Full text of DOI:10.1093/pa/53.3.577

Designing the Northern Ireland Assembly
BY RICK WILFORD
In designing the new constitutional settlement for Northern Ireland, the
signatories to the Belfast Agreement of 1998 were conscious that they
were treading a path littered with failure. Since the introduction of
direct rule in 1972, successive governments had attempted to establish
a power-sharing administration in the province. The first attempt, in
1974, based upon the Sunningdale Agreement reached the previous
year, ended in collapse after five months. In 1975–76, a Constitutional
Convention was created to generate a scheme for self-government: it
too failed, as did both the ‘Atkins initiative’ of 1979, named after the
then Secretary of State, Humphrey Atkins, and the scheme for ‘rolling
devolution’ devised by his successor, Jim Prior.1
The renewed attempt to generate a pact among Northern Ireland’s
parties that would restore devolved institutions was a protracted affair.
Among other things, it required a peace process that was robust and
durable, as well as a certain political suppleness among political leaders,
whether unionist, loyalist, nationalist, or republican. It also needed the
governments in London and Dublin to act in a cooperative, bilateral
fashion, as well as an acceptance by the British government of a role
for external, primarily American, involvement in the conciliation of
Northern Ireland’s rival ethno-national communities. Indeed, parity of
esteem between national communities was an essential ingredient of the
political process. That is, mutual recognition had to be accorded by the
respective communities to the legitimacy of pursuing both Irish unifica-
tion and the maintenance of the Union with Great Britain—provided
each goal was sought through the use of exclusively democratic and
peaceful means. In effect, it meant giving the people of Northern Ireland
the means of expressing their consent about its constitutional future,
and that this principle be accepted by both the Irish and the British
governments as well as the parties within Northern Ireland.
The sequence of events that led to the Agreement on 10 April 1998
was far from uninterrupted. Moreover, until almost the last moment
there was genuine uncertainty whether the negotiations, which had
begun in 1996, would succeed. That they did—notwithstanding the
subsequent difficulties of implementing the Agreement and the later
suspension of devolution in February 2000—is testimony both to the
intricacy of the Agreement’s design and to the stamina and intelligence
of its signatories. It also represents the durability of consociational
thinking among key players, in Belfast, London or Dublin.
Ą Hansard Society for Parliamentary Government 2000
Parliamentary Affairs (2000), 53, 577–590

578
Parliamentary Affairs
To understand the design of the new legislature, it is necessary to
explore the provenance of the Agreement itself. One way of doing this
is to see British policy since the initial imposition of direct rule as an
exercise in variable geometry. There have been two constant elements
in the design of policy, the more obvious of which has been a commit-
ment by successive British governments to devolve power to Northern
Ireland.2 The other has been the ide´e fixe of consociationalism, seen as
an off-the-peg model of how divided societies are to be governed and
derived from the earlier work of Arend Lijphart.3 The third element has
been rather more variable, although since the Anglo-Irish Agreement of
1985 it has become a cornerstone of the design, namely the so-called
‘Irish dimension’. The 1985 Agreement was not simply a means of
governing Northern Ireland on something like a bilateral basis (albeit
that Dublin’s role was, to borrow Garrett Fitzgerald’s phrase, less than
executive but more than consultative). It also required unionism to
accept the Republic’s involvement as a sine qua non of any grand
political design: a prospect that unreconstructed loyalists and die-hard
unionists still refuse to countenance. Thus, although the Irish dimension
waxed and waned between 1972 and 1985, it is an essential part of the
1998 Agreement, albeit in a less muscular form than that expressed in
the Framework Documents of 19954—and certainly in a less developed
mode than that preferred by Sinn Fe´in. These, then, are the three
elements of policy geometry that have converged in the Agreement.
However, the confederal aspects of the Agreement render it best under-
stood as an example of ‘consociationalism plus’.5 It is its external
dimensions, notably in relation to north-south matters, that complicate
the legislative role of the new Assembly.
Consociationalism
In reflecting on the values informing the 1998 Agreement, it is difficult
to escape the conclusion that the inertia of consociationalism had
prevailed among its signatories. Certainly, the defining features of the
consociational model, first elaborated in the Sunningdale Agreement of
1973—which established a local Assembly, a power-sharing grand
coalition government and a prospective north-south institution the
Council of Ireland—are apparent in the1998 Agreement. While
recourse to the model is necessary, however, it is insufficient to capture
its complexity, not least because of its integral confederal character and
a qualified departure from one of its defining features, the unanimity
rule or mutual veto.
