CAIN Web Service

'The Reform of Stormont' by New Ulster Movement (June 1971)

[Key_Events] [KEY_ISSUES] [Conflict_Background]
POLITICS: [Menu] [Reading] [Articles] [Government] [Political_Initiatives] [Political_Solutions] [Parties] [Elections] [Polls] [Sources] [Peace_Process]

Text: New Ulster Movement ... Page Compiled Brendan Lynn

The following pamphlet was published by the New Ulster Movement in June 1971. The views expressed in this pamphlet do not necessarily reflect the views of the members of the CAIN Project. The CAIN Project would welcome other material which meets our guidelines for contributions.






This pamphlet is designed to stimulate informed debate in Northern Ireland and in Great Britain on the need to change Northern Ireland’s political structures as an essential part of uniting our community.

The pamphlet was drafted by a member of the N.U.M. Executive, Dr. John Whyte, Lecturer in Political Science at the Queen’s University of Belfast, but it has been discussed and approved by the Executive as a whole.

June 1971.



Northern Ireland, with a population of just over 1.4 million in an area of 5,452 square miles, has been in existence for fifty years. In some spheres, it has creditable achievements to record. Homes, motorways and hospitals have been built, social services improved, new industries brought in. But politically, the record has been depressing. The province was a divided community when it was formed, and has remained a divided community throughout. There was a time in the early and middle sixties when old-established antagonisms seemed at last to be on the wane, but the violence of the last three years has made them more obtrusive than ever.

To some extent the problem is economic: if Northern Ireland were an area of full employment and high prosperity, its tensions would doubtless diminish, though they would not disappear altogether. To a great extent, the problem is one of attitudes: on both sides there are deep-rooted fears and resentments which cannot easily be cured. However, the psychological factors can be exaggerated too: there are very few people in Northern Ireland who positively court violence, and the great majority on both sides of the community divide would, we believe, be glad to help in building a reconciled and united province. If they have so far been unable to make their full weight felt, the fault lies in a third area: the area of political structures. The purpose of this pamphlet is to discuss ways in which these political structures can be improved, so as to release the beneficent forces that undoubtedly exist in the province.

First, however, it is necessary to analyse where the weakness of the present political structures lies. This can be summed up in one sentence. It is that the Westminster model of government has been applied in a society to which it is not suited. The Westminster model works well only in rather special circumstances. It is a system which can easily lead to tyranny, because it produces a remarkable concentration of power in the government of the day. To avoid this risk, the system requires the existence of two strong parties, competing with each other at elections on fairly even terms. Where this happens, power is not monopolised by a single team of ministers, but alternates between two teams, which between them have the support of the great majority of the electorate. Thus, though at any given moment, one party has 100 per cent of the power and the other has 0 per cent., over a period of time things even out, and no large part of the electorate feels permanently excluded from a share of power. Moreover, in such circumstances the government will treat the opposition with respect, and will even consult it on matters of policy, because ministers know that their own turn in opposition may come soon, and that they wilt then be glad of considerate treatment from their successors in government.

It is easy to see that, in Northern Ireland, the conditions for working the Westminster model have not been attained. Here, instead of two vigorous parties competing on fairly equal terms, we find one party permanently in office while the opposition is split into several fragments. This means that one section of the community has monopolised power, with consequent tendencies to complacency, arrogance and at times injustice: while another section has been excluded from power, with consequent tendencies to frustration, irresponsibility and at times a hankering after violent solutions. The political structures of the province, instead of assuaging community discord, have been so designed that they increase it.



In these circumstances, two possible strategies might be followed to take Northern Ireland out of its impasse. The first is to alter the party system so as to make the Westminster model more workable. The second is to abandon the Westminster model and move to some other system more suited to the situation in the province.

The first of these strategies is already being actively pursued. The Alliance party has been formed to bring Protestants and Catholics together in a viable alternative to the Unionists. The Northern Ireland Labour Party and, on a smaller scale, the Ulster Liberal Party have been working on the same lines for years. The newly formed Social Democratic and Labour Party, while it is so far mainly Catholic in clientele, is in intention a non-sectarian party designed to appeal to Protestants as well as Catholics. There are some Unionists who would like to see their party purged of its right wing, even at the cost of endangering the permanence of its grip on power. If any or all of these Initiatives were successful, they would break the log-jam in Ulster polities, and would set the province on the way to working the Westminster model successfully. They would. then, be a marked improvement on the present situation, and we in the New Ulster Movement wish them well.

At the same time, however, it would be foolish not to recognise the difficulties in their way. The experience of other democracies, as well as our own, suggests that party allegiances are remarkably stable, and that generally speaking it takes decades for a major change to be made in the party system of any country. In Britain, for instance, it took a quarter of a century for the Labour party to replace the liberals as the main alternative to the Conservatives. More rapid changes have been seen in some countries - for instance, the sudden eclipse of the Irish Parliamentary Party in the non-Unionist areas of Ireland, and its replacement by Sinn Fein, at the general election of 1918. But changes as dramatic as that cannot be counted on. Our problem is too urgent for us to rely for its solution on this strategy alone.

It is worth examining, therefore, the second strategy for resolving the province’s problems : namely, to alter not just the party system but the constitutional structures. For we are not bound to stick to the Westminster model. There are many different kinds of democracy: the United States has one pattern of government, France has another. The examples we find particularly interesting are a group of smaller countries on the continent of Europe, whose systems of government differ considerably from Westminster’s but which none the less count as functioning democracies. The states which we have in mind are Switzerland, Austria, the Benelux countries and the Scandinavian countries.

