Freely Available British and Irish Public Legal Information
[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
Supreme Court of Ireland Decisions
You are here: BAILII >> Databases >> Supreme Court of Ireland Decisions >> Heneghan v Minister for Housing, Planning & Local Government & ors (Approved) [2023] IESC 7 (31 March 2023)
URL: http://www.bailii.org/ie/cases/IESC/2023/2023IESC7(MurrayJ).html
[New search] [Printable PDF version] [Help]
AN CHÚIRT UACHTARACH
THE SUPREME COURT
Appeal No.:2022/000006
O’Donnell C.J.
Dunne J.
O’Malley J.
Baker J.
Murray J.
Between/
TOMÁS HENEGHAN
Applicant/Appellant
AND
THE MINISTER FOR HOUSING, PLANNING AND LOCAL GOVERNMENT
THE GOVERNMENT OF IRELAND
THE ATTORNEY GENERAL
IRELAND
Respondents
Judgment of Mr. Justice Brian Murray delivered 31st day of March 2023
I BACKGROUND
Introduction
1. This is an important and difficult case. It arises from the applicant’s claim that Seanad Éireann has since 1979, or shortly thereafter, been composed in violation of the Constitution. The provision of the Constitution which he alleges has been thus breached - Article 18.4.2 - was inserted in 1979 following a referendum. At the time of that referendum, it was suggested to the People that the proposed amendment would, if adopted, result in a change in the electorate for the election of certain members of the Seanad. The anticipated change never occurred. Instead, the Seanad has been elected based on the provisions of an Act - the Seanad Electoral (University Members) Act, 1937 (‘the 1937 Act’) - that was passed forty-two years before the Constitution was amended. Everything has proceeded since as if nothing had changed.
2. The respondents say that nothing had to change. They contend that the effect of Article 18.4.2 was to afford the Oireachtas a power to change the electorate. It was, on this basis, a matter for the Oireachtas to decide if, when and how that electorate might be reconstituted. The applicant says that this is wrong, and that the effect of this Article, when properly read and understood, was to impose a duty on the Oireachtas to act. Because the Oireachtas failed to act in discharge of that obligation - the applicant argues - the relevant provisions of the 1937 Act are unconstitutional.
3. Hogan J. in his concurring judgment describes the drafting of those parts of Article 18.4 that were inserted in 1979 as ‘hapless, incoherent and confused.’ I fully agree. As a result, the Court must in this case confront fundamental issues around how it should interpret an opaque constitutional provision of this kind and, in particular, as to how it should negotiate the relationship between the literal meaning of words used in that provision, other constitutional Articles, principles of equality said by the applicant to inhere in the democratic process, and evidence of what is said to have been the understanding of the People at the time of the adoption of a constitutional Article as to the meaning and effect of that provision.
4. In the light of my consideration of those questions in this judgment, and what I have decided is in consequence the correct construction of Article 18.4.2, I have concluded as follows:
(i) The 1937 Act provided for the election of three members of Seanad Éireann by graduates of the National University of Ireland (‘NUI’) and three by graduates of the University of Dublin (‘TCD’). Article 18.4.2 of the Constitution mandated the introduction by the Oireachtas of legislation to change this.
(ii) The Oireachtas was given a very broad discretion as to how it went about this task. However, at a minimum that reconstitution of the Seanad electorate had to result in the inclusion in that electorate of one or more institutions of higher education other than (but in addition to) NUI or TCD.
(iii) The options available to the Oireachtas in so legislating thus were and are to ensure that the franchise is vested in NUI and TCD and one or more other institutions of higher education, although it would also be open to the Oireachtas to vest the franchise in NUI and one or more other such institutions, or in TCD and one or more other such institutions (the latter two options, in particular, being necessary in the event of the dissolution of either NUI or TCD).
(iv) Sections 6 and 7 of the 1937 Act, in limiting that franchise to TCD and NUI, are invalid having regard to the provisions of the Constitution because Article 18.4.2 requires the reconfiguration of that electorate in one or other of the ways I have just described.
5. As I explain in the final part of this judgment, this is not the first occasion on which a court vested with a constitutional jurisdiction has concluded that legislation governing the composition of a legislative assembly is in violation of applicable constitutional provisions. When this has happened in other jurisdictions, the resulting declarations have been wholly prospective in effect, and have been suspended for such time as is reasonably required to enable laws to be formulated and passed so as to properly constitute the legislature. Apart from everything else, without putting such a declaration in temporary abeyance the problem identified in this judgment could never be solved. The declaratory relief that follows from my conclusions must be similarly suspended.
The applicant’s claim
6. Article 18.4 of the Constitution, as originally enacted by the People on 1 July 1937, provided that three of the sixty members of Seanad Éireann would be elected by TCD, and that three would be elected by NUI. The system for electing candidates to these university seats was specified in the 1937 Act. As I have just noted, the 1937 Act vested the franchise in graduates of these institutions.
7. The Seventh Amendment to the Constitution (Election of Members of Seanad Éireann by Institutions of Higher Education) Act, 1979 came into force in 1979 following a referendum held in July of that year. It resulted in the addition of new provisions to Article 18 (now Articles 18.4.2 and 18.4.3). The effect of these changes was to provide for the enactment of laws allowing for the extension of the university franchise to institutions of higher education in the State other than TCD and NUI.