Though it may be criticised for its rather contrived character, con-
sociationalism does embody the politics of accommodation. Its four
key characteristics—cross-community power-sharing, the proportion-
ality rule, segmental autonomy and the mutual veto—are designed to
effect governance within a divided polity and a society where majority
rule is untenable. Each of the model’s features is present in the new

Designing the Northern Ireland Assembly
579
Agreement, although majoritarianism is provided for in relation to
‘non-key’ decisions within the Assembly, both as a plenary body
and in its committees. Lijphart did not intend consociations to be
ends in themselves, but rather that they could provide a means of
moving towards a more ‘normal’ mode of competitive politics in
the medium to longer term. In the shorter run, a heavy premium is
placed on mutual trust and confidence, initially among the relevant
elites, which—ceteris paribus—descend to envelop contending com-
munities. In this respect, one may depict consociationalism as trickle-
down politics. This interpretation does, however, beg at least one
question: whether political leaders can deliver their followers, as Brian
Faulkner and pro-Sunningdale unionists discovered to their cost in
1974. Faulkner’s readiness to implement the Sunningdale Agreement
led to his loss of the leadership of the Ulster Unionist Party, the
fragmentation of unionism and a near insurrection in the form of the
Ulster Workers Council strike in May 1974. The strike, arguably
called more against the proposed Council of Ireland—which never
met—than against the principle of power-sharing, paralysed the prov-
ince and spelled the end for both the coalition government and Faulk-
ner’s political career.
Despite that earlier failure, and intervening attempts to engineer
social transformation in Northern Ireland by means of what has been
styled ‘inverted consociationalism’,6 the consociational template did
structure the logic of choice in designing the new legislature and its
executive arm. Part of the attraction of consociational democracy is
that its rules are rational in that they seek to eliminate uncertainties,
anxieties and threats among contending elites by building in decision-
rules and procedures which eliminate recourse to simple majoritarian-
ism. Yet, this is true only up to a point in the case of the decision-rules
proposed for the Assembly.
Although a comparison with Sunningdale is a beguiling attraction,
the 1998 Agreement is a subtler political bargain than its predecessor.
Besides the massively changed context created by the IRA’s initial (and
later breached) ‘cessation of military activities’ at the end of August
1994 and the subsequent cease-fires by the major loyalist paramilitary
organisations, its political inclusiveness, it is rendered unique by the
mandatory provision for north-south institutions and the creation of
the British-Irish Council. The commitment to an equality culture, a new
regime of human rights, reform of the criminal justice system as well as
of the RUC, and the accelerated prisoner release scheme, all further
compound the singularity of the Agreement reached on 10 April 1998.
Its endorsement on both sides of the border by referendums also gave it
the popular support that had never existed for the Sunningdale package.
Recall, too, that the referendum in the Republic of Ireland in May 1998
legitimised the proposed changes to Articles 2 and 3 of its 1937
Constitution. These replaced the irredentist claim to the ‘six counties’

580
Parliamentary Affairs
with an aspiration to unite the people of Ireland, a change that the Irish
government had refused to implement in 1973 during the negotiations
on the Sunningdale Agreement.
If one considers the irreconcilable goals of undiluted unionism and
atavistic republicanism, the fact that there was an Agreement in 1998
shows that much was ceded by each side in the negotiations. In
particular, Sinn Fe´in’s position altered significantly during the talks. It
began by opposing: a local, effectively partitionist, Assembly; the
proposition that north-south bodies would be accountable to such an
Assembly; and any change to the Irish Constitution that involved
removing the formal territorial claim to Northern Ireland. It relin-
quished each of these negotiating planks. Correspondingly, a majority
of Ulster Unionists accepted the principle of power-sharing with recon-
structed (that is, decommissioned) republicanism; the integral nature of
institutionalised cross-border bodies with executive powers; the review
of the RUC; an accelerated prisoner-release scheme; and the contingency
of Northern Ireland’s place within the UK (the principle of consent
having become the constitutional imperative on both sides of the
border).
The bargained nature of the Agreement is clear from the interweaving
of each of its three institutional strands: within Northern Ireland,
between north and south, and between east and west. Its designers
clearly understood that if any one of the strands was to unravel, the
wider fabric would fall apart. Of course, the bargain was much more
than an exercise in institution building. At root, it was intended to build
mutual confidence between rival communities, hitherto in short supply.
In particular, the premium placed by pro-Agreement unionists on the
decommissioning issue indicated that the trust would have to be hard
won by the republican movement. In the event, it was the failure by the
IRA to embark on the decommissioning process that led to the suspen-
sion of devolution in February 2000, an option the Secretary of State
deemed preferable to the resignation of David Trimble and his three
cabinet colleagues.
