be addressed for the foreseeable future, by which time the context may have been transformed.

The second flaw, over decommissioning, was serious at the time. It enabled the republican movement to squeeze more concessions from the British government, split the Unionist Party, and undermined unionist support for the Agreement. But the damage is done and cannot now be reversed. The most that we can hope for is that the two governments and the Unionist negotiators will have learned the lessons of that unhappy experience.

The third and fourth flaws - over the devolved institutions and justice - are more serious and cry out for correction. The Strand One arrangements could have succeeded only if the republican leadership had unambiguously renounced violence and committed itself in good faith to making the new Northern Ireland work (which it has still not done). In practice, it delayed decommissioning both because it was difficult to sell to the grassroots and in order to secure further concessions. Thereby republicans let their potential partner, David Trimble, fall, when they could have pulled him back to safety. Unionists, not unreasonably, concluded that Sinn Fein would put their own party interests before the good of the cross-community partnership.

The republican movement continues to celebrate and justify its past use of violence; to deny the legitimacy of both democratic states in Ireland; and to reject their police forces. It remains committed to a strategy of combining political action with "armed struggle" in whatever combination seems most effective at the time in order to achieve its ultimate goal of a 32 county socialist republic. Sadly, the southern state and the republican electorate continue to celebrate the blood sacrifice of 1916 and all the misplaced lethal idealism that has accompanied it down the years.

In terms of practical outcomes, the Agreement has failed in that

(i) the Executive has proved unstable and unsustainable, as cross-national research on the political institutions of divided societies predicted it would;

(ii) the unionist electorate has rejected power-sharing with the present republican leadership; and

(iii) the republican electorate has effectively blocked progress towards accommodation.

Whence the terrible paradox: an Agreement hammered out over five years by some of the, supposedly, best political brains in Britain and Ireland with the finest intentions, has in practice resulted in a more polarised electorate, the dominance of deeply sectarian parties, and widespread popular disenchantment with politics.


Cross-national research shows that the following four essential criteria must be satisfied for power-sharing to succeed:

A commitment by the political leaders of the two sides to maintain and improve the political system. Here, the republican leadership campaigns actively to dissolve Northern Ireland's constitutional status, denies the legitimacy of the state, its policing and criminal justice systems, and refuses to participate in the national Parliament.


Cohesion within each community. The leaders of each community must be able to compromise with each other without losing their followers' support. But here we have a dual two-party system, characterised by intense rivalry between the two main parties on each side and fragmentation within them.


A tradition of cooperation between the leaders of the two communities.   Our tradition has been one of intense hostility. Moreover, we have a history of leaders who showed willingness to cooperate being undermined by their potential partners on the other side and driven from office by hardliners on their own side.


A common external threat.  Not only has there been no common external threat here, but each side fears the intervention of external powers, London or Dublin,  perceived as allied to the other side. Or more accurately, republicans have sought with hope and some success to use Dublin, London and Washington to impose change on Unionists, while Unionists have viewed all external interventions as dangerous.

It will clearly not be possible to satisfy these criteria fully as long as republicans aspire to achieve a united Ireland and unionists insist on preventing them. But, despite the present difficulties, it is possible to imagine new internal governance arrangements which at least encourage progress in the direction of mutual accommodation, and mitigate the risks of failure. Such arrangements would have to be very different from those proposed in 1998. Tinkering with the Agreement may meet the short-term needs of British and Irish Ministers anxious to show that they are still making the peace process work. But NIO officials privately recognise that, in the long run, the problem calls for a fundamental transformation in our political culture and structures of governance.

That transformation began in 1993, when the British and Irish governments agreed on the basic principles which would regulate it. The 1998 Agreement fleshed these out and brought most of the local parties on board, with the essential exception of the DUP. The 2006 Bill picks the process up where it left off in 2002, with the bonus that the republican movement has at last delivered on its promise of decommissioning, and the DUP has joined the game.