What makes these countries so apt for study from the Ulster point of view is that several of them have past histories of community cleavage. Switzerland had a civil war between Catholics and Protestants In the nineteenth century. Austria had a civil war between Catholics and Socialists in 1934. In the Netherlands there have been in the past deep cleavages between Catholics, Protestants and secularists. In Belgium, there has been long-standing friction between French-speaking and Dutch-speaking linguistic groups. The Scandinavian countries are more homogeneous, but even there, there have been issues. like the language question in Norway, which have caused community cleavages, if all these countries have managed to control their tensions and develop into viable democracies, they may have something to teach us.

Thus, while we extend every encouragement to those who are pursuing the first strategy - the vitalising of Northern Ireland’s party system - we feel that the second strategy - the reconstruction of Northern Ireland’s constitutional structures - should be followed at the same time, The rest of this pamphlet is a contribution to this purpose. In it we shall examine various constitutional devices, drawn largely (though not entirely) from the experience of the smaller European democracies just mentioned, whose applicability to Northern Ireland is worth considering.



We start by discussing proportional representation (or P.R.) which is the only possible constitutional change that has, so far, received any widespread attention in the province.

P.R. is an electoral system which entails dividing a country into multi-member constituencies, and then distributing the seats in each constituency so as to reflect, as accurately as possible, the amount of support which each party receives in that constituency. Versions of P.R. are used in all the smaller European democracies we have mentioned as well as in Italy, Israel, Malta, Tasmania and the Republic of Ireland. It was also used in Northern Ireland from 1921 to 1929.

In contrast to P.R. stands the system with which we are familiar in Westminster and Stormont elections, and which is also used in a number of other countries, like Canada, New Zealand and the United States. This involves dividing the country into single-member constituencies, and awarding the seat to whichever candidate wins the most votes, regardless of whether he has an absolute majority or not. It is often known as the ‘first-past-the-post’ system. It usually gives the strongest party a much larger proportion of seats than it has of votes, and correspondingly penalises weaker parties.

We shall now try to assess, as objectively as we can, the merits and demerits of reintroducing P.R. in Northern Ireland.


1. The main argument used against P.R. is that, just because it is fair to all parties, it may encourage splinter groups to set up parties of their own, thus producing a situation where Parliament is fragmented into a number of small parties, and the only governments that can be formed are unstable coalitions.

In some countries - for instance, France under the Fourth Republic, or Germany before Hitler - this has been a real problem. It was probably also the consideration which led Lord Craigavon to abolish P.R. in Northern Ireland in 1929. Northern Ireland had had two general elections under P.R.: in the first (1921) the Unionists had won 40 seats, in the second (1925) their strength had dropped to 32. The losses were due, not to any increase of anti-partitionist sentiment - Nationalists and Republicans gained 12 seats at both elections - but to inroads made by Labour and independent Unionist candidates, who accepted the constitutional position but differed from the official Unionists on other points of policy. If P.R. had not been abolished at that point, the official Unionist majority might have vanished altogether at the next general election.

However, the record shows that unstable government is not necessarily or even usually associated with P.R. Among the countries which have P.R. and at the same time have reasonably stable governments are all the smaller European democracies - Switzerland, Austria, the Benelux countries, the Scandinavian countries - which we have previously mentioned as holding lessons for us in Northern Ireland. In the Republic of Ireland, P.R. has not prevented stable government. Nor has it in Israel.

In the case of Northern Ireland, it could well be argued that, if P.R. did introduce a little more instability into the system, this would be a good thing. For the trouble in this province has been of the opposite kind: instead of instability, we have had rock-like immobility. To have one party in office without a break for fifty years constitutes a world’s record among democracies, equalled only by the hold of the Democratic party on some states in the American deep South. Lord Craigavon may have done the best thing for his party by abolishing P.R. in 1929; it is doubtful if he did the best thing for the province. We have a great deal of margin to spare before enough instability is imported into our system to reach danger point.

2. As well as this general argument against P.R. there is, however, another, more specifically applicable to Northern Ireland. This is that extremists might do better under P.R. than under the present system, and that the province might consequently become even more difficult to govern.

We have tried to calculate what might have happened if P.R. had been in operation at the time of the last Northern Ireland general election, in February 1969. The exercise inevitably includes a great deal of guesswork, because assumptions have to be made about how constituencies would have been divided, and about how the vote in each constituency would have been distributed given the wider range of candidates that would probably have stood under P.R. We have assumed that the reintroduction of P.R. would have been accompanied by a fairer allocation of seats in proportion to votes than actually exists, so that Belfast and Fermanagh (which are at present over-represented) would have lost seats while Antrim, Down and Londonderry (which are at present under-represented) would have gained. On this basis we calculate that the result might have been as follows: -

  Possible Results under P.R. Actual results
Anti-O’Neill Unionists (including Protestant Unionists) 15 13
Pro-O’Neill Unionists (including unofficial unionists) 21 26
N.I. Labour Party 3 2
‘Green opposition’ (i.e. Nationalists, Republican Labour, and independents who have since joined the S.D.L.P.) 13 11

52 52

We stress that these figures can only be a guess, and that other observers working on different assumptions might arrive at different results. But even if the precise figures varied, the main conclusion would probably stand: that, in the circumstances of February 1969, P.R. would probably have weakened moderate Unionists and strengthened extremists. This would have led to a government further to the right than the one we have actually had, and, probably, to even worse community tensions than the ones we have endured.