8. As I have also noted, no step has been taken by the Oireachtas to give effect to this provision. The 1937 Act continues to govern the relevant franchise. Graduates of institutions of higher education in the State other than TCD and NUI have no entitlement to vote for candidates in the university panel in elections to Seanad Éireann.
9. In these proceedings, the applicant (a campaigner for reform of the Seanad who holds the degrees of Bachelor of Arts and Master of Arts in Journalism from the University of Limerick (‘UL’)) challenged on a number of different grounds the constitutional validity and compatibility with the European Convention on Human Rights (‘ECHR’) of various aspects of the process for the election of members of the Seanad. He brought the proceedings following a decision of the first named respondent (in December 2019) refusing his request (of September 2019) that he be registered to vote in the next general election for seats in Seanad Éireann on what he described as ‘the institutions of higher education’ panel. Although not relevant to this appeal, he made a similar request with respect to other panels for election to Seanad Éireann (‘the vocational panels’).
10. One of the claims he made in his proceedings was that he was entitled as a matter of law to be an elector in respect of the university panels. He said that the text of Article 18.4.2 of the Constitution - construed either alone or in conjunction with a principle which he describes as that of ‘equality in the electoral process’ - imposed a duty on the Oireachtas to legislate so as to extend the Seanad university franchise to include, at least, graduates of UL. He contended that the failure to facilitate his registration as an elector in respect of that panel was thus unlawful and in breach of the Constitution, as well as being a violation of the ECHR. On this basis, he sought a declaration that ss. 6 and 7 of the 1937 Act are invalid having regard to the provisions of the Constitution and/or are incompatible with the State’s obligations under the provisions of the ECHR. He also claimed a declaration that in not including UL as a constituency in s. 6 of the 1937 Act, the respondents have failed in their obligation under Article 40.1 of the Constitution to hold the applicant, as a graduate of that University, equal before the law and to respect and vindicate his personal rights as provided for in Article 40.3.2 thereof.
11. This claim (together with complaints as to other aspects of the Senate electoral system that are no longer being pursued [1]) was dismissed following a hearing before a Divisional High Court (Irvine P., Simons and O’Moore JJ.) ([2021] IEHC 716, [2022] 1 ILRM 237). The applicant thereafter applied for leave to appeal to this Court against that part of the High Court decision dismissing his challenge to the constitutional validity and compatibility with the ECHR of the exclusion of graduates of institutions of higher education other than TCD and NUI from participation in elections to Seanad Éireann. He did not seek to appeal the dismissal of his proceedings insofar as it was based upon his exclusion from the electorate for the vocational panels. The reasons for the grant of his application for leave to appeal appear at [2022] IESCDET 70.
The original constitutional provisions
12. Article 18 of Bunreacht na hÉireann provides for an Upper House of the Oireachtas, Seanad Éireann, to be composed of sixty members. Eleven members of the Seanad are nominated by the incoming Taoiseach after a general election. The remaining forty-nine senators are elected members of the Seanad. As originally enacted, Article 18.4 divided these elected members into three groups:
‘4. The elected members of Seanad Éireann shall be elected as follows:—
i. Three shall be elected by the National University of Ireland.
ii. Three shall be elected by the University of Dublin.
iii. Forty-three shall be elected from panels of candidates constituted as hereinafter provided.’
13. Article 18.5 is to the effect that every election of the elected members of Seanad Éireann shall be held on the system of proportional representation by means of the single transferable vote, and by secret postal ballot. Article 18.6 provides that the members of Seanad Éireann to be elected by ‘the Universities’ shall be elected on a franchise and in the manner to be provided by law.
14. Article 18.4 was implemented by the 1937 Act. As I have noted, ss. 6(1) and (2) of that Act provide that NUI and TCD shall each be constituencies for the election of three members of Seanad Éireann, while the effect of ss. 7(1) and (2) (as amended) is that every person who is a graduate of these institutions (together, in the case of TCD, with certain scholars of that college) and is both a citizen and over the age of 18, is entitled to vote in their respective constituencies.
15. Article 18.7 prescribes how the remaining forty-three members referred to in Article 18.4 (iii) are to be elected. It states that five panels of candidates (the vocational panels to which I have earlier referred) shall be formed in the manner provided for by law containing respectively the names of persons having knowledge and practical experience of five ‘interests and services’:
‘(i) National Language and Culture, Literature, Art, Education and such professional interests as may be defined by law for the purpose of this panel;
(ii) Agriculture and allied interests, and Fisheries;
(iii) Labour, whether organised or unorganised;
(iv) Industry and Commerce, including banking, finance, accountancy, engineering and architecture;
(v) Public Administration and social services, including voluntary social activities.’
16. As stated in Article 18.7.2, subject to Article 19, not more than eleven and not less than five members of Seanad Éireann shall be elected from any one panel. Article 19 is as follows:
‘Provision may be made by law for the direct election by any functional or vocational group or association or council of so many members of Seanad Éireann as may be fixed by such law in substitution for an equal number of the members to be elected from the corresponding panels of candidates constituted under Article 18 of this Constitution.’