But it will take decades to achieve the stable just peace which the authors of the 1998 Agreement promised. In effect, the challenge is for both loyalism and republicanism - as ideologies incorporating a willingness to break the law and use violence to achieve their primary political objectives - are themselves decommissioned by their leading advocates and adherents. There can be no guarantee that the process will have positive outcomes: there is no inevitability about a lasting peace. But we who hanker for peace with justice and stability must do our utmost to generate it.


For power-sharing to succeed, the next necessary step has to be taken by the political leaders of republicanism. This is a clear commitment in word and deed to support and uphold the constitutional status of Northern Ireland and its institutions for as long as this is the wish of the majority. It means ending all formal connections, including overlapping memberships, with the IRA, unambiguously rejecting violence - past and future - as a means of pursuing political objectives, and supporting the current arrangements for policing and criminal justice.

This would not preclude developing proposals and campaigning for a united Ireland: indeed, it would be a most welcome development if they would seriously reach out to Unionists with ideas for an agreed Ireland which would accommodate Unionists' concerns and aspirations.

The purpose of these conditions is not to humiliate republicans or make them wrong – many, including Unionists, are prepared to forgive and move on from the mayhem of the past - but to earn the trust of the majority of the electorate, without which power-sharing simply can't succeed. Unionist voters will not support imaginative new departures while they are in survival mode.

In return, the DUP has already indicated its willingness, if republicans take this step, to give an equally clear and binding commitment to sharing power with them and to participate constructively in the NSMC.

In relation to the second criterion, community cohesion, anything which contributes to reducing fragmentation within parties, or rivalry between parties on the same side, can be expected to make leaderships more secure and hence more willing to co-operate across the community divide. If the Unionist Party continues to fragment and decline as it has over the past eight years, there is a real prospect of the DUP maintaining political supremacy in the Unionist community for the foreseeable future. This would reduce the risk to its leaders of negotiating deals with Sinn Fein.

But the dynamics of our electoral system - especially with the STV method of proportional representation in place for Assembly elections - mean that any such dominance is unlikely to last for long. Over a ten year period, the probability is that a new party would emerge to challenge the DUP's hegemony. It could be formed out of the remnants of the Unionist Party, possibly combined with Alliance: or it could comprise breakaway factions from the DUP opposed to any compromise with republicans. Paradoxically, a first past the post system would probably improve the prospects for a sustainable power-sharing administration by producing greater cohesion within the political leaderships on both sides, and raising the barrier against newly formed rivals.

On the third criterion - building a tradition of cooperation between the two leaderships - there has been some progress over the past thirteen years. In local government and on quangos, DUP and Sinn Fein councillors are doing business together in a variety of settings without fuss. The current reform of local government provides an opportunity to establish new structures and procedures which would require, encourage and protect cooperation on a voluntary basis. But to create a culture of cooperation, the dominant parties will have to demonstrate that they are willing and able to put accommodation before party advantage, even when this means taking serious risks with their internal cohesion.

Both to meet this criterion and to respect the basic principles of parliamentary democracy, the arrangements for forming the Executive need to be overhauled. The d'Hondt system shoehorns into government ministers from parties which are diametrically opposed on the main constitutional, economic and social issues of the day, thereby rendering effective decision-making impossible. The Executive thus formed is a gaggle of departmental heads rather than a coherent team united behind a common policy platform. D'Hondt fails to provide an effective opposition capable of commenting critically on the actions of the Executive and of offering an alternative to the incumbent administration: in theory, one or more of the parties might eschew the spoils of office to take on this role, but in the real world this is unlikely.

The rules should be changed to allow for a variety of possible voluntary cross-community coalitions, as in 1973. There are numerous ways in which this could be engineered: the essential democratic principles are that there should be participation by both Unionist and nationalist/republican political leaders in government and an effective opposition.

