‘This is a sovereign assembly’: popular sovereignty, parliament and the people in the Irish Free State

ABSTRACT

The stability of Irish democracy since 1922 is often a cause for contentment with lines of succession drawn from ‘Grattan’s parliament’, O’Connell’s popular movement, and the achievements of the Irish Parliamentary Party (IPP) at Westminster through to Dáil Éireann. The IPP certainly disrupted the Commons, yet Irish politics was also shaped by the ‘mother of parliaments’ – MPs mastered parliamentary procedure and home rule legislation drew on British and colonial parliamentary models. By 1918, the party’s machine and parliamentary behaviour were rejected as much as British rule – but as Michael Laffan suggested, Sinn Féin triumphed by beating the home rulers at their own game. However, the 1922 constitution gave generous expression to the sovereignty of the people, and for judicial review. But, within short succession, facility for popular initiative was overridden. Ironically, the Irish Free State would never be amended by popular will or judicial review but by parliamentary majority in 25 amendments as the people only voted to adopt it in 1922 and replace it in 1937. In practice then, it was closer to a vision of Dicey's supremacy of parliament dogma that prevailed, so much so, that Seán Lemass acrimoniously proclaimed in 1934: ‘this is a sovereign assembly’.

KEYWORDS:

• Popular sovereignty
• democracy
• parliament
• Irish Parliamentary Party
• Sinn Féin
• Irish Free State

Power to remove Judges is provided by the Constitution. If it becomes necessary in the mind of the Government to exercise that power, it will be exercised. The removal of the Comptroller and Auditor-General could be effected by legislation at the present time. The mere existence of the Seanad does not prevent it, has not prevented it, if the Government thought it necessary to do it. It is the Government are entitled to decide these questions, on the principle enunciated by Deputy Hogan that the majority has a right to do what it likes. The majority in this Parliament is entitled to enact any legislation it likes. This is a Sovereign Assembly. Seán Lemass, Dáil Éireann debate, 18 April 1934, vol. 51, no. 14.

The stability of Irish democracy since 1922 has been the subject of much comment with lines of succession drawn from at least the eighteenth-century ‘Grattan’s parliament’ on College Green in Dublin, O’Connell’s popular movement of the early nineteenth century, and the achievements of the Irish Parliamentary Party (IPP) at Westminster (Farrell, 1973; McConnel, 2013, pp. 19–20; Mulvagh, 2016). The IPP certainly disrupted the House of Commons from the 1880s, yet Irish politics from 1801 was also shaped by Westminster – Irish MPs mastered parliamentary procedure, and home rule legislation drew on British and colonial models (Ward, 1994). By 1918, the IPP’s machine and parliamentary strategy was rejected by voters as much as British rule – but as Michael Laffan has suggested, Sinn Féin triumphed by beating the home rulers at their own game (Irish Times, 25 April 2012). The Irish state has remained a parliamentary democracy and many have noted the resilience of Irish democracy when so many European states succumbed to authoritarianism in the interwar period (Barry, 2023). Yet, as work on the Irish revolutionary decade, as well as studies of the 1922 constitution, highlight, parliament was not seen as the only or indeed central well of sovereignty, much less democracy – particularly as those who gained their political grounding in Sinn Féin looked askance at the history of the deposed IPP (Cahillane, 2016).

The IPP in many ways had shaped the parliament at Westminster – but the Oireachtas also came to resemble it in the decades after independence. Analysing these developments, this article will first examine the home rule movement, its relationship to parliament and its legacy in assessing Irish political traditions before the writing of the 1922 constitution. In so doing, the aim is to analyse how Irish parliamentarians behaved and how home rule parliaments were conceived alongside the reality of how the Oireachtas operated after 1922. The article asks to what extent the home rule era bequeathed a parliamentary tradition as much as a democratic one (Kissane, 2011, p. xv; chapter 2). The second part of the article then examines how the Free State constitution was debated and drafted - embracing the sovereignty of the people, and provision for judicial review, but reverting in many ways to a view of parliamentary sovereignty as expressed by Lemass in relatively short order.

A house in College Green? The place of parliament in home rule legislation

Home rule as a movement and a popular political catchcry began with the Home Rule League, which was founded in 1873 by Isaac Butt. Conservative in origin, it was dedicated to land reform and offering a constitutional counterweight to Fenianism (Bull, 2009). As Colin Reid has shown, Butt envisaged home rule – or, in its barest terms, a parliament for Ireland in Dublin – in the context of political reform across the United Kingdom (UK) and empire, and not ‘simple repeal’ or O’Connellism redux (2014). Yet, followers of home rule often saw it differently – it was, Alvin Jackson argues, ‘about mobilising patriotic support in nineteenth-century Ireland for a repeal of the union’, but with ‘an acceptance (however conditional) by the Irish that Westminster was the supreme or sovereign parliament’ (2003, p. 2). While Butt was succeeded by William Shaw as party leader in 1879, it was under Charles Stewart Parnell that the IPP at Westminster and in Ireland took on a different visage. Scion of a Protestant landed family in Wicklow (though Avondale was already ‘heavily encumbered’ when Parnell’s father died), Parnell showed a contrasting attitude to parliament before assuming leadership (Callanan, 2009). While Butt, a lawyer, had displayed reverence for the conventions of Westminster, Parnell, and Joseph Biggar, engaged in obstructionism to filibuster and bring its proceedings to a halt (Thornley, 1960). While a return to an Irish parliament in College Green remained the goal (McConnel, 2013), upon assuming leadership in 1880, Parnell ensured that party rhetoric became stronger in articulating Ireland’s claim for autonomy.