17. Professor Ó Cearúil in his authoritative analysis (‘Bunreacht na hÉireann: A study of the Irish text’ (1999) (at p. 301), literally translates the Irish text [2] of Article 19 as follows:
‘Provision may be made/It will be possible to make provision by law so that any functional or vocational body or council, would be able to elect directly as many members to Seanad Éireann as will be determined by law, in place of the same amount of the members who will be elected from the co-rolls of candidates which will be arranged under Article 18 of the Constitution.’
18. The Seanad Electoral (Panel Members) Act, 1947, as amended, makes provision for nominating bodies and the electorate for the forty-three Seanad seats. The electorate is comprised of incoming members of Dáil Éireann, outgoing members of Seanad Éireann and members of local authorities. As of the last election, the electorate for those forty-three seats consisted of 1,169 voters. The electorate for the six university seats comprised approximately 177,000 persons (approximately 112,000 NUI and 65,000 TCD).
The Seventh Amendment
19. The referendum on the Seventh Amendment to the Constitution was held on 5 July 1979. It is common case that at least one of the reasons for bringing forward proposals for this Amendment was to enable a then current proposal to dissolve NUI and to convert the constituent colleges of NUI into independent universities. That said, reform of the university constituencies had been suggested some time before this: in its important 1967 Report, the Committee on the Constitution (Pr. 9817) had recommended a change to these constituencies, suggesting that the six representatives ‘shall be elected from the Irish Universities in a manner prescribed by law without any specific reference to particular Universities’ (para. 82). [3]
20. As a consequence of the Seventh Amendment, the previous Article 18.4 was renumbered Article 18.4.1, while the new Articles 18.4.2 and 18.4.3 were inserted. The result was that Article 18.4 now reads as follows:
‘1° The elected members of Seanad Éireann shall be elected as follows:—
i. Three shall be elected by the National University of Ireland.
ii. Three shall be elected by the University of Dublin.
iii. Forty-three shall be elected from panels of candidates constituted as hereinafter provided.
2° Provision may be made by law for the election, on a franchise and in the manner to be provided by law, by one or more of the following institutions, namely:
i. the universities [4] mentioned in subsection 1° of this section,
ii. any other institutions [5] of higher education in the State
of so many members of Seanad Éireann as may be fixed by law in substitution for an equal number of the members to be elected pursuant to paragraphs i and ii of the said subsection 1°.
A member or members of Seanad Éireann may be elected under this subsection by institutions grouped together or by a single institution.
3° Nothing in this Article shall be invoked to prohibit the dissolution by law of a university mentioned in subsection 1° of this section.’
21. It will be observed that the highlighted language used in part of Article 18.4.2 as thus inserted (the emphasis is mine), is strikingly similar to that appearing in the text of Article 19. The Irish text of this part of Article 18.4.2 differs somewhat from Irish language version of Article 19 (see Ó Cearúil at p. 301). Article 18, when translated literally from the Irish text, changes from ‘of so many members of Seanad Éireann as may be fixed by law in substitution for an equal number of the members to be elected’ to ‘that number of members of Seanad Éireann that will be provided for by law in place of an equal/equivalent number of the members who will be/are to be elected’. In Article 19, ‘of so many members of Seanad Éireann as may be fixed by such law in substitution for an equal number of the members to be elected’ becomes ‘would be able to elect directly as many members to Seanad Éireann as will be determined by that law, in place of the same amount of the members who will be elected.’
The 1937 Act
22. Section 6 of the 1937 Act, provides:
‘(1) At every Seanad election -
(a) the National University of Ireland shall be a constituency (in this Act referred to as the National University constituency) for the election of three members of Seanad Éireann, and
(b) the University of Dublin shall be a constituency (in this Act referred to as the Dublin University constituency) for the election of three members of Seanad Éireann, and
(c) every person who is for the time being registered as an elector in the register of electors for the National University constituency shall be entitled to vote in that constituency, and
(d) every person who is for the time being registered as an elector in the register of electors for the Dublin University constituency shall be entitled to vote in that constituency.
(2) No person shall be entitled to vote at an election in a university constituency unless he is registered as an elector in the register of electors for that constituency.
(3) Nothing in this section shall entitle any person to vote at an election in a university constituency while he is prohibited by law from so voting, nor shall anything in this section relieve any person from any penalties to which he may be liable for so voting.
(4) The National University constituency and the Dublin University constituency are in this Act referred to as university constituencies and the expression “university constituency” shall in this Act be construed accordingly.’
23. Section 7(1) as it now stands provides that every person who is a citizen of Ireland and has received a degree (other than an honorary degree) from NUI and has attained the age of eighteen years (the original provision, insofar as it referred to persons over the age of twenty-one was amended by the Electoral (Amendment) Act, 1973) shall be entitled to be registered as an elector in the register of electors for the NUI constituency. Section 7(2) makes similar provision in respect of graduates of TCD and the TCD constituency (although in the case of the latter it also extends the franchise to certain scholars of that university).
24. It should be said that while Article 18.4 mandated the NUI and TCD university panels thus provided for by law, the Article did not prescribe who the electors for that constituency were to be. In theory, the Oireachtas could have vested responsibility for electing candidates to those panels in, for example, the Presidents or governing bodies of the institutions in question. Indeed, having decided to vest the franchise in graduates of those institutions (as s. 7 did) there is no reason why the franchise cannot be extended to undergraduates of the institutions (provided they are of voting age, as indeed was done in the case of certain scholars of TCD), nor any particular reason why it could not have been limited, for example, to post-graduates. In fact, on one view of the text of the provision alone, there is no reason why graduates of other institutions could not in theory be allocated a vote on these panels.