The Assembly
The constraints of a divided society are manifest in a number of key
features of the Assembly’s consociational design. For example, each of
the 108 Members of the Legislative Assembly (MLAs) elected by Single
Transferable Vote proportional representation, must designate him/
herself either as a ‘Nationalist’, a ‘Unionist’ or ‘Other’.7 Suchself-
ascription is required to allow application of the tests of
cross-community support in what the Agreement describes as ‘key’
decisions. These, including the election of the Assembly’s Presiding
Officer (Speaker), the determination of its standing orders, the planned
‘programme for government’ and the budgetary procedures, are subject
to two alternative special voting procedures, each designed to realise

Designing the Northern Ireland Assembly
581
the principle of bicommunalism: ‘parallel consent’ and ‘weighted major-
ity’.8 The Agreement also provides for a ‘Petition of Concern’ through
which 30 MLAs can designate an issue as a key decision, thereby
triggering the cross-community voting procedures.
The special voting procedures, another reflection of consociational-
ism’s characteristic proportionality rule, were first employed for the
twinned election of David Trimble and Seamus Mallon as First and
Deputy First Minister Designate at the inaugural meeting of the shadow
Assembly on 1 July 1998.9 Their joint election epitomised the power-
sharing character of the Agreement: they were intended to be co-equals,
enjoying a shared status that exemplified accommodatory politics.Until
December 1999, when two junior ministers were nominated jointly by
Trimble and Mallon, the First and Deputy First Ministers were the only
members of the Executive Committee subject to a vote in the Assembly.
The remaining ten were nominated by the relevant (four) parties (as
were their preferred departments) through the application of the
d’Hondt rule.10 They did not require the endorsement of the Assembly,
although they could be removed by a resolution proposed by at least 30
MLAs and obtaining cross-community support. In such circumstances,
the minister (or indeed any party that loses the Assembly’s confidence)
would be excluded from office for not less than twelve months. This
device provides a potential check on ministerial autonomy, offsetting
the possibility that a department could be turned into a party fiefdom
and, as with ‘key’ decisions, underlines the model of partnership that is
intended to infuse the new institutions. A bloc vote by unionist MLAs
(the only ‘grouping’ capable of mustering a simple majority) would be
insufficient to exclude a minister. Should an incumbent be excluded,
his/her successor would be nominated from within the ranks of the
same party. If, however, a party loses the confidence of the Assembly,
then d’Hondt is reapplied: the Executive seats would be reallocated to
that party with the next highest number of MLAs or, in the event of a
tie, the party with the higher number of first preference votes cast at the
Assembly election of 25 June 1998
Assembly committees
The legislative competence of the Assembly is consistent with prior
attempts to devolve power to Northern Ireland. The 1998 Act provides
for its authority to be primarily in ‘transferred’ matters, that is those
functions administered by the six Northern Ireland Departments that
had evolved under the direct rule regime (as re-engineered by the inter-
party agreement of December 1998). The Assembly is not designed to
be a mere legislative cipher or rubber-stamp for the Executive Commit-
tee. In addition to its more conventional legislative and scrutinising
roles, ‘statutory committees’ (a term deliberately chosen to convey the
various roles: scrutiny, policy advice, and part of the legislative process)
enjoy considerable authority. These committees are charged to ‘advise

582
Parliamentary Affairs
and assist’ each of the departments with which they are associated ‘in
the formulation of policy’: in addition, they may initiate primary
legislation. Legislative authority will not, then, be vested solely in the
proposed Executive Committee but shared by the committees, whose
chairs and deputy chairs are also appointed to reflect the Assembly’s
composition. Given ten departments, a total of 20 chairs and deputy
chairs were nominated by the parties (ministers are excluded). In
making their nominations, they were constrained by the principle that
they should ‘prefer committees in which they do not have a party
interest’: this prevented nominations drawn from the same party as the
relevant minister—a further means of realising the power-sharing
principle.
However, no provision was made in either the Agreement or the
Northern Ireland Act 1998 for the Office of the First and Deputy First
Ministers to be subject to a statutory committee. This issue proved
highly contentious when the draft standing orders were debated in
March 1999, spilled over into the short period of devolution, and was
not fully resolved at the time of suspension. During its shadow phase,
the Assembly’s committee on standing orders (itself jointly chaired by a
unionist and a nationalist) had proposed the creation of a ‘Committee
on Conformity with Equality Requirements’ that would test any policy
or legislative proposal against the statutory obligation to promote
equality of opportunity. Given that an ‘Equality Unit’11 constitutes one
element of the Office of the First and Deputy First Ministers (subse-
quently called the Office of the Centre), this meant that a close working
relationship would develop between the Unit and the proposed
Committee.