The IPP’s partnership with the Land League also brought grassroots political ballast – and of course, in some British eyes, an association with agrarian crime (O’Callaghan, 1994). The party did, however, succeed in wringing from the British government a series of land acts which helped to provide tenants with a pathway towards owning their land. From 1883, Labourers Acts were introduced which helped some of the country’s agrarian poor, even if the extent to which the IPP represented all segments of rural Ireland is debated (Aalen, 1986; Jones, 1986). While fears for democracy were often expressed by political actors and thinkers in the age of reform (Reid, 2014), Parnell’s party profited from major democratic revisions including the 1872 Secret Ballot Act and extension of the franchise in 1884. The latter helped it to gain more seats nationally, and in Ulster more seats than unionists, even encouraging what James Loughlin termed the ‘seductive conclusion’ that the province was an ‘electoral problem’ (2007, p. 161; Jackson, 1989). Parnell shaped the IPP into how it is often seen: the template for the modern mass-based party with a disciplined parliamentary caucus standing squarely behind its leader (O’Malley, 2015).

The first home rule bill, proposed in 1886 to grant Ireland self-government within the UK, saw Westminster and its imperial mechanisms inspire the draft. While Ireland has been the subject of much debate over whether its legal status was analogous to colonies within the empire or to the metropole due to the unusual status of its government (Fitzpatrick, 1999), as Ward (1994) has pointed out, Gladstone used ‘constitutional language’ from Canada in drafting the legislation, including the 1840 colony of Canada constitution and the 1867 federal constitution. This legislation proposed that the lord lieutenant would represent the Crown in Ireland and hold the power to ‘appoint ministers; summon, prorogue or dissolve’ the parliament. Money bills would also have to have his recommendation (Ward, 1994, pp. 63–64). While the Irish parliament was to be subordinate to Westminster and the lord lieutenant would possess executive authority, Gladstone insisted home rule was ‘strictly and substantially analogous’ rather than ‘parallel’ to the Canadian constitution. A unicameral parliament of two orders was proposed – a First Order of 103 members and a franchise with property restrictions, while the Second Order (206 members) would be more popularly elected (Ward, 1994, p. 67). The provision that the First Order might pay closer attention to property owners assumed importance given British concerns about the ‘fitness’ of the Irish for self-government. As Collombier-Lakeman (2016) has shown, this occupied a significant part of the House of Commons debate on the home rule bill.

Many debates, however, centred on imperial unity rather than democracy per se, and on UK-wide implications. On the latter, the constitutional lawyer and famous opponent of home rule, Albert Venn Dicey, feared that satisfying the demands of the Irish pro-home rule ‘minority’ (in a UK context) fed the ascendancy of party rather than parliamentary sovereignty, flying in the face of Benthamite ideas of the common good (Conti, 2023, pp. xxv–xxix; Dicey, op cit). Allusions to home rule in the context of imperial federation or ‘home rule all around’ notwithstanding, Dicey was especially critical of Irish representation remaining at Westminster alongside a Dublin parliament. For the IPP, however, it was the unity of Ireland that trumped everything. Liverpool IPP MP T.P. O’Connor even expressed willingness to sacrifice democracy to help win unionists over to home rule:

To foster this idea of common nationhood above our strong Party differences, our class hatreds, and our distinctions of creeds will be the first work of Irish statesmanship; and by way of starting that work favourably and accelerating its progress, we are quite willing not merely to submit to, but even to welcome, restraints on the rights of majorities, which in ordinary circumstances we would reject (Commons, 3 June 1886, vol. 306, cols 848–849).

The Commons rejected the first home rule bill. The second, in 1893, ‘used the nomenclature of the colonies’ rather than Westminster, but still made provision for a lower house (‘Legislative Assembly’) which would comprise 103 members for five years and a ‘Legislative Council’, with 48 members on a more restricted franchise for eight-year terms (Ward, 1994, p. 67). While it managed to clear the House of Commons, Westminster’s upper house, the House of Lords, rejected it. The third home rule bill introduced by the Liberal government in 1912 followed a similar path – as in 1893, it proposed two parliamentary chambers rather than two orders (O’Donoghue, 2018, p. 44). The 164-member chamber was to be called the House of Commons, like at Westminster, but with a senate of 40 members appointed by the lord lieutenant to be designated as the second chamber (though in 1914 at committee stage the provision for the upper chamber was amended so that it would be elected after a five-year transitionary period, with the four provinces to be used as constituencies (Ward, 1994, pp. 67–69)). The Westminster parliament would retain supremacy with powers over the Crown, peace and war, military force and foreign policy. However, the power of the proposed Dublin parliament in key areas such as tax, customs and excise would be limited, and it would not be given immediate authority to oversee land acts or control policing. It would also be prevented from endowing any religion and Irish representation at Westminster would be retained in the form of 42 MPs (Jackson, 2003, pp. 109–111; O’Donoghue, 2018, p. 45; Ward, 1994, p. 69).

The home rule negotiations led by Liberal Prime Minister David Lloyd George after the 1916 Easter Rising followed from these proposals, with a home rule parliament still to be denied having power over military affairs and matters of peace and war, the postal service, or the Dublin Metropolitan Police in time of war (John Redmond to John Dillon, personal communication, 2 July 1916). Following the collapse of this scheme, some Irish home rulers looked to the dominions of South Africa and Australia or to the post-First World War peace conference (Daniel O’Leary to John Redmond, personal communication, 14 July 1917; to Matthew Keating, 18 July 1917). As political currents ran away from the IPP, ‘home rule’ would arrive in the form of the 1920 Government of Ireland Act, but this would only function in the new Northern Ireland as a bicameral legislature, with a House of Commons and Senate opened in Belfast in June 1921.