25. Finally, it is to be noted that the outcome of the challenge to the constitutional validity of these provisions depends almost entirely on the meaning of Article 18.4. If that Article read as a whole enables the constituencies to be validly vested in NUI and TCD to the exclusion of all other institutions of higher education, the provisions are valid. If it requires the Oireachtas to legislate so as to expand the constituencies, they are not valid.
The respondents’ case and the High Court judgment
26. The judgment of the Divisional Court records the applicant’s case insofar as based upon the university seats, as depending on three propositions - (a) that the enactment of the Seventh Amendment meant that the current form of election was invalid, (b) that Article 18 must be read in conjunction with the equality provisions of the Constitution, and in particular Article 40.1, and (c) that Article 18.4.1 is in the nature of a transitional provision. In 1937, the applicant stressed, a decision was made to include all of the universities in the State within the franchise. He argued that in 1979 the franchise was extended to reflect the fact that there were more institutions of higher education at that time. The intention, as one of his witnesses put it, was to make matters ‘more equal’ having regard to the fact that by the time of the referendum there were many more such institutions than there had been in 1937.
27. The respondents’ case was that s. 7 of the 1937 Act could not be unconstitutional by reason of the exclusion from the Seanad electorate of graduates of third level institutions other than TCD and NUI for the simple reason that that is precisely what the Constitution itself provides. Thus, Article 18.4.1 (i) and (ii) state that three senators ‘shall’ be elected by each of these universities. Section 7 of the 1937 Act, the respondents argued, does precisely what the Constitution thus expressly envisages. Insofar as Article 18.4.2 enables the election by ‘other institutions of higher education’, the respondents said that the Constitution does not mandate this. On the contrary it grants only a power: ‘[p]rovision may be made by law…’ (emphasis added). On this argument, to read these words as imposing a mandatory obligation would be to up-end the whole structure of Article 18.
28. Before the Divisional Court, the parties called evidence addressing the historical context within which the composition of the Seanad was originally fixed, and thereafter the subject of amendment in 1979. [6] The Divisional Court began its analysis of the issues by explaining why it did not believe this evidence was of assistance to it. It recorded a number of principles of constitutional interpretation as identified in the judgment in Bacik v. An Taoiseach [2020] IEHC 313, [2021] 3 IR 283 (at para. 80) which, the Court noted, were agreed by both sides to be relevant. One of these was the following:
‘If a literal interpretation of one provision might bring it into conflict with the literal meaning of another provision, then it is legitimate to resort to the harmonious approach with a view to interpreting both provisions in a way which avoids inconsistency. In this context, while Murray C.J. in Curtin did not expressly say that the harmonious interpretation favoured by Henchy J. in O’Shea should be applied, he did not dissent from the observations of Henchy J. to that effect in the passage quoted by him. It was interesting to note that, although [Tormey v. Ireland [1985] I.R. 289] was cited by Counsel in Curtin, the judgment of the Court does not refer to it. Having regard to the emphasis placed by the Supreme Court in Curtin on the principle of the words of the Constitutional provision in issue should be the first port of call, it seems to us that the harmonious approach will only be taken in cases of apparent inconsistency. It will not be necessary to go beyond a literal interpretation of a Constitutional provision unless such an interpretation gives rise to an apparent conflict with some other provision of the Constitution.’
(Emphasis added).
29. Based on this, the Court observed (at para. 80) that it ‘will have regard to factors such as a relevant amendment or the historical context only in cases of doubt, ambiguity, inconsistency or silence’. It explained that by ‘historical context’ it meant ‘facts that may (in appropriate circumstances) help to inform the Court as to the meaning of a constitutional provision by reference to the circumstances of its enactment.’ From there, the Court expressed the view that neither in respect of the university nor the vocational panels was there the required level of doubt, ambiguity, inconsistency or silence to require consideration of historical context or the outcome of any referenda in respect of Amendments to the Constitution in relation to the Seanad. Moreover, even if the Court had come to a different view, the evidence available to it was, it believed, equivocal. The historical evidence was, therefore, viewed by the Court as both inadmissible, and of little assistance to the decisions the Court had to make (at para. 91).
30. The judgment of the Divisional Court referred to the 1979 Referendum, stressing that while the People added a provision - Article 18.4.2 - they also left in place the original express provision mandating three Senators to be elected by NUI and three Senators elected by TCD to the exclusion of any other institution of higher education in Ireland. As such, the Court said, it was difficult to overstate the significance of the plain provisions of Article 18.4. These, it was suggested, mean that the decision whether or not to enact legislation changing the composition of the university panel is exclusively within the discretion of the Oireachtas; the word ‘may’ in Article 18.4, it was said, can have no other meaning. The applicant’s case, the Court said, involved asking it to decide that a form of election in respect of these six seats which is expressly permitted by the Constitution, is itself unconstitutional.