Before rejecting P.R. on this ground, however, we should attempt to project what might happen at the next general election. It is generally agreed that, since 1969, there has been a marked polarisation in Northern Ireland. and that extremists on both sides have gained ground. In these circumstances, and under the current electoral system, the next general election could produce an unpleasant setback for moderates of all kinds and a confrontation between intransigent blocs. P.R. on the other hand would ensure that moderates at least won seats in proportion to their support in the population. P.R. has thc effect of rescuing weaker groups. In 1969, it might have helped the extremists; but at the next general election, it is moderates who would be more likely to benefit.


The disadvantages urged against P.R. do not, then, turn out to be decisive. Furthermore, there are some positive arguments that can be urged in favour of P.R.:

1. A point that might not be important in every country but which certainly applies in Northern Ireland is that P.R. should give the ordinary electors a bigger say in the choice of their M.P.s than they have had under the existing electoral system. Northern Ireland general elections have been notorious for the number of uncontested seats: 23 in 1965, 24 in 1962, 27 in 1958, and as many as 29 back in 1933. This means that in many constituencies the electors did not choose their representatives at all: the effective choice rested with a small coterie of selectors, Unionist or Nationalist. True, the pattern broke down in 1969, when only seven seats were uncontested: and it may be that, with the formation of the Alliance Party and renewed determination in the Labour party, the ruling party will never have so easy a passage again. But the reintroduction of P.R. would increase the likelihood that all constituencies would be contested. Under the present electoral system, a party cannot hope to win a seat unless its candidate polls more votes than anybody else. Under P.R., a party can hope to pick up at any rate one seat in a multi-member constituency, even if its supporters are only a minority in that constituency, and there is therefore much more incentive to put up candidates. This was evident when Northern Ireland last had P.R.: in the general election of 1921 every constituency was contested, and in the general election of 1925 every constituency save one.

2. P.R. is more manifestly fair than the existing electoral system. We are not talking here of gerrymandering - i.e. the deliberate drawing of electoral boundaries in such a way as to maximise the advantage of the party in power. Gerrymandering is possible to a limited extent even under P.R., and some would say that the Fianna Fail government has indulged in it in the Republic. The way to guard against gerrymandering is to take the drawing of boundaries out of the hands of the government, and entrust it to an impartial commission. In the Republic, such a commission does not exist; in Northern Ireland, it has been provided for by the Electoral Law Act (Northern Ireland) 1968.

When we say that P.R. is fair, we mean simply that, under P.R., seats are allocated as closely as possible in proportion to votes cast. This is in itself a guarantee of fairness which inspires confidence in both sides of a divided community. In all the divided countries of the continent which were mentioned earlier in this pamphlet - Switzerland, Austria, the Benelux and Scandinavian countries - P.R. has been introduced. In all these countries its introduction was found necessary to maintain confidence in the fairness of the country’s political system.

3. However, the strongest argument for P.R. probably is that it provides psychological underpinning for other and more important reforms which can also be introduced. A country which introduces P.R. gets used to the idea of power being shared between groups in the community in proportion to their strength. Once this principle has been adopted, it can be extended in ways that will be explored later in this pamphlet.


The argument so far suggests that the balance of advantage lies with restoring P.R. in Northern Ireland. There is one more question to be settled, however, before we move on to discuss other possible reforms. This is to decide which kind of P.R. might be best suited to the province.

There are two kinds of P.R.: the list system, and the single transferable vote (S.T.V.). The former is the more widespread, being found in all the continental democracies which have P.R., and in Israel. The latter is more popular in the English-speaking world, being found in the Republic of Ireland, Malta and Tasmania; it was also the type used in Northern Ireland between 1921 and 1929. The list system is now (since the party conference of May 1971) advocated by the Northern Ireland Labour Party; S.T.V. is the system which most other advocates of P.R. in Northern Ireland appear to have in mind. In our opinion, either type of P.R. would be an improvement on the current electoral system, but there is just enough difference between the two types to make it worth discussing their respective merits.

The most obvious difference between the two kinds of P.R. lies in the role accorded to the voter, Under the list system each party puts up a list of candidates for the constituency, and the voter simply puts an X against the list of the party he prefers - just as he puts an X against the individual candidate whom he prefers under the present system. Under the ST.V. system on the other hand, while each party puts up as many candidates as it likes, the voter expresses his preference not for a party but for an individual candidate. What he does is to put the figure 1 against the candidate he likes best, 2 against the candidate whom he likes next best, and so on. He need not number them all, and his ballot is still valid even if he does no more than put 1 against the candidate of his first choice. The more preferences he expresses. however, the more effective his vote is likely to be.

From these differences between the two systems, the following advantages and disadvantages arise:

1. The list system is the easier to operate. Quite simple rules ensure that seats are distributed between parties as accurately as possible in proportion to the votes which each party received. Under S.T.V., however, the rules for counting and distributing votes are much more complicated. (For readers interested in detail, we include, as an appendix to this pamphlet, an example of S.T.V in operation). However, this is not a problem which need concern the ordinary voter. Provided the returning officer and his staff can master the rules - and they have proved perfectly capable of doing so in every country which has used S.T.V. - the system is quite practicable.

2. The main argument which the Northern Ireland Labour Party puts forward for preferring the list system is that it ‘allows the voters sharp and clear cut choices between opposing political philosophies,’ while S.T.V. tends to blur the sharp edges of debate. This is because, under the list system, the voter chooses one party and one party only, while under S.T.V. he can give his lower preferences to candidates of parties other than the one he likes best.