‘Nothing could be fairer than a fairly-constituted convention’: The IPP and the Irish democratic tradition

While no putative home rule chamber would ever have full sovereignty, the IPP was unquestionably in the vanguard of advancing Irish commitment to parliamentary democracy with the Westminster model as its lodestar. In the view of Maurice Manning, it can be credited with having ‘in a sense socialised the vast majority into seeing parliament as the normal forum for the working-out of democracy – in other words, created a political culture which had parliament at its centre’ (2010, pp. 15–17). Yet, before 1918 there was a growing sense among many that the IPP’s democratic credentials were weaker than its parliamentary means might suggest. The campaign for women’s suffrage had been long denied in the UK as elsewhere and an enhanced franchise was set to be introduced at the end of the war (Bingham, 2018; Gleadle, 2009). It was a time of varied strands of political thought about what a ‘new Ireland’ – whether independent, socialist, feminist or still unionist within the UK – might entail (Bourke & Gallagher, 2022). The IPP had benefitted from the democratic reforms of the 1870s and 1880s it is true, but by 1910 it possessed electoral hegemony in nationalist Ireland. Many constituencies had not been contested since the Parnell split in 1890–1891 and unless there was a unionist or schismatic, O’Brienite, challenger, MPs were often unaccustomed to fighting for their seats. According to Senia Pašeta, the IPP failed to ‘tolerate, let alone welcome, politically active women’ (2013, pp. 63–64). It did not take the side of workers in the 1913 Lockout – but even more general was the perception of the IPP as a machine, maintaining its dominance to the detriment of real debate and open competition (McConnel, 2005).

If Dicey feared the sovereignty of party over parliament, opponents of the IPP came to fear the dominance of party over the people. Nor did the IPP have a monopoly on democratic language. As Reid has demonstrated, the language of democracy was a key part of debates on all sides of the Irish question from 1912, as unionists argued for their right to remain British citizens within the UK. While Dicey had attacked home rule from a UK-wide perspective, Irish unionists articulated their rights with a different frame of reference – a northern focus in ‘an attempt to reformulate the unionist position to suit the wider Edwardian political zeitgeist’ (Reid, 2017, p. 214). The IPP had to argue its democratic case for home rule across Ireland with respect for minorities against Carson’s Ulster strategy, while outside parliament, the rise of the Ulster Volunteer Force and later the Irish Volunteers meant that parliamentary tactics struggled to maintain pre-eminence over the threat of political violence (Bowman, 2007). While John Redmond had skilfully negotiated the two general elections of 1910 so as to hold the balance of power at Westminster and leverage the third home rule bill from the Liberal government, the outbreak of war in 1914 also meant the suspension of what might be termed ‘normal’ electoral politics. No general election was held for eight years though the party machine remained largely healthy in the years up to the 1916 Easter Rising (Bew, 1994).

However, the rivalries within the IPP betrayed tensions – and to outsiders only reinforced the sense of a closed shop and top-down dominance, even if Redmond never exerted the kind of control over the grassroots that Parnell enjoyed (McConnel, 2013). While sympathetic newspaper editors insisted as late as 1916 that nothing ‘could be fairer than a fairly-constituted convention’ (Enniscorthy Guardian, 4 March 1916), the hostility to the Healyite Bernard Hamill in Louth for opposing the favoured IPP candidate Patrick Whitty in the ‘unnecessary’ North Louth byelection did little for the IPP’s democratic credentials (Dundalk Democrat, 5 February 1916). While the Easter Rising may not have had a popular mandate, IPP hegemony began to fray very quickly from 1917. The 1918 election in general was not a cure for all the ills of its old convention system. Many (though fewer) constituencies were still uncontested – either due to IPP disillusionment or fear of intimidation from republicans. In the midst of the influenza pandemic, voter turnout at 68% in contested constituencies was also not as high as it would be in the Irish Free State in 1932 and 1933 (77% and 81% respectively) despite the fact that women over the age of 30 and men over the age of 21 had gained the right to vote (Callinan, 2018; Gallagher, 1993, pp. 115, 147, 176). Yet, there is no doubt 1918 saw an enormous increase in democratic participation, as the electorate had almost tripled and new voters had gone to the polls. The South Donegal IPP MP J.G. Swift MacNeill had not faced a contest in 23 years and withdrew, reflecting that he had ‘too long been a member for this constituency to be able consistently with sincerity or self-respect to solicit as a favour votes to secure my election’ (Ferriter, 2004, p. 184).

After 1918 Sinn Féin clearly placed its own de facto parliament, Dáil Éireann, at the heart of its counter-state. It eschewed Westminster, but the Dáil convened (when possible, with members in prison and on the run) from 21 January 1919. Despite its difficulties, the Dáil certainly preserved in the circumstances of guerrilla war and martial law the basic concept of responsible government. It passed the Dáil constitution, appointed a cabinet in which ministers were given responsibility, and the Dáil courts were later established (Farrell, 1969; Kotsonouris, 2020). On the other hand, the circumstances meant that communication between cabinet and parliament was not always smooth or unproblematic. The IPP’s withdrawal also meant that the 1921 elections for ‘southern Ireland’, which were used to constitute the Second Dáil, were an effective fait accompli in almost all constituencies (Laffan, 1999, p. 339). Yet, it was this chamber which would debate and approve the Anglo-Irish Treaty between December 1921 and January 1922.

Kissane has described 1922 as a 'new beginning' and noted nationalist critiques of the home rule constitutional tradition, but the revolutionary parliament established between 1919 and 1922 undoubtedly bore echoes of Westminster and helped set the template for the Irish state with a dominant lower house and standing orders similar to the British parliament (Kissane, 2011; Ward, 1994, pp. 159–160). Yet, while assessing the pre-1922 period requires scrutiny of the ideas of sovereignty, parliament and democracy and how they informed constitutional developments, the writing of the Free State constitution was clearly crucial to the construction of the independent Irish state. The drafting process and how these articles were interpreted and managed by political leaders therefore demands closer scrutiny.