31. The Court rejected the argument that an Amendment to the Constitution ‘permitting the Oireachtas to enact legislation requires it to do so’ and held that the Seventh Amendment to the Constitution did not place any legal obligation on the Oireachtas to exercise its power to legislate to widen the enfranchisement of the university seats. To accept the applicant’s submission as to the meaning of Article 18.4.2 would lead to a ‘direct conflict’ between that provision and 18.4.1 and would thus place two sequential provisions ‘at loggerheads’. This, the Court stated, would not be a desirable interpretation of the basic law of the State.
32. It then said this (at para. 148):
‘The Court is satisfied that the facts of this case themselves illustrate precisely why a discretion should not be turned into an obligation. There is a large number of different institutes of higher education that could be included in an expanded franchise for the six seats. There is therefore a kaleidoscopic range of different possibilities available to the Oireachtas, should it decide to expand the franchise beyond the NUI and TCD. The Oireachtas is thus said to be under a legal obligation to legislate, but no guidance is given by Counsel for Mr. Heneghan as to what level of extension of the franchise (or what level of consideration of extension of the franchise) by the Oireachtas is sufficient to discharge this alleged legal duty.’
33. The Divisional Court similarly rejected an argument advanced by the applicant to the effect that in construing the provisions of Article 18, the equality provisions of Article 40.1 should be taken into account. This argument, the Court found, would lead to a construction of the Constitution which is not harmonious, and would involve taking general principles of equality and using them to override a specific provision which permits graduates of NUI and TCD to be preferred over graduates of other higher education institutions. As a result, the Court held that it could not under any circumstances draw a ‘blue pencil through a provision of the Constitution’, especially in circumstances where the People had ratified the continued operation of the provision by enacting the Seventh Amendment. The Court specifically recorded its disagreement with the proposition advanced by the applicant that Article 18.4.1 was retained only as a transitional provision, stating that the natural reading of Article 18.4 is inconsistent with the suggestion that the first portion of it is of transitional or temporary application alone.
34. As I have noted, the applicant also sought declarations pursuant to s. 5 of the European Convention on Human Rights Act 2003 that insofar as graduates of institutions of higher learning other than TCD and NUI are excluded from the electorate for the Senate university seats, Irish law is inconsistent with the State’s obligations under the Convention.
35. The State in response argued that firstly, the 2003 Act is subordinate to the Constitution; secondly, that the Court is not obliged to interpret the Constitution in a manner compatible with the State’s obligations under the Convention; and thirdly, that the 2003 Act does not apply in as much as the 1937 Act reflects what is prescribed in the Constitution.
36. In respect of this issue, the Divisional Court stated that the State’s threshold objection to the applicant’s case was well made because:
(i) the Constitution expressly sanctions the current franchise for the election of the six senators as provided for by Article 18.4.1;
(ii) the 2003 Act cannot override the constitutional mandate that the six senators may be elected by NUI and TCD;
(iii) inasmuch as the 1937 Act is challenged, the extent of that challenge is in relation to an electorate which the Constitution itself expressly permits;
(iv) the argument that the 1937 Act must be interpreted by reference to the provisions of the 2003 Act would lead to a finding that the composition of an electorate (itself in accordance with the provisions of the Constitution) is nonetheless unlawful, would be fundamentally wrong, and would entirely ignore the basic legal reality that domestic legislation is inferior, rather than superior, to the provisions of the Constitution; and
(v) in inviting the Court to declare the relevant portions of the 1937 Act incompatible with the provisions of the Convention, the applicant was asking the Court to declare Article 18.4.1 (i) and (ii) incompatible with the provisions of the Convention. This is not what is contemplated by s. 5 of the 2003 Act, which is confined to declarations concerning statutory provisions and rules of law.
37. The Divisional Court thus found that the system for electing the six university panel senators, as provided for in the 1937 Act is not invalid having regard to the provisions of the Constitution and that the applicant was not entitled to a declaration that the relevant Irish law is inconsistent with the State’s obligations under the ECHR.
II ARTICLE 18.4, THE APPLICABLE PRINCIPLES OF INTERPRETATION, AND THE EVIDENCE
Article 18.4
38. At first blush, the Divisional Court’s interpretation of Article 18.4 may appear unexceptional. Had the People, when adopting the Seventh Amendment, wished to impose an obligation on the Oireachtas to legislate so as to expand the franchise for the university constituencies beyond TCD and NUI, they could easily have said so. Instead, the new Article 18.4.2 is framed in terms that are seemingly and - one must assume, deliberately - permissive rather than mandatory. In this regard, as I have earlier observed, they reflect in part the provisions of Article 19, which are clearly intended to be discretionary. When that choice of language is combined with the retention in Article 18.4.1 of the terms originally appearing in Article 18.4 and expressly mandating the election to the university seats by only two institutions, and when read in the light of the seemingly broad discretion vested in the Oireachtas as to which institutions would benefit from any revision of the university franchise (‘one or more of the following institutions …’) the overall effect seems, on initial consideration, clear. The university franchise is limited to TCD and NUI (as it has always been), but the Oireachtas is free in its discretion to change this so as to include other institutions, and indeed free in its discretion to decide which institutions it will so include. Whatever controversy may exist around the methodologies properly brought to bear on the interpretation of the Constitution, the analysis must always start with the words, and the words are (on this argument), if nothing else, clear. Further support for that construction can be derived from the policy decisions involved in extending the franchise as identified by the Divisional Court: the fact that in exercising the power (as it is said to be) the Oireachtas would have to make potentially difficulty choices as to which ‘institutions of higher education’ would be brought within the franchise might be thought to point towards the Constitution vesting in the Oireachtas a broad discretion as to not merely how, but also whether, the franchise should be thus extended. Any other conclusion, it could be said, would cast unacceptable doubt over the composition of a key organ of State.