This however, in the circumstances of Northern Ireland, seems to he an advantage and not a disadvantage. It could help to diminish polarisation in the community. A Unionist might use his lower preferences to favour a moderate opposition candidate as against an extremist; and an opposition voter could use his lower preferences to favour the less objectionable Unionists (from his point of view) as against the more objectionable. In this way, both sides could learn to see the good in some of their opponents. S.T.V., then, could have an educative effect on the electorate which would not he obtained under the list system.

3. A further disadvantage of the list system is that it leaves meet’ power in the hands of the party machines than does S.T.V. Let us take a six-member constituency in which party A has the support of roughly one-third of the electorate. Under either type of P.R., party A is likely to receive two seats out of six. But under the list system, the two seats will go automatically to the two candidates whom the party has put first and second on its list: it is not the voters, then, but the party leaders who decide which of the party candidates are most likely to be elected Under S.T.V., on the other hand, the seats will go to whichever two candidates of the party are most favoured by the electors. We think this important in an area like Northern Ireland, where one party label can cover such a variety of opinions.

Our conclusion, then, is that the balance of advantage definitely favours the S.T.V. type of P.R. as against the list system.



When people discuss the strengthening of parliamentary committees at Westminster or Stormont, they generally argue over whether it will increase the efficiency of Parliament or the influence of back-benchers. We wish to approach the question from a different angle. We wish to discuss whether the development of such committees would assuage the community tensions of Northern Ireland.

First, a word should be said about the current position of such committees. At Stormont, they are almost non-existent. Virtually all legislation is discussed by the two houses in full session. There are hardly any committees for scrutinising administration the Public Accounts Committee, which examines the expenditure of government departments. is the only important exception.

At Westminster, the development of such committees has gone considerably further. The detailed examination of most bills goes on in standing committees, of which there are six. Alongside these. there is a set of select committees which scrutinise various aspects of government administration, and which have expanded greatly in number and powers over the past five years.

On the continent, things have gone further still. Here, most of the work of Parliament has been delegated to committees. The exact functions of these committees vary from one country to another, but in general, the two tasks of examining legislation and scrutinising administration, which at Westminster are divided between two types of committee, are in continental countries both handled by the same committees. Each committee has oversight of a particular subject, and both examines legislation and keeps an eye on government departments dealing with that subject. In pursuit of these tasks, committees are entitled to call for documents, including official documents and examine witnesses, including civil servants. In some countries, they are empowered to hire specialist staff so that they will have independent expert advice to set against what the civil servants tell them.

The proposal which we are examining here is that a committee system should be developed at Stormont which goes at least as far as that at Westminster, and might be taken as far as that to be found in continental countries.


1. The practical objection most likely to be raised against any extension of the committee system at Stormont is that it is unnecessary. Stormont is only a subordinate parliament, and its volume of business is far below that of a sovereign legislature such as is found at Westminster or in continental capitals. The Stormont parliament generally sits only three days a week, and, even without a committee system, can comfortably handle all its business.

As against this, however, it can be argued that if such committees were set up, they would easily find plenty to do. Even if they did not have much legislation to consider, there is still much room for expansion in the scrutiny of government administration, which so far is hardly done at all at Stormont.

Furthermore, once the Macrory proposals on local government reform are implemented, the volume of business at Stormont will grow, because these proposals entail transferring to Stormont control many services now provided by local authorities. The Macrory report itself suggests that Parliament may wish to establish committees to oversee the services transferred to Stormont. (Report of the Review Body on Local Government in Northern Ireland, 1970, Cmd. 546, para. 92). If committees are established to keep an eye on these services, it would be only logical to extend the committee system so as to oversee all services provided by government.

2. Another objection to introducing such committees is that party whips find them bad for party discipline. It is true that in all countries where such committees exist, the rule is that a party is represented on each committee in proportion to its strength in the house as a whole, so that a government with a safe majority in Parliament ought to have a safe majority in each committee. But in practice, things do not work out like that. Members of such a committee tend to develop a common interest in their problems, and to generate an esprit do corps which cuts across party loyalties.

But In the circumstances of Northern Ireland, this would seem a good, rather than a bad, thing. Any device which generates loyalties cutting across party boundaries is, in the circumstances of this province, desirable. The reason why party whips dislike these committees is the very reason why they should be encouraged.

3. Furthermore, a developed committee system em in Parliament has the advantage of involving representatives of all parties on equal terms. In the intimate atmosphere of a committee members tend to forget each other’s party labels and to assess each other as persons. This means that an opposition member with a grasp of the subject will earn the respect of even government supporters, and can have a real influence on the committee’s decisions. The effect of this in reducing frustrations on the opposition side of the house can easily be imagined. P is probably no coincidence that, in all those smaller European democracies which we have singled out for their success in controlling community tensions, a developed committee system operates in Parliament.

A number of problems would have to be settled before such a committee system could be introduced at Stormont. One would be to decide how many committees should be established. At present there are nine government departments (including the prime minister’s) at Stormont, and a simple decision might be to have one committee to watch each department. On the other hand, it might be possible to have one committee overseeing two of the less busy departments, thus reducing the total number of committees Again it would have to be decided whether Senate and House of Commons should each have their own set of committees, or whether joint committees could he established. On the continent, most parliaments have distinct sets of committees for each house, but Sweden has joint committees, and the latter might seem an arrangement better suited to the limited manpower available in Northern Ireland. On the other hand, to establish joint committees would entail more drastic changes in parliamentary procedure. however, these are all problems of detail which could easily be settled. On balance, we would strongly recommend the establishment of such committees at Stormont.