The Constitution Committee, 1922 and popular sovereignty

While the Anglo-Irish Treaty prescribed the constitutional relationship of the Irish Free State with Britain (and remained key in British constitutional theory (Coffey, 2012)), Michael Collins, as Chairman of the Provisional Government, appointed the Constitution Committee in January 1922. It determined in large measure the form of government for the new Irish state. It drew on the British tradition, but there were novel departures: judicial review, provision for external ministers, and creative opportunities to amend the constitution by popular referendum and introduce legislation by popular initiative. In addition, although proposed by the committee, an article requiring a referendum on declaring war was not included. Nor was the proposed requirement that government-initiated amendments be subject to popular vote. Gerard Hogan noted that the Irish judiciary found review to be an alien concept in practice and rarely engaged in the process, whether out of suspicion, or fear, that political circumstances were too sensitive (Hogan, 2012 pp. 3–4). The second novelty, external ministers, fell away quickly from 1925, as Cahillane has illustrated, with the modifications in the applicable process limiting the original scope of the mechanism to act as a check on the party system (2016, chapter 8). The instruments of referendum and initiative are, however, most instructive on the relationship of the people to the constitution in the lifetime of the Irish Free State. Likewise, the fate of the popular veto on war declarations and government amendments suggests a preference amongst politicians for the prevalence of parliament.

All the latter instruments and proposals of popular sovereignty emanated from the Constitution Committee’s deliberations, which resulted in three separate drafts (A, B and C). The Provisional Government favoured B, which was in many ways similar to A, but very different to C; though all were infused with popular sovereignty. Given the eclectic composition of its non-party membership, it is hardly surprising the committee veered from party tradition in some of its proposals and Kohn noted of the final document: ‘its model is to be found less in the older American, Australian and Swiss precedents than in the post-War Constitutions of the new Continental Republics’ (Kohn, 1932, p. 238). The committee consisted of five lawyers – Hugh Kennedy, James Murnaghan, John O’Byrne, Clement France and Kevin O’Shiel (the first three of whom were subsequently supreme court judges); one serving Dáil deputy and Chairman of the Provisional Government, Michael Collins; a writer heavily connected with the revolutionary government, Darrell Figgis; an academic co-opted after the inaugural meeting, Alfred O’Rahilly; a businessman, James Douglas; and James McNeill, formerly a senior official in the Indian civil service and a future Governor-General of the Free State (Farrell, 1970, pp. 117–119).

The committee’s inaugural meeting was held on 24 January, and the drafting was rushed to allow the Provisional Government to participate in the League of Nations conference in Geneva – with British permission to attend contingent on delivering a draft constitution. The committee met 27 times up to 7 March (Cahillane, 2016, p. 9; Farrell, 1970, pp. 119–120). Collins chaired the first meeting but only attended one more. Deputy Chairman and Secretary to the Committee was Figgis, a surprise appointee of Arthur Griffith disliked by Collins and something of a loose cannon in Irish politics. He was instrumental in producing Draft A, which was signed by McNeill and O’Byrne. Kennedy, a prominent lawyer and adviser to the government – not yet Law Officer (Attorney General) but quickly to assume that role – became Chief Justice in 1926. Murnaghan, an academic and lawyer, was appointed to the Supreme Court in 1925. O’Byrne, a lawyer who worked closely with Murnaghan on drafting articles concerning the judiciary, succeeded Kennedy as Attorney General and later served as a judge of both the High and Supreme Courts. O’Shiel, a lawyer heavily involved with the Dáil courts, did not participate sufficiently to warrant signing any draft. Clement France, an American lawyer in Ireland on White Cross business, was recommended by Douglas from his dealings with that body. France proved productive but embarrassing as he assumed a much broader but unwarranted role in Irish affairs, and his return to the United States was a relief. Douglas, a Quaker and humanitarian businessman, and friendly acquaintance of Collins’s, brought more balance as all the others, bar Figgis and France, were Catholic. He was heavily involved in the drafting, and largely responsible for ensuring Draft B, co-signed by Kennedy and France, was adopted. O’Rahilly, a polymath but also self-declared expert on most things, brought original if singular views not widely shared (though Murnaghan co-signed his Draft C) (Cahillane, 2016, p. 14; Farrell, 1970, pp. 117–119; Gaughan, 1989, pp. 151–154).

Ideas of sovereignty informed many of the views of the Constitution Committee’s members. Collins’s advice to it was to assume sovereignty as a given and ‘that [the] constitution should contain only what is necessary for constituting the machinery of our government following upon the declaration that authority is derived from the people’ (Gaughan, 1998, p. 85). The committee enshrined this with Article 1 in both drafts A and B, asserting: ‘The Nation’s sovereignty extends not only to all men and women of the nation, but to all the material possessions of the nation’ and continuing in language drawn equally from the 1916 Proclamation and the 1919 Democratic Programme of the First Dáil. O’Rahilly’s Draft C, in contrast, was shorter and much closer to that text eventually agreed. He proposed that ‘legislative power resides directly in the people of Ireland and through them in Congress’ (Cahillane, 2016, p. 225). The article eventually adopted read: ‘All powers of government and all authority, legislative, executive and judicial, in Ireland are derived from the people of Ireland and shall be exercised … through the organisations established by … this Constitution’ (Cahillane, 2016, p. 186).

After about one month, Figgis, who was already widely disliked, attempted a solo run and presented the committee with a draft constitution. This resulted in tensions and three separate drafts, with O’Rahilly objecting that from Figgis ‘he had received a translation into English of the German Weimar Constitution’ (Gaughan, 1998, p. 80). Figgis’s draft was partly modelled on the Weimar constitution but with significant Swiss influences, including proposals for referendum and initiative. It also envisaged that all amendments to the constitution from the outset would require plebiscites, as well as popular votes on declarations of war. Contemporary opinions on the instruments differed. A young Harvard-based economist, William Rappard, felt the Swiss model was a favourable one and staunchly defended it:

The initiative and referendum have sometimes been accused of making party government impossible. This criticism, which would perhaps more justly apply to proportional representation, another novel electoral scheme which is making rapid progress in Switzerland, is not borne out by Swiss experience. All that can be said, is that popular votes have somewhat strengthened the influence and self-confidence of minority parties (Rappard, 1912, p. 364).

However, an older Harvard-educated commentator and former literary editor of the New York Sun, Mayo Hazeltine (Ware, 1912, p. 36), was quite dismissive:

So much for the Swiss Referendum, which, of course, has a purely negative effect, merely enabling the people of Switzerland to reject measures passed by their representatives in the Federal or Cantonal Legislatures … The conception of the Initiative may be bold, but those who have observed the institution longest and studied it most carefully pronounce it unlikely to be of any great use to mankind (Hazeltine, 1907, pp. 211–213).