The simmering dissatisfaction within the Assembly about the limited
accountability of Trimble and Mallon’s Office afforded by the above
Committee boiled over shortly after the devolution of powers. On 6
December, the standing orders committee revisited the matter and,
instead proposed two new standing committees to oversee much of the
work of the Office: the Equality, Human Rights and Community
Relations Committee and the European Affairs Committee. The motion
for the two cross-cutting committees, akin to those in the Welsh
Assembly, was passed, and a week later a draft order was tabled for a
third committee to examine and report on the Office’s remaining
functions, thereby making it fully accountable to the Assembly. How-
ever, the First and Deputy First Ministers tabled an amendment, seeking
to revoke the two committees agreed on 6 December and replace them
with a single Committee of the Centre to report on a limited range of
functions carried out by their Office. Their intention was to insulate the
external/confederal aspects of their remit from the focused scrutiny of a
standing committee. As Mallon said: ‘It is essential that discussions
which take place in determining the decisions reached by the Executive
Committee, or the negotiating position for the Northern Ireland admin-

Designing the Northern Ireland Assembly
583
istration in relation to the North-South Ministerial Council or the
British-Irish Council, should remain private.’
The Trimble/Mallon view was that such matters were more properly
dealt with on the floor of the Assembly than in a committee room, but
it drew opposition from a wide range of parties, including Sinn Fe´in
and the Democratic Unionist Party (DUP), appalled at the limited
scrutiny it afforded. Nevertheless, the combined weight of Ulster Union-
ist Party (UUP) and SDLP members ensured that the wish of the First
and Deputy First Ministers prevailed. Thus, there was only one commit-
tee, and that with a limited remit. Among its initial tasks was a role in
the ‘mainstreaming’ of equality of opportunity required by the Agree-
ment. It would scrutinise legislative proposals, whether referred by the
Executive Committee or the statutory committees (voting on a cross-
community basis) to test its consistency with statutory equality require-
ments. Equality is further safeguarded by the requirement to seek the
view of independent Northern Ireland Human Rights Commission, also
created under the terms of the Agreement, on whether legislative
proposals are compatible with the European Convention on Human
Rights or any future Northern Ireland Bill of Rights, also flagged by the
Agreement. And, of course, there is further protection through the
courts, which can overrule Assembly legislation on grounds of
inconsistency.
The Committee’s oversight of proposed legislation, whether it ema-
nates from the Executive Committee, a committee or an individual
member, meant that it would work in harness with the Equality Unit
mentioned above. The latter’s responsibility includes ensuring that in
the formulation and review of policies and in their delivery departments
and other public bodies will comply with ‘Policy Appraisal for Fair
Treatment’, first implemented as a set of administrative guidelines in
1994 and made a statutory duty by the Northern Ireland Act 1998.
Sifting policies and legislation through this net is further proof of the
intention to mainstream equality of opportunity.12
With the exception of a decision to refer matters to the Committee of
the Centre, other matters before the statutory committees will be
decided by simple majority (no single party having a majority). How-
ever, where a committee is considering a key decision, any recommenda-
tions it makes, and on whatever basis, will be subject to the Assembly’s
special voting procedures. All non-key decisions will be resolved in the
Assembly on the basis of a simple majority vote, although MLAs,
through a ‘Petition of Concern’, could designate the matter a key
decision, thus activating the cross-community voting rules.
The ordinary memberships of the statutory committees were formally
allocated by the Assembly’s Business Committee (chaired by the
Speaker), although effectively by party whips on a proportional basis.
Each has eleven members, including the chair and deputy chair. Propor-
tionality meant that the opportunity to exercise executive patronage

584
Parliamentary Affairs
was narrowed, though not entirely foreclosed. The First and Deputy
First Ministers were able to nominate junior ministers to the depart-
ments, including their own, but no formula was specified in either the
Agreement or the Northern Ireland Act 1998 for the nomination
procedure, unlike the application of the d’Hondt rule for ministers and
committee chairs and deputy chairs. This, too, caused a row in the
Assembly. Section 19 of the Act provides that the First and Deputy First
Ministers may at any time determine that a number of members should
be appointed as junior ministers and the functions they should exercise.