This follows logically from our last proposal. At present Stormont is rather small to handle a developed committee system, and its numbers would need to be increased. The Macrory report has already suggested (pan. 91) that Parliament might need to be enlarged in order to rope with the greater volume of business which its proposals will generate for M.P.s. If, on top of this, a developed committee system is introduced. the case for an enlargement will be stronger still.

The only question that would remain to be settled is how substantial the increase should be. If it is too large, there would be physical difficulties in providing for all thc new members at Stormont. If it is too small, it will not make enough improvement to the working of the house. Probably an increase of 50 per cent. would be about right. This would raise the strength of the House of Commons from 52 to 78, and of the Senate from 26 to 39.



If Parliament were to be enlarged, it would be worth considering new ways of selecting the additional members of the Senate.

First, it might be as well to outline the functions of the Senate, which on paper are substantial. Every bill has to be approved by the Senate as well as by the House of Commons. In case of deadlock, a decision may be taken by the two houses sitting jointly. As the House of Commons is twice as large as the Senate, this would mean that, in such a joint sitting, the Senate would have one-third of the decision-making power.

In practice, the Senate has made little use of its paper powers. Only twice has it rejected a bill coming up from the Commons. The right to resolve a dispute by a joint sitting has never been used. If the Senate has made so little impact, part of the reason may lie in the ways in which Senators are chosen. At present, Senators reach the house in one of two ways:

(a) The Lord Mayor of Belfast and (when the office is not in abeyance) the Mayor of Londonderry hold their seats ex officio.
(b) The remaining 24 Senators are chosen, by P.R., by members of the House of Commons. They serve for eight years; twelve of them are elected every four years.

The result of this system of selection is that the Senate is largely a replica of the House of Commons, both in party composition and in the type of person selected. There would be obvious advantages in adopting other means of selection for certain of the Senators, so as to bring in a wider variety of outlooks and backgrounds, and thus invigorate the work both of the house as a whole and of parliamentary committees.

We examined a number of possible ways of recruitment, and we shall state briefly the merits and demerits of the methods we considered:

1. University representation might be restored, but put in the Senate rather than the House of Commons. It is generally agreed that the four Queen’s University M.P.s who sat in the Stormont House of Commons until 1969 made a better-than-average contribution to the work of the House. In the Republic, where university representatives sit in the Senate, it is generally agreed also that they make an important contribution. On the other hand, to reintroduce university representation could he criticised as being undemocratic: it would mean that one section of the community - university graduates - would have a greater say in the composition of Parliament than any other.

2. Certain interest groups - e.g. the Irish Congress of Trade Unions, the Confederation of British Industries, the Ulster Farmers’ Union, the Northern Ireland Council of Social Service - might be given the right to nominate Senators. This could be a particularly valuable way of introducing highly-qualified people into the Senate who might otherwise never get there. On the other hand it could also be criticised as undemocratic. Furthermore, some of these interest groups might not be at all anxious to have the right of nominating senators bestowed upon them. As things are, these interest groups are influential enough to obtain a hearing from government on any matters which interest them. If they had a man in the Senate, he would have to take a stand on every issue, not only those of concern to his interest group, with the consequent danger of causing unnecessary friction with their own members and with other groups.

3. The government of the day might be allowed to nominate a small number of senators, on the understanding that the nominations would be used to bring into the senate men with talents and experience that would otherwise not be found there. The difficulty here would be to ensure that the government used the right in this way, and not to bolster its own majority.

4. The number of representatives of local authorities might be increased. This proposal has already been made in the Macrory report. which suggests (para. 94) that chairmen of district councils might elect a few of their number to sit in the Senate for a term. Such a provision would add to the number of Senators with practical experience of operating public services on the ground. It would also be less open to the complaint of being undemocratic than the proposals numbered 1 to 3 above. On the continent of Europe. the commonest method of selecting the upper house of a parliament is to have it elected by local government councillors. On the other hand, this method might be the least effective of the four we have considered for bringing Into the Senate different types of expertise and background.

We do not, therefore, make any firm recommendation as between the four methods of selecting Senators that we have considered. But we do consider that if any one of them, or any combination of them, were adopted, it should make some improvement to the working of Parliament.



This is a more radical proposal than any of those already discussed. It would involve the abolition of the distinction between government and opposition, and the establishment of a government in which all parties are represented in proportion to their strength in Parliament. Even on the continent, it is much less widespread than the other devices we have discussed. A number of countries have had all-party governments in times of emergency - just as the United Kingdom had during the two world wars - but there are only two countries where the principle of P.R. in government has been tried for any long period. One is Austria, where the two main parties, the Catholics and the Social Democrats. ruled jointly from 1945 to 1966. The other Is Switzerland, where, after a long period of development from the eighteen-nineties to the nineteen-fifties, it seems now to be accepted as a permanent feature of the Swiss way of life.

In Switzerland, where the system has been most fully developed, it works as follows. The people elect their members of parliament, for a fixed period. (In the lower house, the period is four years; in the upper house, it varies with the canton, and may be anything between one and four years; but the important point is that all members of parliament are there for a fixed term), Parliament in its turn elects the seven members of the government, also by a system which is in effect proportional representation, and also for a fixed period (in this case, four years). There is no prime minister: the chairmanship of the government is held, for a year at a time, by the different ministers in rotation. The system seems strange to people brought up on the Westminster model; and yet it works well in Switzerland, a country which in the nineteenth century had tensions between Catholic and Protestant as acute as our own. The merits and demerits of applying it to Northern Ireland are at least worth discussing.