Despite other tensions, Figgis’s draft articles 48 and 49 concerned the referendum and initiative and were supported by the majority of the committee, becoming the text approved by the Provisional Government in Draft B. Draft Article 48 read that, excluding legislation covering taxation, money, health and public safety:

Any law deemed to be passed by the Oireachtas may be suspended for a period of ninety days on the written demand of two-fifths of the members of Dáil Éireann or of a majority of the members of Seanad Éireann presented to the Uachtarán. Such a law shall be submitted to the people if demanded before the expiration of ninety days by petition signed by not less than 30,000 voters.

Draft Article 49 read that:

After … five years … the Oireachtas may provide for the initiation by the people proposals for laws or constitutional amendments … [and if not] it shall on the petition of not less than 50,000 voters on the register of whom not more than 20,000 shall be in one constituency, either make such provision or submit the question to the people for decision in accordance with the ordinary regulations governing the Referendum. Any legislation passed by the Oireachtas providing for such initiation for the people shall provide: (1) that such proposals may be initiated on a petition of fifty thousand voters on the register; (2) that if the Oireachtas rejects a proposal so initiated it shall be submitted to the people in accordance with the ordinary regulations governing the Referendum; (3) that if the Oireachtas enacts a proposal so initiated such enactments shall be subject to the provisions respecting ordinary legislation (Cahillane, 2016, pp. 203–204).

While the Provisional Government adopted these articles, they were different to O’Rahilly’s Draft C, which had interesting suggestions. O’Rahilly had been exploring sovereignty for several years (O’Rahilly, 1961, pp. 52–53). The part Stonyhurst-educated Jesuit novitiate was doubtless prompted by the mid-First World War prediction of the Jesuit philosopher, Lewis Watt. In an article in Studies, the Jesuit journal co-founded by O’Rahilly in 1912, Watt surmised:

after the war we shall have a period of destruction and construction in political thought … In view of such an eventuality, it is well that Catholics should be ready to explain what their traditional philosophy has to say about society and the State. In this article one problem only will be discussed – that of the origin of sovereignty (Watt, 1916, p. 188)

On this basis, O’Rahilly teased out the origins of sovereignty of the people; and satisfied that it did not spring from the Reformation, nor was it tainted by association with the French Revolution, he embraced it fully (O’Rahilly, 1921a; O’Rahilly, 1921b) and it informed his drafting. Despite his protestations about Figgis’s translation of the Weimar document, O’Rahilly based his proposals on the Swiss model; although he considered reducing the three-month delay that the Swiss facilitated in order to organise a petition by 50,000 voters to 75 days (Footnotes to the Draft Constitution of 1922, p. 3). However, he later favoured three months and 100,000 petitioners, in effect different in form (if not greatly different in substance) from Draft A, but most likely only to be different from Figgis’s draft. O’Rahilly also embraced the initiative with proposals which envisaged 80,000 voters demanding general legislation which ‘Congress’ (his designation of both houses of parliament) could enact or, if it rejected, the matter be submitted to the people, and if a majority voted in favour, Congress would bring forward legislation. Also, 80,000 voters could propose a specific bill which, if rejected by Congress, would be put to the people; though not before Congress had the opportunity to put an alternative bill before the people for majority vote (Cahillane, 2016, p. 231). From his notes, it is clear that O’Rahilly was influenced by Michael Cronin, Professor of Ethics and Politics in University College Dublin (Cronin, 1921, p. 343), whom he (mis)quoted. Cronin, in O’Rahilly’s citation, held ‘the initiative is a powerful instrument in the hands of the voters as against an intriguing, and more particularly against a designedly inactive, parliament’ [sic] (Footnotes to the Draft Constitution of 1922, p. 2).

The notion of popular sovereignty was taken further by Figgis who, if not quite looking for neutrality, wanted to embed in the constitution a majority popular vote on any declarations of war. While O’Rahilly, again closer to the adopted article, would have granted war and peace decisions to ‘Congress’, Figgis’s wording, which is identical in Draft B, proposed that:

Save in the case of actual invasion, Saorstát Éireann shall not declare war upon any other State or Nation or be committed to participation in any way without the assent of a majority of the voters on the register obtained on a Referendum.

Significantly, it would not be a qualified majority, such as two-thirds of votes cast, but a majority of the eligible voting population; a benchmark that would require a high turnout. This was initially accepted by the Provisional Government, but that veto was subsequently removed, and the article placed before the Dáil for ratification in September simply read:

Save in the case of actual invasion, the Irish Free State/Saorstát Eireann [sic] shall not be committed to active participation in any war without the assent of the Parliament/Oireachtas (Bill to enact Constitution of Irish Free State, 18 September, 1922).

This version was passed but at the Dáil ratification stage, Figgis, an Independent deputy for Dublin County since the June ‘Pact Election’, supported by Thomas Johnson, the Labour Party leader, moved that the original wording from the Constitution Committee be reinstated. While his amendment was comfortably defeated (42-19), the debate made some notable points. Minister for Home Affairs Kevin O’Higgins, tasked with piloting the constitution through the Dáil, explained:

We choose Parliament … because war is almost invariably a matter of sudden emergency. You have not six or eight months or years’ notice of war, but some incident occurs, some tension or strain for a short time, and the war clouds burst. Therefore, the assent of Parliament rather than referendum of the people was chosen (Dáil Éireann debate, 5 October 1922, vol. 1, no. 18).

Figgis countered with language that resonated with future generations:

The direct voice of the people … should be given in the matter. It is the people’s children who will be lost in that war; it is the people who will have to bear the privations of the war, and it is the people will ultimately have to pay the cost of that war. Therefore the decision in that case should be given by the people themselves.