In December 1999, Trimble and Mallon tabled a joint determination
for the appointed of two such ministers to their Office, subject to a
simple majority vote in the Assembly. However, the party identities of
the two nominees—UUP and SDLP—angered the other parties which
spoke of ‘jobs for the boys’ and a ‘closed shop’ for the UUP and the
SDLP. Trimble and Mallon argued that the junior ministers had to
command their confidence—which would have been sorely tried had
they nominated members of the DUP and Sinn Fe´in. The nominations
were eventually endorsed by 49 votes to 38, the UUP and SDLP voting
en bloc.
Mutual veto/unanimity
The Agreement provides for the unanimity rule, or mutual veto, char-
acteristic of consociational design, in relation to the procedures of the
Executive Committee. A four-party voluntary coalition (UUP, DUP,
SDLP and Sinn Fe´in) seems an unlikely setting within which the
convention of collective responsibility could take root. Given the DUP’s
antipathy towards the Agreement in general and, in particular, to Sinn
Fe´in’s participation in government, there was no prospect that full
meetings of the Executive Committee would take place, nor did they
during the ten weeks of devolution. This has raised some intriguing
questions about its intended operating procedures, although there is
clear guidance in the Agreement/Act, especially in relation to key
decisions and the operations of the North-South Ministerial Council.
For instance, the Executive is charged by the Agreement to ‘provide
a forum for the discussion of, and agreement on, issues which cut across
the responsibilities of two or more ministers, for prioritising executive
and legislative proposals and for recommending a common position
where necessary’ (theexample cited in relation to the latter is external
relationships). In addition, it ‘will seek to agree each year, and review
as necessary, a programme incorporating an agreed budget linked to
policies and programmes, subject to approval by the Assembly, after
scrutiny in Assembly Committees, on a cross-community basis’. The
Agreement also states that ‘Ministers will have full executive authority
in their respective areas of responsibility, within any broad programme
agreed by the Executive Committee and endorsed by the Assembly as a
whole’. All ministers, as a condition of their appointment, must also

Designing the Northern Ireland Assembly
585
affirm the terms of the ‘Pledge of Office’ set out in the Agreement. This
requires them to ‘participate with colleagues in the preparation of a
programme for government’; to ‘operate within that ‘when agreed
within the Executive Committee and endorsed by the Assembly’; and
‘to support, and to act in accordance with, all decisions of the Executive
Committee and the Assembly’.
The Pledge reinforces the interlocking relationship between the Exec-
utive and the Assembly. Moreover, and this seems paradoxical given
the inclusion of emphatically anti-Agreement parties in the Assembly,
the lack of a formal (let alone loyal!) opposition underlines the unique-
ness of the Northern Ireland case. In particular, the refusal of the DUP’s
ministers to sit alongside their Sinn Fe´in counterparts in the Executive
Committee, unless and until there was full and verifiable decommission-
ing by the IRA, threatened to create gridlock rather than to facilitate
such government. (However, the First and Deputy First Ministers’
meetings with the DUP’s two ministers independently of those with the
other eight ministers did appear to work relatively smoothly during the
brief (first?) phase of devolution.)
Programme for government
During its shadow phase between July 1998 and early December 1999
little attempt was made by the then prospective Executive to discuss, let
alone agree, a programme for government. Apart from an initial
exploratory meeting in January 1999, involving the party’s policy
advisers at which civil service briefing papers were tabled, and a follow-
up ‘brainstorming session’ later that month, no substantive progress
was made towards a common programme.13 Habituated to opposing
the Northern Ireland Office and one another, both within and across
the political divide, the process of agreeing a programme for govern-
ment confronted the parties with a steep learning curve. During the
shadow phase there was a ‘Transitional Programme’ prepared by the
Northern Ireland Office for MLAs, a digest of information on the policy
agendas of the departments. The parties were also faced with a raft of
policy proposals from the team of direct-rule ministers, including a
regional economic strategy, a regional planning strategy and reform of
the health sector.
The prospective Executive also inherited the government’s expendi-
ture plans for Northern Ireland for 1999–2000 to 2001–02, announced
in December 1998. This followed the Comprehensive Spending Review
initiated by the government in 1997 with its additional £1.4bn for
Northern Ireland. During the early stages of the Review, the then
Secretary of State, Mo Mowlam, consulted the parties, district councils,
trades unions, industry and commerce, the voluntary and community
sectors, on which expenditure programmes should be given priority.