1. The system seems at first sight to be peculiarly liable to deadlock. In most European countries, the government can be unseated by Parliament if it ceases to have the support of a parliamentary majority. In Switzerland, however, the government is in office for a fixed term. and remains there even if Parliament does not entirely approve of its policies. Again, in most European countries a minister who works badly with his colleagues can be dismissed by the prime minister. But in Switzerland, where there is no prime minister and all ministers are in office for the same fixed period, there is no way of getting rid of a difficult colleague.

In practice. however, these difficulties do not arise in Switzerland. Those who have to work the system have developed the appropriate attitudes of mind. Realising that they cannot get rid of each other, they have learnt how to live together. Ministers, in particular, have developed an outlook more akin to our senior civil servants than to our ministers, They put forward proposals; if these are rejected either by their colleagues or by Parliament they do not feel It necessary to resign; they accent the situation philosophically and try to find an alternative which will have more chance of winning approval.

The system works successfully in Switzerland despite considerable divergences of opinion among the main parties. The present composition of the government is as follows:

Socialists 2
Radicals (the party of the Protestant bourgeoisie: despite its name, rather conservative 2
Christian Conservatives (the Catholic Party) 2
Farmers' Party 1

This probably represents as wide a spread of opinion as would be found in Northern Ireland, if such a system were introduced here. Radicals and Christian-conservatives represent different religious interests; Socialists and Farmers represent different economic ones. A coalition including Mr. Paisley, Mr. Faulkner, Mr. Vivian Simpson and Mr. Hume would hardly be more divergent; yet in Switzerland they all manage to work together.

2. A more serious argument against P.R. in government is that it deprives ministers of the stimulus of a vigorous opposition. Under the Westminster system, when it is working well, the government is faced by a lively opposition, constantly probing its weaknesses, suggesting alternative policies, and posing a real threat at the next general election. This is a powerful guarantee that the government will not become complacent or overbearing. Under an all-party government, faced with no real competition, there is more danger that ministers will become lazy and unimaginative.

We think there is real force in this argument, and, in countries where the Westminster model can work well, we would not advocate its abolition. But in Northern Ireland this option is not open: for reasons explained at the beginning of this pamphlet, the Westminster model does not work well here. In this province, the choice is between carrying on with a system of government which in local circumstances works badly, and adopting a system which has worked well in another deeply divided community.

Moreover, we feel that this particular argument against the Swiss system has more force in a sovereign parliament than in a parliament with devolved power like Northern Ireland’s. For the scope for independent decision at Stormont is in practice extremely narrow. Financial and economic policy is almost wholly dictated by Westminster; social policies develop step by step with Westminster’s. In the past Northern Ireland ministers have had rein for their ingenuity in the field of electoral law and of policing; but since Westminster developed an interest in these fields as well, there is less scope for independent decision even here. There is little room left for independent action except in the details of commercial, agricultural and educational policy. To argue that a Swiss-type system might hamper the formation of imaginative and lively policies is then, a much less serious charge in Northern Ireland than it would be in an independent country.

3. A Swiss-type executive system would have one incidental merit: it would widen the field for ministerial recruitment. At present, ministers in the Northern Ireland government must either he members of one or other House of Parliament, or must obtain election to one or other House within six months of appointment. The result is that the reservoir of potential ministers is lamentably small. Under the Swiss system, ministers have the right to speak before either House of Parliament hut are not themselves members of Parliament: if a member of either house is elected to the government, he resigns his parliamentary seat. The result is that parliamentarians, in electing members of the government, can choose anyone they like. The pool of potential talent is enlarged to include the entire population.

We think, then, that despite the drastic nature of the proposal. the introduction of P.R. at governmental level is well worth considering in Northern Ireland. It is a more radical change than the others advocated in this pamphlet, and more time might be needed to prepare the ground for it psychologically. It might therefore best be brought in after other reforms had begun to take effect. But it should certainly not be lost sight of as a possibility.


We have now discussed five proposals which seem to have positive merits. There is, however, another device which has been used in divided societies elsewhere, and, to complete the discussion, the suitability of this device for Northern Ireland should be examined.

This device can be described as ‘weighted representation for minorities.’ An extreme form of It can be found in the Cyprus constitution of 1960, which gave the 17 1/2 per-cent Turkish minority extensive safeguards as against the Greek majority. It was allotted 30 per cent. of the seats in government and parliament; and on certain matters, such as taxation, any change in the law had to be approved not just by a majority of Parliament as a whole, but by majorities of the Greek and Turkish deputies voting separately - which gave the small Turkish minority power in these matters to thwart altogether the wishes of the large Greek majority. A much more attenuated version of the same principle can be found in a constitutional amendment recently adopted in Belgium. Belgium is divided into two communities deeply distrustful of one another: the French-speaking Walloons (40 per cent of the population), and the Dutch-speaking Flemings (60 per cent). Under the new arrangements, each community is given a limited veto over legislation. The way it works is that if 75 pr cent of either the Walloon or the Flemish deputies oppose the Hill, it is referred back to the government for reconsideration even if it is supported by a majority of deputies as a whole. Though the government would be legally entitled, after it had reconsidered its proposals, to refuse to alter them, the presumption is that it would hesitate to do so if it knew that so large a proportion of one or other of the linguistic communities was opposed to its measure.


Such arrangements have the apparent advantage of giving added security to minorities; hut on closer examination they seem to have more serious disadvantages

(1) They are not very democratic. They mean that minorities have, not just their fair share of power, but more than their fair share. Under the Cyprus constitution, the 17 1/2 per-cent Turkish minority had on some issues veto powers over the wishes of the 82 1/2 -per-cent Greek majority. In Belgium, three-quarters of the Walloon deputies can hold up a measure even though they are only 30 per cent. of the parliament.