Johnson supported Figgis, warning against being drawn into a war by parliament, one instigated by foreign countries – ‘more likely’ Britain – and predicted: ‘this Parliament will probably change its character. We need not have very long memories to know how the elected of the people a very few years ago enthusiastically backed a war overseas. That might … very easily happen again.’ O’Higgins pointed out some practicalities but carefully avoided the realities of articles 6–9 of the Treaty (granting Britain access to specified assets ‘in time of war’): ‘We must not put the Irish Parliament of the future in a position where the enemy – or even a potential enemy – could be using its home waters and harbours because the votes were not yet counted, and give them another two or three days because we have still to count the votes’. However, Britain’s ‘in time of war’ access to facilities meant that irrespective of what declarations the Free State made, any war Britain engaged in would embroil it, not to mention the expectation that all dominions would come in on Britain’s side – as had all the empire in the First World War (and would indeed all the Commonwealth bar Ireland in the Second). President of the Executive Council (Prime Minister) W.T. Cosgrave spoke from lofty principles that belied his government’s future action on the referendum, initiative and the right of the people to validate a government-initiated amendment, when he said:

When Deputies speak of the people having to be consulted they apparently forget that the Parliament in this case is in very close association with the people … and are in a position to feel exactly the pulse of the people in a matter of that sort, and it is unlikely that a Parliament … devised under a Constitution so democratically fashioned, would for a moment embark on a war which would not have the support of the people (Dáil Éireann debate, 5 October 1922, vol. 1, no. 18).

That right of the people to validate government-initiated amendments seemed straightforward at the drafting stage and went with unaltered wording to the Dáil as:

Amendments of this Constitution within the terms of the Scheduled Treaty may be made by the Parliament/Oireachtas but every such amendment must be submitted to a Referendum of the people and shall not be passed unless a majority of the voters on the register record their votes and either a majority of the voters on the register or two-thirds of the votes recorded are in favour (Bill to enact Constitution of Irish Free State, 18 September, 1922).

The debate took an unexpected turn that ultimately proved disastrous for the constitution when on 5 October O’Higgins moved to tidy up ‘latent defects’ and avoid a ‘cumbrous process’ of popular referenda ‘to have to effect changes in the Constitution – changes about which there might be unanimity in the Dáil and in the Senate’. He proposed ‘every such amendment passed by the Parliament after the expiration of a period of five years from the date of the coming into operation of this Constitution must be submitted to a Referendum’. Johnson endorsed it but felt five years were not enough and suggested seven to ten; or two parliaments or eight years, the latter fully supported by Figgis. The latter formula was adopted and while none of the deputies could have foreseen how it would be used in 1929 to extend the period of amendment by parliamentary majority, it is difficult to understand why those least likely to be in power were so willing to extend the government’s control over the constitution.

Much as the Constitution Committee embraced sovereignty, it was more circumspect when it came to considerations of minority representation. Ultimately, the proposals it made for the 40-seat senate were superseded by Cosgrave’s (and O’Higgins’s) determination to stand over an agreement reached on the fringe of the Treaty negotiations between Arthur Griffith and southern unionists for a 60-seat assembly (O'Sullivan, 1940, pp. 75ff). Cosgrave read the heads of this agreement into the Dáil record (Dáil Éireann debate, 18 September 1922, vol. 1 no. 6). Southern unionists had been anxious about minority political representation since the Irish Convention (1917-1918), and expectations of minority representation were strongly etched in the 64-member Senate of Southern Ireland established under the Government of Ireland Act, 1920. And while the Constitution Committee was not influenced in any way by it, it is noteworthy that some of the configuration and composition of the 1920 senate would, ironically, be quite similar to that in de Valera’s 1937 constitution which drew partially on vocationalism. Depending on nominations of the lord lieutenant and voting inclinations of its voting constituencies, the 1920 senate would have (had it ever existed) a potentially strong minority bias with membership clearly delineated: its membership would comprise entitled officers (three – the lord chancellor, and lord mayors of Dublin and Cork; 17 nominees of the lord lieutenant representing commerce (including banking), labour and the scientific and learned professions; six elected by two church hierarchies (Catholic four, Church of Ireland two); 16 resident rate/taxpaying peers, elected by resident rate/taxpaying peers; eight resident rate/taxpaying Irish privy councillors elected by the Irish Privy Council; 14 county councillors, two from the southern Ulster counties and 4 each from the other provinces, elected by their peers (Government of Ireland Act, 1920, Schedule 2, parts I–III).

The Constitution Committee was sceptical about creating an upper house, but since Collins made it one of the only stipulations (Cahillane, 2016, p. 145), it had to be included. The committee disagreed widely, with different suggestions in the three drafts, including O’Rahilly going as far as to suggest its membership should be elected by the Dáil to avoid conflict over competing claims of representatives of the people. Draft B, initially accepted by the government but later amended, proposed a 40-member body, with terms of office for 12 years and ten members elected every three (Cahillane, 2016, pp. 25–26). In the event, although its powers were limited, it was largely dominated, as O’Sullivan (1940) illustrated, by minority membership.

The fate of popular sovereignty and the triumph of party politics

With the constitution ratified by October 1922, the Free State could congratulate itself on joining over 20 other European states that introduced postwar constitutions and secured democratic government. From a political perspective, operating under proportional representation, a single party with about 30% support nationally could form a government, either solely or in coalition. Cosgrave’s pro-Treaty Sinn Féin secured 38.5% of the vote at the Pact Election in June 1922 (Gallagher, 1979, p. 414). However, because of 36 anti-Treaty Sinn Féin abstainers, his party had an absolute majority in the Dáil, where it held 58 seats, versus Labour’s 17, the Farmers’ Party’s seven and 11 Independents. This left a potentially huge swathe of voters who might not share the government’s objectives. In the event, with post-Treaty abstention, pro-Treaty Sinn Féin/Cumann na nGaedheal dominated Dáil Éireann until 1927. But, in the event of a popular referendum, it did not necessarily have a majority, and as republican voters exercised the franchise, there was potential to undermine Cosgrave’s government. For example, as Gallagher demonstrated, at the Pact Election, Labour (21.3%) and anti-Treaty Sinn Féin (21.2%) secured more of the vote than pro-Treaty Sinn Féin (38.5%) (Gallagher, 1979, p. 415). This potential threat may not have affected the thinking of the Constitutional Committee, as Sinn Féin had not formally split when it met, and some such as Figgis and O’Rahilly openly anti-party, but it would undoubtedly have registered with Cosgrave (Kissane, 2011, p. 30).