The safeguards for cross-community decision-making also need to be changed. The current requirement that MLAs identify themselves as Unionist, Nationalist or Other reinforces the salience of sectarian issues, notably partition. It generates a sense of separateness, disadvantages cross-community parties and works against the development of a shared Northern Ireland identity. As long as these designations persist, there will be no prospect of a new non-sectarian politics evolving, and parties will be encouraged to campaign on divisive platforms. Again, there is a long list of possible safeguards: the essential thing is that all major decisions (such as the selection of Executive members) command substantial (although not necessarily majority) support from both sides of the community.

On the fourth criterion, nationalists will continue to look to Dublin for support and to criticise London for supporting Unionists, while both governments will in fact be working together towards their common objective of reducing the bother this place gives them. In this context what matters is that:

(i) both governments speak with one voice on Northern Ireland issues;


(ii) neither the DUP at Westminster nor Sinn Fein in the Dáil - nor the electoral threat Sinn Fein poses to Fianna Fail - is permitted to influence their joint policies in sectarian directions. The risk here is of the next Dáil election campaign - due next year - stimulating assertive rhetoric from Sinn Fein, and a consequential shift in other parties' positions on the North to see off the challenge. This would lead to a negative reaction amongst Unionists, who are deeply concerned at the prospect of what a coalition in Dublin dependent on hardline republican support might induce London to attempt.


As spokesmen for the two largest political parties have pointed out, the only interests to have signed up for the St. Andrews "Agreement" are the two governments. Whatever reservations there may be amongst NIO officials, Peter Hain has declared it a success, and clearly believes that his best tactic at this stage is to keep talking it up and rush it through. When he bulldozed the Act through the Commons on 21 November, he actually introduced it by saying, "This Bill is potentially the most significant for generations": and concluded the debate, with what others might view as a cruel self-parody: "Anyone from Parliament, Northern Ireland or the rest of the UK watching our debate will be filled with bubbling enthusiasm for the prospect that the St. Andrews agreement will be implemented."

Our ebullient viceroy may even achieve a March election (which will polarise the electorate further). But for the longer term, the proposed election will do no more than tickle the surface of the issues that have to be tackled, as outlined above: by then of course it won't be his problem any more than it now is Willie Whitelaw's.

On policing and justice, the Act provides for Sinn Fein to join the District Policing Partnerships and to take a Pledge of Office in which they will undertake (without having actually to utter the dread words):

"… to uphold the rule of law based as it is on the fundamental principles of fairness, impartiality and democratic accountability, including support for policing and the courts as set out in paragraph 6 of the St Andrews Agreement",

which says

"We believe that the essential elements of support for law and order include endorsing fully the Police Service of Northern Ireland and the criminal justice system, actively encouraging everyone in the community to co-operate fully with the PSNI in tackling crime in all areas and actively supporting all the policing and criminal justice institutions, including the Policing Board.”

This wordy formulation allows for more wriggle-room than strict constructionists on the Unionist side will feel comfortable with: the only binding commitment is "to uphold the rule of law". The final paragraph can reasonably be interpreted as merely aspirational; a Sinn Fein Minister, having taken the Pledge with his fingers crossed behind his back could argue that "we believe etc…, but sadly we are not yet satisfied that the policing arrangements as they are at present justify our giving that support". What is wrong with the more straightforward:

"I pledge to uphold Northern Ireland's constitutional status and the consent principle; respect the rule of law; and actively support all Northern Ireland's policing and criminal justice institutions, including the Police Service and the  Policing Board"?

But the more fundamental problem is not with the wording of the Pledge, since many Unionists won't trust anything the Sinn Fein leaders say: what matters is what they do and are seen to do. At the very least this must mean joining the Policing Board and the District Policing Partnerships, and encouraging their constituents to work with the Police Service. The DUP insists that Sinn Fein must do these things before entering government, and this is not an unreasonable requirement. Indeed, for the process to work it must be done before the planned start of the election campaign on 30 January.