(2) As a result, they may not be very effective even as a means of keeping the peace. In Cyprus, the constitution quickly broke down because the Turkish minority used its constitutional powers in a way which seemed unreasonable to the Greek majority. In Belgium, it is too early to say whether the new arrangements will work. If they do work, it may be due not to their intrinsic merits, but to the fact that there will lie little occasion to use them. Over decades of agitation, the grievances of Fleming against Walloon and of Walloon against Fleming have been exhaustively thrashed out, and on practically every issue that could arise between them a settlement has already been reached.

(3) However, even if this objection turns out to be too pessimistic, there is another objection to proposals such as these. They tend, by the very fact that they are designed to protect community groups, to emphasise community divisions. In Belgium, deputies on election have to register themselves as Flemish or Walloon. In Cyprus, there are separate electorates, which means that not only deputies, but electors as well, have to enrol themselves as Greek or Turkish. In the Northern Ireland context, such devices would presumably mean that every UP., or even every elector, would have to declare himself Protestant or Catholic. This would mean emphasising a line of cleavage which, in our view, should in every possible way be bridged over.

For all these reasons, therefore, we do not consider that arrangements to give weighted representation to minorities arc appropriate to Northern Ireland,


As a contribution to the solution of Northern Ireland’s problems we therefore recommend:

1. The introduction of proportional representation into parliamentary elections.

2. The establishment of a system of all-party specialist committees in Parliament, with powers of oversight over government departments.

3. The enlargement of Parliament by 50 per cent.

4. The selection of some Senators by methods other than those at present employed, so as to widen the range of expertise in the house.

5. At least as a long-term possibility - the introduction of proportional representation in government, on the Swiss model.

To introduce these measures into Northern Ireland would, we freely concede, require a transformation in habits of thought. Politicians, civil servants and the general public would all have to get used to a system markedly different from the one in which they were brought up. For the changes to have any chance of success, a large-scale education programme would have to be undertaken. One could envisage plane-loads of politicians, journalists and civics masters taking off for Berne, Stockholm and Amsterdam to see how things were managed there. There would have to be programmes on television and radio, articles in the press, extra-mural courses put on by the universities, to explain the reforms to the public.

At the same time, the extent of the reforms need not be overdramatised. They would in no way affect the constitutional position of Northern Ireland as an integral part of the United Kingdom. They would not do away with political parties, but would simply make them work together for the common good. The changes could be brought about separately, by stages, or all at one time. The necessary legal and constitutional changes would require no great demands on the parliamentary draughtsmen’s skill to draw up, and could be enacted without disruption of the life of the community in Northern Ireland. These are proposals for improving the political structures of Northern Ireland, not for turning the province upside down.

Finally, even if some readers have doubts about the specific proposals which we make, we hope that the principle on which this pamphlet is written will find wide acceptance. This is that the experience of countries with problems comparable to our own has been far too little explored. It is only common sense to examine the solutions which they have evolved so as to learn what lessons we can for ourselves.



A case study of proportional representation in operation.

To illustrate how P.R. under the single-transferable vote system works, we have taken an example from Northern Ireland’s previous experience with this system. The instance we have chosen is the constituency of West Belfast in the general election of 1925.

The first thing which the returning officer had to do, when the ballot papers came in, was to ascertain how many first Preference votes had been cast for each candidate.

The result was as follows:

Devlin Nationalist 17,558
Woods Independent Unionist 9,599
Lynn Unionist 8,371
Burn Unionist 4,808
McConville Republican 3,146
Dickson Unionist 3,133
McMullen Labour 2,869

The next thing the returning officer had to do was to work out the quota. The quota is the minimum number of votes which a candidate needs in order to secure election. The formula used for calculating it is as follows:

West Belfast was a four member constituency, and the valid poll was 49,484, so the quota was 49,484, so the quota was 49,484/5 +1 which equals 9,897 Thus any candidate who won more than 9,897 votes would be elected.

Once the returning officer had worked out the quota, it became clear that one candidate, Devlin, had exceeded It on the first count, and so he was declared elected. The returning officer’s next task was to distribute Devlln’s surplus. Devlin had gained many more votes - to be precise, 7,661 more votes - than he needed to secure election, and it was only fair to allow his surplus voters to express a second preference once their first preference was home and dry. So the returning officer’s next task was to redistribute 7,661 of Devlin’s votes to other candidates in accordance with the second preferences expressed by the voters. (1)

The result was to alter the figures for the other candidates as follows:

Woods Independent Unionist

+ 1,472

= 11,071

Lynn Unionist

+ 136

= 8,507

McMullen Labour

+ 4,368

= 7,237

Burn Unionist

+ 70

= 4,878

McConville Republican

+ 1,310

= 4,456

Dickson Unionist

+ 305

= 3,438

This meant that another candidate, the Independent Unionist, Woods, now reached the quota and was declared elected. His surplus of 1,174 was then redistributed according to the voters’ next preferences and the results were as follows:

Lynn Unionist

+ 92

= 8,599

McMullen Labour

+ 558

= 7,795

Burn Unionist

+ 43

= 4,921

McConville Republican

+ 55

= 4,511

Dickson Unionist

+ 185

= 3,623


At this stage no further candidate had reached the quota (2), so the returning officer now started at the other end and eliminated the bottom candidate, one of the Unionists, Dickson. Dickson’s votes were then redistributed according to the next preference of his supporters, with the following results:


Lynn Unionist

+ 1,838

= 10,437

McMullen Labour

+ 207

= 8,002

Burn Unionist

+ 1,068

= 5,989

McConville Republican

+ 34

= 4,545

As a result of this redistribution, the first of the official Unionists, Lynn, reached the quota and was declared elected. The redistribution of his surplus gave the following results:

McMullen Labour

+ 13

= 8,015

Burn Unionist

+ 520

= 6,509

McConville Republican

+ 7

= 4,552

No further candidate had reached the quota, so the returning officer now eliminated the bottom candidate, McConville. No fewer than 2.216 of his votes were non-transferable. i.e. the electors had expressed no preference as between the two candidates still in the race. But of those which were transferable, the great majority favoured the Socialist as against the Unionist, so the result of the final count was as follows:


McMillen Labour

+ 2,330

= 10,345

Burn Unionist

+ 6

= 6,515

Thus McMullen was elected for the fourth seat and the final result in West Belfast was this:



Independent Unionist






This system of counting votes was obviously far more cumbrous than the present one, and the reader may well ask : was it worth the extra trouble? What difference would it have made if West Belfast, instead of being one large four-member constituency, had been divided into four small single-member constituencies? An answer can be attempted by following the subsequent electoral history of the area. For in 1929 it was divided into four small single-member constituencies - St Anne’s, Falls, Central and Woodvale - and it has remained that way ever since. In the general election of 1929, the seats were won as follows:

Nationalists 2 (Falls and Central)
Official Unionist 1 (St Anne's)
Independent Unionist 1 (J.W. Nixon in Woodvale)

Since then, each of the four seats has had a distinctive history. St Anne’s has always been held by an official Unionist. Woodvale was held by the Independent Unionist, Nixon, until 1950. It was then captured by official Unionists, who held it until 1958, lost it then to Northern Ireland Labour, but recaptured it in 1965. The remaining two seals have always been in opposition hands. Central was held by Nationalists till 1946. Since then its M.P.s have worn a variety of labels - Labour, Independent Labour, National Democrat, Republican Labour. Falls was held by Nationalists till 1942, and since then by Republicans, Socialist Republican, Republican Labour. In 1969 it as captured for Northern Ireland Labour by Mr. Paddy Devlin, but Mr. Devlin has since left that party and joined S.D.L.P. Thus the overall picture of the area’s representation has not been so very different from what it was under proportional representation, with Unionists (either official, or independent) generally holding two seats, and opposition groups generally holding two.

However, before we dismiss P.R. as just an expensive way of reaching a similar result to that under our present electoral system, the following points may be noted:

1. The electorate have less range of choice than under P.R. Although this area has had fewer uncontested elections than some parts of Northern Ireland, the range of choice in any one constituency is generally restricted. In Falls and Central it has tended to be between various shades of green and pink; in St. Anne’s and Woodvale between various shades of Orange. No constituency has ever had the range of choice offered to West Belfast in 1925. when electors could choose between Nationalist, Republican, Labour and two kinds of Unionist.

2. There is some indication that the groups which have lost most from the change have been moderate, bridge-building ones. The Northern Ireland Labour Party, which is the only party to have fought all 4 seats in the area at one time or another, and which has considerable support in the area, has never, since the abolition of P.R. been sure of a seat there. Moderate Unionists, too, seem to be underrepresented. The two Unionists in the area at the moment, Mr. Laird of St. Anne’s and Mr. McQuade of Woodvale are both hard-liners.

Yet at the general election of 1969, unofficial pro O’Neill Unionists in these two constituencies won 7,414 votes, a volume of support which might well have won them a seat under P.R.

3. There is also evidence, from the 1925 election results, of the depolarising effects of P.R. This is specially evident if we note what happened on redistribution to Devlin’s surplus. More than half of it went to Labour, and about a sixth of it to the Republican. But nearly 30 per cent. of it went to Unionists of one kind or another. In other words, something like 30 per cent of Devlin’s voters were either Nationalists who were prepared to give their second preference to a Unionist, or Unionists who were prepared to give their first preference to a Nationalist. Either way the phenomenon was a heartening one. It is made impossible by our present electoral system.


(1) For readers interested in minutiae, we shall explain how the returning officer selected the 7,661 votes for transfer. He could obviously not be allowed to please himself as to which ones he picked, as this could lead to gross unfairness. Instead, strict rules of selection were laid down. First, the returning officer had to go through all of Devlin’s 17,558 votes and sort them into parcels according to the second preferences expressed. At this stage he discarded any of Devlin’s ballot papers on which no second preference was expressed. Suppose there were 1,000 of these. This would leave 16,558 votes on which a preference was expressed. The returning officer had to select for transfer 7,661 out of these 16,558. He did so in such a way that the proportions between second preferences expressed for various candidates were accurately maintained. For instance if x per cent of the 16,558 votes showed a second preference for Woods, then x per cent of the 7,681 votes which the returning officer transferred would also show a second preference for Woods.

(2) The discerning reader may note that though Wood’s surplus was 1,174, only 933 votes were transferred. This would have been because only that number of his voters expressed a further preference. All his other voters would have put either ‘Woods 1’ or ‘Devlin 1’. Woods 2’, or ‘Woods 1, Devlin 2’, without expressing any further choice. If they had expressed a further choice, the returning officer was bound to take account of it.


CAIN contains information and source material on the conflict and politics in Northern Ireland.
CAIN is based within Ulster University.

go to the top of this page go to the top of this page
Last modified :