That threat materialised in 1927 when, as detailed by O’Sullivan (1940, pp. 228–230) Fianna Fáil attempted to use the initiative to abolish the Oath of Allegiance and Fidelity. By that stage, Cosgrave’s government had begun reviewing the referendum but did not yet act on a 1926 committee recommendation to abolish it (Cahillane, 2016, p. 116). In June 1928, however, the government did move to close off numerous avenues of challenge. On 7 June, Cosgrave introduced five constitutional amendments dealing with the removal of the direct election of senators, the alteration of tenure of senators, the lowering age of eligibility for Seanad electability, and the abolition of articles 47 and 48 – the referendum and initiative (Dáil Éireann debate, 7 June 1928, vol. 24, no. 2). As de Valera had earlier agreed to a joint committee deliberating on the status of his initiative petition, he deemed the latter proposal ‘the nearest thing to sharp practice’ as the Dáil debated the first reading. While Fianna Fáil opposed it bitterly at every stage, the initial opposition of T.J. O’Connell, Johnson’s successor, gave way and Cosgrave carried the vote comfortably (64-41).

O’Neill surmised that Cosgrave’s distrust of the referendum and initiative was based on a fear that they might be used by a more populist and better organised Fianna Fáil to overturn the Treaty (O’Neill, 2000, p. 309). If so, then Cosgrave also did not have faith in the judiciary to uphold the illegality of such a move – with the 1931 Statute of Westminster still some time in the future. In fact, Cosgrave, unwittingly or otherwise, handed de Valera the very tools to dismantle the Treaty, with Constitution Amendment No. 16, 1929 which extended the power to amend the constitution by parliamentary majority for another eight years, locking the people out of adjudication on the 1922 constitution. This was a curious move as the political arithmetic after the second 1927 election saw Cosgrave propped up by informal support from the Farmers’ Party, whose leader Michael Heffernan accepted a parliamentary secretaryship (equivalent to a junior ministry) in the final Cumann na nGaedheal administration. Cosgrave clearly intended pushing the measure through rapidly, as he indicated when he introduced it on 21 November 1928, stating that he proposed ‘to take all the Stages of this Bill on this day week’ (Dáil Éireann debate, 21 November 1928, vol. 27, no. 4). He was delayed, however, until 13 March 1929 but when it was presented the opposition acquiesced with astonishing ease. De Valera, with perhaps a scent of power and a political inclination, indicated ‘we hope to see it availed of to make changes which will make this Constitution one which will be more satisfactory to the Irish people’. T.J. O’Connell also agreed, though less enthusiastically, noting ‘if the government or governments who will be in power for the next eight years are as prolific in amendments as … [those] for the past seven years, there will not be much of the original Constitution left’ (Dáil Éireann debate, 13 March 1929, vol. 28, no. 10).

A seeming inability to control public opinion, or guarantee a referendum outcome, likewise influenced de Valera and Fianna Fáil after they assumed power in March 1932. They selectively went about partial reversals; suspending the Military Tribunal set up under Article 2A, which authorities such as Hogan have shown to be contentious (2013, pp. 8–10), and the insertion of which into the constitution in 1931 they had opposed. But significantly de Valera never broached the idea of reinstating the referendum and initiative instruments. Nor did Labour, now led by William Norton, make any ‘price of power’ demand as its much-depleted parliamentary party of seven members propped up de Valera, who was five seats short of a majority. The extent of Fianna Fáil’s preference for party control was later proclaimed. It was particularly evident in the 1934 debate on the abolition of the Seanad, under the Constitution (Amendment No. 24) Bill. From the outset this was a contentious debate and reinforced the impression that Fianna Fáil preferred majority party control of both the Oireachtas and constitution. It was when the opposition rose in that debate Lemass laid bare certain realities in stark fashion when he proclaimed the ‘sovereign assembly’.

Although in May 1934 de Valera rejected a proposal from future Fine Gael Taoiseach John A. Costello to protect certain Articles of the Constitution being amended by an increasing powerful parliament without the imprimatur of a general election, ‘under pressure from the opposition’, he accepted that the Comptroller and Auditor General and the judiciary could only be removed by ‘a four-sevenths majority of the Dáil’ (Coffey, 2012, pp. 290–292; Ward, 1994, p. 237). He also set up the constitutional review committee to examine the state of fundamental rights under the now heavily amended constitution. Yet, by this point, Fianna Fáil, the ‘slightly constitutional’ party of 1929, had begun to find the constitution convenient for governing. More worryingly, Lemass had further contended that the government now had powers to do the unthinkable: ‘there is nothing to prevent the Government at the present time, in so far as the Constitution permits, prolonging the life of this Parliament’ (Dáil Éireann debate, 18 April 1934, vol. 51, no. 14). The following month, de Valera disputed Fine Gael Deputy Patrick McGilligan’s claim that the Third Dáil differed to the present one having been a constituent assembly convened to draft the constitution: as was later judged by Fitzgibbon, J in Ryan v Lennon. He implied that this interpretation would mean that all parliaments in the earlier period had been sovereign – a position he likened to the instability he saw as existing in states with unwritten constitutions (Dáil Éireann debate, 17 May 1934, vol. 52, no. 9). Yet this debate was before the Ryan v Lennon case. While a government memorandum actually put forward an argument that the Oireachtas was indeed a ‘perpetual constituent assembly’ (Coffey, 2012, p. 299), the Supreme Court found that in Irish constitutional theory, the Third Dáil was a constituent assembly with greater status than subsequent Dála. It was therefore questionable if later legislation ‘repugnant’ to the Treaty was then legally sound. As Coffey has observed, this ‘placed the entire Fianna Fáil constitutional project in jeopardy’ (2012, p. 302) and de Valera began on the path towards writing a new constitution – one which would, in time, establish on firm grounds the referendum as a mechanism for constitutional amendment.