39. But the provision is in no sense as simple as this suggested reading implies. To understand the interpretative issues that arise, it is helpful to quote Article 18.4.2 again:
‘Provision may be made by law for the election, on a franchise and in the manner to be provided by law, by one or more of the following institutions, namely:
i. the universities mentioned in subsection 1° of this section,
ii. any other institutions of higher education in the State
of so many members of Seanad Éireann as may be fixed by law in substitution for an equal number of the members to be elected pursuant to paragraphs i and ii of the said subsection 1°.
A member or members of Seanad Éireann may be elected under this subsection by institutions grouped together or by a single institution.’
40. The Irish text [7] is translated by Professor Ó Cearúil as follows (at p. 301):
‘Provision may be made by law so that there will be elected in accordance with an electoral system, and in the manner, (that will be) provided by law, by one or more of the following institutions, that is/namely:
i. the Universities mentioned in subsection 1 of this section,
ii. any other institutions of higher education in the State,
that number of members of Seanad Éireann that will be provided for by law in place of an equal/equivalent number of the members who will be/are to be elected pursuant to paragraphs i and ii of that subsection.
A member or members of Seanad Éireann may be elected under this subsection by institutions which will be collected together or by a single institution.’
41. A close examination of the text shows that the relationship between Article 18.4.1 and 18.4.2 cannot be understood by reference only to a literal reading of the words in each provision. This is important, as the Divisional Court attached considerable significance to the fact that the former provision was not amended in 1979 (at para. 90):
‘while the 7th amendment created the potential for the Oireachtas to widen the electorate in that regard, it nevertheless left intact the provision whereby the six university members could be elected solely by the NUI and TCD.’
42. Later in its judgment, it said the following (at para. 150):
‘Even if Mr. Heneghan’s submissions on this point are correct, and Article 18.4.2° now requires the Oireachtas to make provision by law for election on a franchise of the type contemplated by that clause, there remains the fact that Article 18.4.1° remains in situ …. on the face of it his submissions about the meaning of Article 18.4.2°, would, if correct, lead to a direct conflict between it and Article 18.4.1°. This would not involve a harmonious reading of the Constitution; rather, it would set two sequential provisions at loggerheads with each other, which is not a desirable interpretation of the basic law of the State.’
43. However, no matter how one reads them, there is a conflict between Article 18.4.1 and 18.4.2. The respondents acknowledge (but significantly understate) this when they say that the provision is ‘somewhat unusual in granting the Oireachtas the power to substantially modify the effect of a Constitutional provision through legislation’.
44. Article 18.4.1 (i) and (ii) state that three of the elected members of Seanad Éireann ‘shall’ be elected by NUI and that three ‘shall’ be elected by TCD. Article 18.4.2 provides that the Oireachtas ‘may’ by law change this. Clearly, it cannot both be the case that six Senators must be elected by TCD and NUI and that those same members can be elected by other institutions of higher education. This tension can, of course, be easily resolved by reading Article 18.4.1 (i) and (ii) (but not Article 18.4.1 (iii)) as being subject to the enactment of legislation in accordance with Article 18.4.2 and there is no doubt but that this was the intention. What is relevant is that the implementation of that obvious intention requires the implication into one or other of the provisions of words that are not, but could easily have been, incorporated into it. On either party’s construction of Article 18.4, Article 18.4.1 has been converted from a provision which mandated NUI and TCD as the relevant constituencies, to one which acknowledges at least a discretion on the part of the Oireachtas to change this. Indeed, counsel for the respondents in the course of his oral submissions described these parts of Article 18.4.1 as ‘the default position’. And, if that is so, the effect is that the ‘shall’ in Article 18.4.1 - in substance - means ‘may’.
45. When that implication is made, it is clear that Article 18.4.1 is of little assistance in resolving the issue of construction that arises here. Properly read, the Article is a conditional prescription of the university constituencies that is dependent on the Oireachtas not legislating in accordance with Article 18.4.2. If Article 18.4.2 merely confers on the Oireachtas a discretion, then the Oireachtas has the power to decide whether Article 18.4.1 continues to define the composition of the university panel. If Article 18.4.2 imposes a duty on the Oireachtas to legislate, Article 18.4.1 defines the composition of the Seanad unless and until that duty is discharged. It can certainly be said that if the latter is the case, Article 18.4.1 expresses the intention in an awkward manner, and it is not inaccurate to say that this renders the provision ‘temporary’ or ‘transitional’ insofar as the position of TCD and NUI are concerned. But even on that interpretation this can be said: Article 18.4.1 prescribes pro tempore the composition of the university panels unless and until the Oireachtas legislates. So, while the Divisional Court rejected the contention suggested by the applicant that the provision was intended to be merely transitional ‘it would have been worded very differently when the 7th Amendment was introduced’, the fact is that the provision should have been worded differently were it intended to be a conditional provision which, on any version of the text, it was.