Perhaps the biggest weakness in the Act is that, while it is littered with deadlines for the DUP, there is no deadline for Sinn Fein to sign up for policing: the one step without which nothing else can happen. Peter Hain insists - makes it "crystal clear" - that all the other deadlines are absolutely essential and firm, but on this he says it would "not be helpful".  Sinn Fein says it can't make any commitment without a special party conference, which it can't organise until it's ready, so that's that: it's a matter for Sinn Fein.

It also says it can't sign up until policing is devolved, and the Act tries to meet SF on this by envisaging the possibility of devolution within a year of the new Assembly convening. However, since the DUP insists on the principle that the majority community as well as the minority must trust the Ministers responsible for law and order, that isn't going to happen. The longer the republican movement prevaricates the longer it will be before the necessary trust can be built. Even the SDLP recognises that it would be a mistake to try to devolve policing before the Executive has successfully established itself and earned the confidence of both traditions.

On human rights and equality, the British Government has committed itself to doing some things which it was already doing or should already have done. Language activists in their respective communities will welcome the proposed support for the Irish language and Ulster Scots heritage. The Act does not provide for an Irish Language Act as required in the St. Andrews document, since this is being carried forward separately on a fast track by the Department of Culture, Arts and Leisure. The St. Andrews Act does however require the Executive to adopt a "strategy" on the language. This and the Language Act should ensure that safeguards are in place to ensure that the policy of promoting Irish is not used to legitimate discrimination: the corresponding policy in the South if applied in the North would probably be deemed illegal on the ground of indirect discrimination against Protestants (who unlike those educated in Catholic schools generally don't have qualifications in Irish) applying for posts in the public sector.

The language commitments and the provision in the document to ensure access for EU nationals to posts in the Northern Ireland Civil Service should have been matched by a parallel undertaking from the Irish government to amend its legislation so as to open up its public services to all applicants on equal terms, whether or not they have Irish. The opportunity should have been taken also to end the effective discrimination which exists in Northern Ireland against Protestant teachers seeking employment in maintained schools.

Strand One Changes

On "practical changes to the operation of the institutions", as the document described them, the Act falls far short of the radical overhaul which is needed to create a sustainable devolved administration: DUP talk of "replacing the Belfast Agreement" is - like Peter Hain's "extraordinary" £50bn peace dividend - thick smoke and broken mirrors.

Before looking at the changes foreshadowed in the St. Andrews document, there is a provision in the Act which, depending on the outcome of the March election, could prove more important than anything else in it. The effect of clause 8 - covering the appointment of the First Minister - is that if the largest Unionist party is not also the largest political party overall, then the largest Nationalist party has the power to nominate a candidate to the post of First Minister.

What this means in practice is that if Sinn Fein wins more seats than the DUP, Sinn Fein will nominate the First Minister even if a majority of MLAs are designated as Unionist. In reality this is unlikely to happen: the Unionist Party would have to make a comeback and the SDLP vote to collapse. At first, DUP MEP Jim Allister condemned this as a "monstrous and a gross affront to democracy". But the DUP MPs did not try to amend it, and we must conclude that they intend to turn this "affront" to advantage by encouraging the Unionist electorate (a) to turn out and (b) not to jeopardize the DUP's overall supremacy by voting for the second largest Unionist party. Correspondingly, Sinn Fein will use the glittering prize of a Nationalist First Minister (and more importantly the rage this will create amongst Unionists) to hammer the SDLP.

The first of the changes actually signalled in the St. Andrews document is that the Act provides for a new Ministerial Code, intended to strengthen the accountability of individual Ministers to the Executive. But the weakness in Ministerial accountability flows from the d'Hondt system. Ministers in the previous Executive were subject to a Ministerial Code, and it did not prevent disputes arising over Ministerial accountability. There is no reason to suppose that putting the Code on a statutory footing will improve the coherence and efficiency of a cobbled-together Executive. Incidentally the Act provides for the Secretary of State to impose a Code if the parties can't agree on one, which is hardly a compelling display of confidence in the parties' ability to tackle difficult issues!