Before this, much as Lemass proclaimed the potential for it, and the Blueshirt ‘march’ of 1933 notwithstanding, there was no lurch to abuse of power or serious risk of dictatorship in the Free State – unlike in Weimar Germany on which it was partially modelled. Whereas with the increasing erosion of the centre ground even before the Nazi takeover, by 1932 the Weimar Republic was run by presidential decree, the Irish Free State had effectively been run by government decree since 1922. But there the similarities ended. As J.J. Lee noted, ‘Nothing so became Cosgrave in office as his manner of leaving it’ (1989, p. 175). Irish political traditions had developed as part of evolving British democracy throughout the nineteenth century, one in which the ‘political party’ dominated; however, in much Irish political discourse, legitimacy rested on respect for public opinion. The landmark election victory of Sinn Féin in 1918 implied a democratic mandate for those from that political tradition; the party had campaigned on a poll they had heralded at the time as ‘Ireland’s Independence Day’ (O’Donoghue, 2019). Even as Cosgrave’s government effectively overrode fundamental rights with the introduction of the Military Tribunal under Article 2A, the extent to which it traduced them might be considered restrained. Ó Longaigh has documented the conviction rate of the tribunal (2006, pp. 294–299). In 1931, on its introduction, 38 IRA members were tried, with five found either not guilty or not prosecuted; 14 receiving custodial sentences of two years or less; and two sentenced to more than five years. Under de Valera, the pattern was similar. Of the 483 IRA members brought before it between 1931 and 1937, 19 were nolle prosequi, 42 were found not guilty and four were sentenced to five years or more. A similar pattern was also observed in the tribunal’s treatment of the Blueshirts from 1933 to 1935. Some 494 members were tried, with 14 nolle prosequi, 49 found not guilty and none receiving sentences greater than five years.

Conclusion

It is ironic that fear of party dominance in 1922 had some echo of Dicey, who, interested in the Swiss model, was persuaded a referendum should be employed before home rule was introduced by a parliament dominated by the Liberals and the IPP (McHarg, 2023). Perhaps this is not so surprising; both unionists and Sinn Féin had been enemies of the IPP before 1918. If unionists remaining in the Free State had to be satisfied with the Seanad to guarantee minority rights, the leaders of the new state drew on their political education in Sinn Féin. Cahillane has argued that ‘the anti-party theme was introduced into the Constitution in order to prevent domination by parties and to bring politics closer to the people’ – a view that chimed with criticisms of the pre-1918 status quo. Dominion models were not given as much consideration as other international models like Weimar Germany by many drafters (Cahillane, 2016, p. 152; Coffey, 2019), and regardless of any emerging view in Britain that the Crown represented the people (Cahillane), real innovations were offered in 1922 – including the powers of referendum and initiative. These were genuinely novel inclusions in a state with Westminster and Common Law heritage.

In Diarmaid Ferriter’s view, however, independence brought ‘an excessively centralised state, weak local government, and national parliamentarians focused on local constituency work to secure their re-election’ (Ferriter, 2016, p. 171). While the Free State administration’s desire for weak local government may be traced to disillusionment with how Redmondites operated councils since 1899 (Daly, 1997), despite the potentially radical clauses inserted by the constitution drafters, the other features of the state suggested that in practice parties after 1922 behaved like the IPP had at Westminster. The IPP deserves credit for establishing a set of norms around the primacy of parliament even while direct action around land and other forms of political activism prospered, but the suspicions of party expressed in 1922 may not have been assuaged by the development of a system with so many echoes of the Westminster model. By December 1933, even Michael Tierney, a former Cumann na nGaedheal TD and Professor of Greek at UCD, lamented that proportional representation had not reduced the influence of parties and complained that parliament had, despite all efforts to the contrary, been given too much power by the 1922 constitution. It was, in his words, ‘the father of red tape, the shield of anonymous bureaucracy’ (Tierney, 1933, p. 558).

Before Fianna Fáil took its seats in 1927, Cumann na nGaedheal TDs had not been under serious pressure to ‘sit, act and vote’ as Irish MPs had at Westminster. Conversely, Cosgrave and his colleagues developed a strong executive resembling the British model. Fianna Fáil brought more intense parliamentary competition – and greater attention to constituency detail than its competitors but was not so different in other ways. Fianna Fáil instituted strict discipline and a strong whip system behind de Valera as parliamentary party leader and, later, leader of an executive which showed little inclination to operate the 1922 constitution in its original form. As the practice of parliament in the Free State took shape alongside de Valera’s constitutional changes, it was little wonder that by April 1934, Lemass was happy to reflect the Dáil was a ‘sovereign assembly’. The precedent for popular sovereignty clearly did not entirely disappear, though; few countries are so regularly involved in popular democratic constitution-making as Ireland is today and has been over the last three decades, exhibiting a propensity for an average of a referendum a year. In 1937 de Valera alluded: ‘we have used, and propose to use in the new Constitution, methods which we have used here and found satisfactory in the past. I think there is wisdom in that’ (Dáil Éireann debate, 11 May 1937, vol. 67, no. 1). Popular sovereignty was not subsumed into parliamentary sovereignty over the century of the Irish state.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Andrew McCarthy

Andrew J. McCarthy is a lecturer in History in Cork, teaches and has published on modern Irish history, including the 1937 constitution, and is editor of the Journal of the Cork Historical and Archaeological Society.

Martin O’Donoghue

Dr Martin O'Donoghue is a Postdoctoral Research Fellow at the Max Planck Institute for Legal History and Legal Theory in Frankfurt. He is the author of The Legacy of the Irish Parliamentary Party in Independent Ireland, 1922–1949 (Liverpool, 2019) and has previously taught history at the University of Sheffield, Northumbria University and the University of Limerick.