46. This question of whether the Oireachtas is free to decline to trigger the alteration of the constituencies that is enabled by Article 18.4.2 thereby releasing the university panel from the constraint otherwise imposed by Article 18.4.1 must in the first instance be therefore resolved within the text of that provision alone. But there, again, the language is not free from doubt. Clearly, the word ‘may’ generally confers a power rather than a duty, but there is ample legal authority that in some contexts it presages a mandatory obligation. I will return to some of these cases later.
47. In deciding which of these meanings should be applied here, it is relevant that in at least one situation the Oireachtas must change the constituencies. It will be recalled that the Seventh Amendment was proposed - at least in part - for the purposes of enabling the dissolution of NUI and its replacement with several universities comprising its constituent colleges. But, following the Seventh Amendment, Article 18 made no provision for what was to happen if this very thing occurred. In that event, Article 18.4.1 would have been inoperable, as it would not have been possible to allocate the three seats provided for in Article 18.4.1 (i). The failure to make provision for what would happen in that event (which is clearly envisaged by Article 18.4.3) is one of the many surprising features of the provision.
48. There was, admittedly, a suggestion pressed rather lightly by counsel for the respondent that this eventuality might have been addressed through reliance on the general language of Article 18.6, and indeed at one point it was said that it might be possible to accommodate graduates of universities or institutions of higher education on the electorate for one of the other panels. However, the only way I can see that one could have confidently ensured that there would have continued to be forty-nine elected members of Seanad Éireann following the dissolution of NUI would have been by legislating in accordance with Article 18.4.2. So, at least in some situations, the ‘may’ in Article 18.4.2 for all practical purposes meant ‘must’, and it is of some considerable significance that the one situation in which that is clearly the case, is the specific circumstance the respondents say the Seventh Amendment was intended to address.
49. These are not the only interpretative issues arising from the Seventh Amendment. It will be seen that Article 18.4.2 introduces two separately numbered categories of institution - ‘(i) the universities mentioned in subsection 1° of this section [being, of course, TCD and NUI] … (ii) any other institutions of higher education in the State.’ These are introduced by the words ‘the following institutions’. It is entirely unclear why this was done in this way, or indeed what the effect of thus framing the provision is. It may be that - perhaps as the only universities then in existence in the State and/or because provision had already been made for them in Article 18.4.1 - TCD and NUI were being categorised as a single institution for this purpose (which would explain why they were iterated together) or it may be that they were being designated as separate institutions (which would beg the question of why the enumeration was required at all - an issue to which I later return).
50. That is a distinction with a difference. If they were categorised as a single institution the effect of the phrase ‘by one or more of the following institutions’ would appear to be that if the entitlement to elect members were vested in TCD it would also have to be vested in NUI. Given the absence of an express facility to do otherwise, it seems likely that that entitlement would have to be evenly distributed between them.
51. If, on the other hand, TCD and NUI are to be categorised as separate institutions, then the entitlement to elect members could be vested in one, but not the other, and indeed it could be allocated unevenly as between them. [8] But even then, literally construed, Article 18.4.2 would mean that the franchise in respect of all six seats could be vested in one of those two universities. More fundamentally and however it falls to be construed, without qualification all of the seats could be vested in an institution other than TCD or NUI. On that literal construction, the effect of the Seventh Amendment was to allow the Oireachtas to further concentrate the Seanad university electorate in one institution to the exclusion of all others, or to remove from the ‘university’ franchise, the only two actual universities in the State (as was then the case). All of this begs the question of whether there is an interpretation of Article 18.4.2 that would preclude some or all of these, counterintuitive, outcomes.
52. Article 18.6 in contrast may require not interpolation or interpretation, but surgery. This is also simple and uncontroversial, but it may have to be effected to render the provision consistent with Article 18.4.2. Article 18.6 was not amended in 1979. It reads thus:
‘The members of Seanad Éireann to be elected by the Universities shall be elected on a franchise and in the manner to be provided by law’.
53. This gives rise to no difficulty as long as the Oireachtas does not legislate in accordance with Article 18.4.2 (if it is permitted to do this). If the Oireachtas does legislate in accordance with that provision a question arises as to what happens to Article 18.6. It may be that it too is then rendered redundant, with the authority to determine the franchise and manner of election being governed by that part of Article 18.4.2 that so provides (‘on a franchise and in a manner determined by law’). If that is the effect, the continued operation of Article 18.6 must also be read as being subject to legislation not being adopted in accordance with Article 18.4.2.
54. However, if one assumes that each provision of the Constitution is intended to have some effect unless otherwise provided, it might be concluded that Article 18.6 governs the election of members of the Seanad by ‘the Universities’ with Article 18.4.2 enabling provision by law for elections by other institutions of higher education, or (perhaps) of any redistributed seats formerly elected by one of the universities. Or it may be that both provisions apply, with Article 18.6 being the primary provision empowering the Oireachtas to determine franchise and manner of election, and the short statement in Article 18.4.2 to which I have referred confirming this. In that event, ‘Universities’ in Article 18.6 includes not only TCD and NUI (which as enacted in 1937, and indeed in 1979, is what it clearly meant) but also any other universities subsequently established together with ‘institutions of higher education’. That exercise of interpretation would require the re-phrasing of the provision so that a word with a clear and defined legal meaning reflected in Article 18.4.1 (‘Universities’) would be interpreted as including not merely additional universities but also another category of institution (‘institutions of higher education’) which itself lacks any clear definition either in the Constitution or, for that matter, in the general law. [9] Either way, any interpretation of Article 18.6 that does not consign the entire provision to redundancy if the franchise were extended, assumes that ‘the Universities’ continue to be part of the franchise, at least (having regard to Article 18.3) for as long as they both continued to exist.