After the Review, she decided on a further round of consultation with
Trimble and Mallon and with the shadow Assembly. This was an

586
Parliamentary Affairs
induction into the realities of governing for the parties, although the
government recognised that the expenditure plans could/would change
subsequently. Work on the programme was continued by the First and
Deputy First Ministers Designate. In reporting on this to the shadow
Assembly in mid-January 1999, they set out their shared agenda for the
programme, albeit in understandably broad terms. Their priorities were:
‘to address the needs of the most vulnerable and disadvantaged; to
imbue the community with a sense of enterprise and self-reliance; to
tackle educational disadvantage; and to put behind us the tragic years
of trauma and separation’. In words reminiscent of New Labour, they
continued: ‘We want to agree upon and implement a programme for
government that will succeed in delivering efficient, accountable and
transparent government and enable us to achieve economic growth and
development, the benefits of which will be shared throughout the entire
community.’14 However, the deepening impasse over decommissioning,
fraying relationships among the pro-Agreement parties and the unflinch-
ing posture of the DUP towards the Agreement effectively stymied the
prospect of an agreed, budgeted programme for the new departments.
Shortly after the conclusion of the Mitchell review of the implementa-
tion of the Agreement in mid-November, and in the immediate wake of
the transfer of powers on 2 December 1999, work on the programme
for government was renewed in earnest. However, by this stage the
Executive had no alternative to adoption of the spending plans
bequeathed by the Northern Ireland Office.
North-South relationships and the mutual veto
The difficulties involved in devising a programme for government are
compounded by Strand Two of the Agreement, which seeks to institu-
tionalise the relationship between Belfast and Dublin. The negotiations
on this north-south (confederal) dimension was a protracted (and at
times acid) affair. Its outcome was six new cross-border implementation
bodies and a further six matters for cooperation through existing bodies
in each country.15 Though the detail of structures and functions of these
bodies is significant, of interest here is their standard operating pro-
cedures and their consistency with the consociational template. The
first, perhaps paradoxical, point to note is the seeming inconsistency
with that template. The Pledge of Office in the Agreement requires
ministers ‘to discharge in good faith all the duties of office’. Given the
indispensability of the North-South Ministerial Council, one might
conclude that participation in it by local ministers was not an option.
Indeed, the Agreement stipulated that such participation is an ‘essential
responsibility’ for them. However, the same paragraph also states: ‘If a
holder of a relevant post will not participate normally in the Council,
the Taoiseach in the case of the Irish Government, and the First and
Deputy First Minister in the case of the Northern Ireland Administra-
tion, to be able to make alternative arrangements.’ This enabled the

Designing the Northern Ireland Assembly
587
DUP’s members of the Executive Committee to boycott the Council
without, on the face of it, breaking the Pledge of Office.
Though politic, even expedient, the provision sits uneasily with the
intention to facilitate inclusive government that is a hallmark of conso-
ciationalism. It meant that only three of the four parties in the Executive
would participate in the Council, allowing, potentially, the DUP to
frustrate at least some of its business because its agenda must be agreed,
as must all decisions, by Northern Ireland ministers and their counter-
parts in the Republic. Moreover, any Council decisions that are ‘beyond
the defined authority of those attending’ must achieve the consent of
both the Assembly (on a cross-community basis) and of the Irish
parliament. There is, in effect, provision for a reciprocal veto on each
side of the border, although it seemed clear that the North-South
Ministerial Council would enjoy a measure of autonomy, provided its
participants could agree on the adoption of common policies and their
implementation.
The East-West dimension
The North-South Ministerial Council is one aspect of the Agreement’s
confederal character. The other is contained in Strand Three which deals
with the British-Irish Council and the British-Irish Intergovernmental
Conference. The Intergovernmental Conference (which subsumes the
Anglo-Irish Intergovernmental Council and Intergovernmental Confer-
ence established by the 1985 Anglo-Irish Agreement) has attractions for
both the unionist and nationalist communities. For the latter, it under-
pins the binationalism of the Agreement by providing for the continuing
involvement of the Irish Republic in the internal affairs of Northern
Ireland. Equally, unionist anxieties that a new Anglo-Irish agreement
might evolve into some form of joint authority were assuaged, since the
Conference did not entail a derogation of sovereignty by the British
government.
The remit of the Intergovernmental Conference is to foster ‘bilateral
cooperation at all levels on all matters of mutual interest within the
competence of both governments’, essentially those that are not to be
devolved in the first instance. These include prisons, policing and
criminal justice, issues upon which the Republic’s Minister for Foreign
Affairs, as ‘co-chair’ alongside the Secretary of State, ‘may put forward
views and proposals’, given the Irish government’s ‘special interest in
Northern Ireland’. Both governments will ‘intensify cooperation on the
all-island or cross-border aspects of these (non-devolved) matters’ and
any decisions it makes will be on an agreed basis. Indeed, the Agreement
commits them to ‘make determined efforts to resolve disagreements’.
Unanimity is clearly the rule for participants.