On a related point, the Act says that the Code should provide for cross-community voting by the Executive: but it does not explain what will happen if one or more of the Ministers designates himself as "Other". Presumably his vote just wouldn't count, as in the Assembly: so much for encouraging cross-community political activity! Ministers might like to reflect on the fact that, according to the 2005 Life and Times Survey, roughly one third of the electorate identify themselves as "Other". Even a designation which made the category less obviously residual would show willing: how about "Anti-sectarian"?

This is but one example of a tendency in Ministers' thinking to pre-empt the outcome of the election by assuming that the DUP and Sinn Fein will be the dominant parties: another is the requirement for these two parties to identify a First and deputy First Minister before the March election. Surely this tends to prejudice the outcome of the election, and as such should be illegal?

The second change is a provision for "Assembly referrals for Executive review". Again, this is intended to constrain maverick Ministers. It would allow the Assembly to refer important Ministerial decisions - such as the abolition of the 11+ - to the full Executive for a collective decision. But the Executive already had the power under the Northern Ireland Act 1998 to consider any major issue and any issue cutting across the responsibilities of two or more Ministers, and Ministers were already obliged under the old Code to act in accordance with all decisions of the Executive. So this cannot reasonably be interpreted as a significant change. Martin McGuinness was able to get away with his last minute decision on the 11+ only because the Executive did not have time to reverse it before collapsing. This provision would not add significantly to the Executive's power to constrain individual Ministers, or to the Assembly's power to hold the Executive accountable.

The third change is a series of amendments to the Pledge of Office, including those discussed earlier under "policing". I have argued that a stronger Pledge is needed in order to address the first criterion for successful powersharing. But what really matters is not what the Ministers promise, but what they do.

The fourth change is in the process of appointing the First and deputy First Ministers. The two largest parties would nominate their candidates for the two posts separately rather than jointly as before. The obvious consequence is that Ian Paisley would avoid the embarrassment of publicly nominating a former IRA commander as deputy First Minister: in practice, if DUP MLAs failed to vote for the person nominated by Sinn Fein, there would be no Executive.

The fifth change is that the Assembly would appoint a standing Institutional Review Committee to oversee the operation of the Strand One institutions; and the First Minister and deputy First Minister would appoint an Efficiency Review Panel to examine their efficiency and effectiveness. Whether these groups matter will depend on the reports they produce and the action taken as a result. Potentially, they could contribute to streamlining a bloated system: but if the parties are serious about efficiency, they could prove their credentials in March by agreeing to sacrifice some plum Ministerial appointments.

The sixth change is that the UK government would repeal the Northern Ireland Act 2000, which empowers it to suspend devolution. This is cosmetic, since the devolved administration would still exist by virtue of the 1998 Act, and Westminster would retain the power to amend that Act like any other. The current administration has shown an unhealthy willingness to produce, amend and suspend supposedly binding constitutional legislation at the drop of a hat.

The last Strand One change from the St. Andrews document worth mentioning is that an MLA would not be able to change community designation from "Unionist", "Nationalist" or "Other" during the term of the Assembly. This was included at the request of the DUP to prevent cross-community members from re-designating themselves as Unionists in order to vote with the Unionist Party against them (a device previously deployed to sustain David Trimble in office). It serves no general public interest.

Overall, these changes in Strand One represent modest and largely presentational gains for the DUP. They can credibly be presented as reworking the 1998 Agreement, but not as ditching it: the fundamental principles negotiated on behalf of Unionism by David Trimble are still in place.

Strands 2 and 3 Changes

A statutory right for a Minister to attend an NSMC/BIC meeting in which he had a leading interest, and a corresponding obligation on the First and deputy First Ministers to nominate both a replacement if the lead Minister is not going to attend, and another Minister to satisfy the requirement for cross-community representation. This meets a demand from Sinn Fein: it would prevent a Unionist First Minister from withholding a nomination, a tactic used by David Trimble. A modest gain for Nationalists.

A review of the efficiency and effectiveness of the existing implementation bodies and the case for additional bodies and new areas of cooperation. This offers something to both sides. In practice Sinn Fein and the SDLP are unlikely to agree to cut back any of the existing bodies or their expenditure, however useless the review finds them, without offsetting increases in cross-border activity elsewhere. But there is still scope for developing new practical forms of cooperation to mutual advantage, so this could result in efficiency improvements. On balance, a potential gain for Nationalists.

Assembly scrutiny of implementation bodies. The Assembly already had the power to require the Chairs and Chief Executives of the bodies to appear before its committees, so there does not appear to be any substantive gain here. A presentational gain for the DUP.

North/South Parliamentary Forum and Independent Consultative Forum. To be "encouraged" and "supported" by the Executive. These are hardly major threats to the Union. Depending on how they are designed and managed, they could fail to take off, degenerate into talking shops, or actually help promote good neighbourly relations. Modest gains for Nationalists.

Secretariat for the British Irish Council. To be welcomed as a move towards equality with the NSMC, although still subject to the agreement of BIC members: but even with this the BIC will still have less dedicated capacity than the NSMC Secretariat and North/South implementation bodies. Modest gain for Unionists.

East-West Inter-parliamentary Framework.  To be welcomed as another means to strengthen political relationships across these islands. Modest gain for Unionists.


The St. Andrews Act is a hasty piece of legislation, and it is barely conceivable that the professional draftsmen who produced it are content with their handiwork. Conservative spokesman Lord Smith has described it as "…a wretched Bill…a patchwork of cobbled-together partisan clamourings with a touch of half-baked NIO ingenuity".

It tinkers with the 1998 Act, and falls well short of the fundamental transformation which a long term strategy for tackling the problem requires. It does almost nothing to progress the criteria for successful powersharing which we have identified.

What it does do is provide a series of rather tiny and transparent fig leaves to cover the Secretary of State's embarrassment at his failure to get the parties actually to agree, and to allow the leaderships of the DUP and Sinn Fein to go into government together if they decide they want to, and can sell it to their parties.

The main gain from the Act will come if Sinn Fein signs up for policing - which is still far short of a commitment to upholding Northern Ireland's constitutional status. The process of building trust can then begin. But it will take far longer than the government seems to imagine before the DUP and its supporters will trust former IRA members with custody of the Police Service. The forthcoming election campaign - if it happens - might leave the party leaders more secure; but equally, it might have the opposite effect, if hardline factions become too disaffected.

Trivial though its substance is (relative to the 1998 Act), the 2006 Act might just trigger a process of internal consultations and inter-party negotiations which results in the formation of an Executive by next March. But even if it scores that limited short term goal, a coalition dominated by Sinn Fein and the DUP would be volatile and unlikely to survive in office for a full four year term, if only because of the compromises each party would have to make in discharging the responsibilities of office.

Imagine the concessions required to tackle effectively the thorny issues associated with the devolution of policing and criminal justice, or to take the community relations agenda forward towards the British government's vision of a shared future, in which sectarian differences are steadily reconciled and a common identity forged. Then visualise the strains and conflicts that will occur within the two parties and their constituencies if their leaders offer such concessions, and the opportunities for their rivals to exploit these for electoral advantage.

The best we could reasonably expect would be a division of powers within the Executive, with the DUP and Sinn Fein each running their Departments almost autonomously within their agreed budgets. The thorniest issues which most need to be tackled courageously would be deferred or neglected. They would undoubtedly make it their first priority to finish off the work of the March election (if necessary) by finishing off their rivals forever. The end result could be a regime effectively dominated by the two most uncompromising parties on each side, with the smaller parties (like the residual Others) soon ceasing to exert effective influence, even as a critical opposition.

Whether it succeeds or fails in creating an Executive by March 2007, the Act does not bode well for the future of liberal and progressive politics, or for the influence in government of the 35% of the population who describe themselves as neither Unionist nor Nationalist.

What it offers is not so much powersharing between the two communities, as dividing the spoils of office between the leaders of the two dominant parties.