This hefty shove towards Anglo-Irish cooperation, welcomed by the
nationalist and republican communities, is much more acceptable to
unionists than the Inter-Governmental Conference established by the

588
Parliamentary Affairs
1985 Anglo-Irish Agreement. There are three reasons for this. Its
meetings will be attended by relevant members of the Executive Com-
mittee; they will also be able to participate in the triennial reviews of
the 1998 Agreement conducted under the aegis of the Conference; and
fit will be unable to override the democratic institutions created by the
Agreement. It was the accessibility to and relative transparency of this
aspect of the Agreement that made it more palatable for pro-Agreement
unionists. Moreover, the prospect that policing may be devolved to
Northern Ireland—an objective endorsed by the Patten report—will be
generally welcomed by MLAs, although their reasons for so doing may
well differ.
The British-Irish Council is the final confederal aspect of the Agree-
ment and one that was of particular interest to unionist negotiators. It
brought together representatives of the British and Irish Governments,
the Scottish Parliament, the Welsh and Northern Ireland Assemblies,
representatives of the Channel Islands and the Isle of Man, thereby
realising a broadened, all-islands scope to the Agreement. Within this
institutional framework, the representatives were to discuss, consult
and exchange information on matters of mutual interest. However, it
was intended as more than a rather grand talking-shop. Members, on a
bilateral or multilateral basis and subject to approval by their own
elected institutions, could agree common policies and their implemen-
tation in fields including transport, agriculture, health, education and
culture. In this way, the Council supplied potential substance to the
‘totality of relationships’ first essayed during the Thatcher-Haughey
summits of the early 1980s. Although it met for the first time in early
December 1999, as did the North-South Ministerial Council, the Brit-
ish-Irish Council appeared to be the least developed of the Agreement’s
institutions. Of undeniable attraction to the pro-Agreement unionists,
who saw it as a vehicle for better integrating cooperation with other
elected representatives, especially in Britain, the suspension of devolu-
tion halted its operation. If direct rule again becomes more or less
permanent, it is unlikely to survive, while the North-South Ministerial
Council will wither on the vine.
Conclusion
Though the trajectory of its underlying values can be traced back to
1973, none can deny the novelty and subtle complexity of the 1998
Agreement. Bolstered by a new regime of human rights and a commit-
ment to a culture of equal opportunity, it has something for (nearly)
everyone. Steeped in a pluralist, inclusive philosophy, it was an imagi-
native attempt to move from zero-sum to positive-sum politics. For its
proponents, whether nationalist, republican, loyalist, unionist or other,
it proved a difficult bargain: as David Trimble put it in commending
support for the Agreement at the referendum in May 1998, ‘It’s as good
as it gets’.

Designing the Northern Ireland Assembly
589
How ‘good’ it can become remains largely untested, since the ten-
week phase of devolution, from 2 December 1999 to 11 February 2000,
was tantalisingly brief. Although too short a period to agree and unveil
the programme for government, the new ministers had taken to their
respective briefs with enthusiasm, while the Assembly’s committees had
prepared their agendas for scrutinising the departments. In addition, the
Assembly’s initial legislative programme was announced at the end of
January by David Trimble, albeit that it was relatively uncontroversial
and designed to match parity with relevant British legislation.16 During
the devolved period the Assembly had also nominated the chairs and
deputy chairs of its committees by proportional representation and had
elected three Deputy Speakers by means of cross-community voting.17
The Assembly’s longer shadow phase that had begun on 1 July 1998
had also demonstrated the preparedness of members from all parties to
work together, as they cooperated in a variety of shadow committees
including the Committee to Advise the Presiding Officer, the Committee
on Standing Orders and the Shadow Assembly Commission. Two ad
hoc committees were also established during this period, one to examine
the procedural consequences of devolution, the other to examine the
future of the port of Belfast. Though relatively modest, such instances
indicated that in some areas the Assembly’s members were able to
engage with one another on a cooperative basis.
Controversy, however, was never far from the surface. During the
devolved phase, the Assembly rejected the recommendations of the
Patten report for the reform of the Royal Ulster Constabulary and
voted against the decision made by one of Sinn Fe´in’s two ministers to
site Belfast’s maternity services in a hospital based in her own constitu-
ency of west Belfast, rather in the adjoining constituency of south
Belfast. Moreover, at the final session of the Assembly held three days
before it was suspended, the DUP made its third attempt to exclude
Sinn Fe´in from holding ministerial office on the ground that it was not
committed to non-violence and exclusively peaceful and democratic
means. The DUP’s motion, by way of a Petition of Concern, narrowly
failed to attract the 30 signatures required to trigger the Assembly’s
cross-community voting procedures.
The capacity of the Agreement’s institutions to provide effective
government and a stable, if interim, political settlement in Northern
Ireland remains a potential rather than tangible reality. In particular,
the ability of members of the Assembly to work in tandem with the
Executive Committee in agreeing a programme for government provides
a severe test for Northern Ireland’s consociational democracy. Indeed,
the same is true of the four-party Executive itself. Accustomed to
‘oppositionism’, both in relation to one another and to the direct-rule
regime, its composition seems ready-made to implode. Arguably, the
close attention to institution building that is a hallmark of consociation-
alism, complete with its concern for proportionality and related checks

590
Parliamentary Affairs
and balances, including the unanimity rule, also deflects its proponents
from thinking about the substance of policies. Albeit carefully crafted,
and hard-bargained, the Agreement does have a rather narrow institu-
tional focus, and yet it is more than an instance of orthodox consocia-
tionalism. The fact that its signatories agreed to set in train reform of
the police and the criminal justice system, and to endorse a free-standing
Equality and Human Rights Commissions, means that the process of
social transformation in Northern Ireland will survive—even if the
institutional expressions of the Agreement, including the Assembly,
suffer the same fate as their predecessors.
1
2
For previous attempts to restore devolution see R. Wilford, ‘Regional Assemblies and Parliament’ in P.
Mitchell and R. Wilford (eds), Politics in Northern Ireland, Westview Press, 1998.
In Opposition, both Conservatives and Labour contemplated alternatives. In its 1979 election mani-
festo, the former proposed revitalisation of local government; between 1981 and 1994 the latter
adopted a policy of Irish unity by consent.
3
4
A. Lijphart, The Politics of Accommodation: Pluralism and Democracy in the Netherlands, University
of California Press, 1968; and ‘Consociational Democracy’, World Politics, 1969, p. 21.
There were two such Documents: A Framework for Accountable Government in Northern Ireland,
produced by the British government, and A New Framework for Agreement, produced by the British
and Irish governments, HMSO, 1995.
5
6
R. Wilford, ‘Epilogue’ in P. Mitchell and R. Wilford (eds), op. cit.
R. Wilford, ‘Inverting Consociationalism? Policy, pluralism and the Post-Modern’ in B. Hadfield (ed),
Northern Ireland: Politics and the Constitution, Open University Press, 1992.
7
When Monica McWilliams and Jane Morrice, the Northern Ireland Women’s Coalition’s MLAs signed
the roll at the first meeting of the shadow Assembly in July 1998, they designated themselves as
‘Nationalist, Unionist, Other’. In the face of objections from other MLAs, notably the DUP, they
redesignated themselves as ‘Inclusive Other’. The standing orders adopted by the shadow Assembly in
March 1999 allow change of a party’s designation once in the life of an Assembly on 30 days written
notice to the Presiding Officer.
8
9
‘Parallel consent’ requires a majority of those present and voting, including a majority of those
designated as unionists and nationalists. A ‘weighted majority’ requires 60% of those present and
voting, including 40% of both unionists and nationalists.
Sinn Fe´in abstained from the vote endorsing the Trimble/Mallon ticket.
10 Establishment of ten departments was an exercise in parity of ministerial esteem. The Executive had
six unionists (four UUP, including the First Minister, and two DUP) and six nationalists (four SDLP,
including the Deputy First Minister, and two Sinn Fe´in). See New Northern Ireland Assembly Report,
15.2.99 for a list of departmental functions.
11 The Agreement had suggested that the Assembly might create a Department of Equality.
12 The Agreement committed the British government to create a statutory obligation on public authorities
in Northern Ireland ‘to carry out all their functions with due regard to the need to promote equality of
opportunity in relation to religion and political opinion; gender; race; disability; age; marital status;
dependants; and sexual orientation’.
13 At the ‘brainstorming session’, participants got a copy of the programme for government agreed by the
Irish Fine Gael, Labour Party and Democratic Left coalition in 1994. The intention was to demonstrate
the possibility of brokering such a programme.
14 New Northern Ireland Assembly Report, 18.1.99.
15 The six implementation bodies are inland waterways; food safety; trade and business development;
special EU programmes; language; aquaculture and marine matters. The six matters for cross-border
cooperation are: transport; agriculture; education; health; environment; and tourism.
16 It included bills on ground rents, inter-country adoption, street trading and child support, and bore no
relation to the intended programme for government.
17 The committees included Audit, Business, Procedures, Public Accounts, and Standards and Privilege.
The Deputy Speakers come from the UUP, SDLP Women’s Coalition.