55. It is to be borne in mind that in order to maintain the forty-nine seats provided for in Article 18.1 there always had to be six ‘university’ seats. That being so, had it been the intention that legislation under Article 18.4.2 would simply provide for the filling of the six seats referred to in Article 18.4.1, this could have been easily communicated by including the following text after the words appearing in 18.4.2 (ii): ‘of six members of Seanad Éireann’. In that way, there would have been a ‘clean sweep’ with the legislation enacted under Article 18.4.2 replacing the provisions of Article 18.4.1 (i) and (ii). The issue with the text arises from the fact that the following formula was instead used:
‘of so many members of Seanad Éireann as may be fixed by law in substitution for an equal number of the members to be elected pursuant to paragraphs i and ii of the said subsection 1°.’
56. This is, on any view, a difficult way of expressing the intent. The conjunction of ‘so many ... as may be fixed’ suggests a discretion, while the phrase ‘in substitution for an equal number of’ suggests none. Here, the obverse of the reconstruction demanded by Article 18.4.1 is required: ‘may be fixed by law in substitution for an equal …’ actually means ‘shall be fixed …’.
57. This aspect of the provision brings us back both to the relationship between Article 18.4.1 and 18.4.2 and the reason the institutions are separately iterated in the latter. It may be that the language was used because it appeared in Article 19 and that it was all simply intended to communicate that the six seats provided for in Article 18.4.1 were to be elected by the institutions provided for in the law enacted pursuant to Article 18.4.2. Or it may be that the idea was that instead of a ‘clean sweep’ it was intended to allow, say, one of the three seats allocated to NUI by Article 18.4.1 and one of the three allocated to TCD to be granted to Institutions A and B and that in that situation the two seats given to A and B would be in substitution for two seats allocated in Article 18.4.1 to NUI and TCD. The provision could have been similarly used had NUI been dissolved: one additional seat might have been allocated to TCD or another institution and the two remaining seats formerly elected by NUI might have been allocated to A and B. And, noting that the final clause of Article 18.4.2 enabled the franchise for the election of a single member by a number of institutions, there are numerous complex variations on this. This might explain the otherwise roundabout language. But all of this is to some extent speculative: what is clear is that the same effect could have been achieved by far simpler and more direct language, with Article 18.4.2 allowing a straightforward redistribution or reallocation of the franchise directed by Article 18.4.1 to be shared equally between TCD and NUI. For some reason, this was not done.
58. The Court in construing any part of Article 18.4 must, clearly, strive to achieve an interpretation that resolves as many of these interpretative issues as possible in a manner that is coherent, and workable while at the same time giving effect to the intent of the People in adopting the provision. So, while it might appear that the narrow question of whether Article 18.4.2 is permissive or mandatory, and of whether the Oireachtas could, consistent with Article 18.4.1 continue to vest the Seanad university franchise in TCD and NUI, can be answered easily and by reference to the text of those provisions alone, in fact those seemingly clear questions cannot be divorced from the wholly ambiguous context in which they appear.
59. This is the confused background from which three particular and necessarily related issues around the proper approach to the construction of a provision of the Constitution emerged in argument in this case - (a) when does a close reading of the literal meaning of the words used in such a provision yield to broader considerations of context and purpose, (b) what type of extrinsic evidence (if any) can be admitted to properly interpret the text of a constitutional provision inserted following an Amendment of the Constitution and, if admitted, when and to what end it can be used as part of the process of interpretation and (c) (insofar as it might be said to be different from (a)) whether there exists a defined bedrock of constitutional principles that can be used to elucidate the meaning of individual articles and, if so, when this may be done?
Principles of Interpretation
60. These questions are framed by some familiar propositions. That the process of construing any individual provision of the Constitution begins with a consideration of the text, including the Irish language version, is uncontroversial. Similarly, there is no doubt but that an Article that lacks clarity can be construed by reference to other provisions of the document, and that gaps and silences within the text can be filled by reference not merely to other provisions, but also by taking into account some fundamental principles that have been found to underpin the Constitution. Fair procedures is the most obvious and incontrovertible of these, but the same has been said of the guarantees afforded to other fundamental rights.
61. A consideration of the authorities shows that these propositions can now be confidently supplemented. While there have inevitably been diversions along the route, the terminus of the case law today is that it can never be said simply because a provision appears clear on its face, no other Article is relevant to its construction, just as it is in error to conclude that if all provisions pertinent to an issue appear unambiguous, broader questions of purpose and context are immediately out-ruled in considering their meaning. The relevance of each of these considerations will in any given case depend on the clarity of the text, the purpose of the provision, and the relationship the Article bears to other parts of the document. But they are each, always, potentially engaged: a provision of the Constitution that, although clear on its face, is in conflict with another such provision cannot be said to have a ‘plain meaning’ and it will be necessary to at least consider whether - having regard to their respective objects - they can be reconciled. An Article that although seemingly unambiguous in its terms but which conflicts with fundamental principles underlying the Constitution, on closer analysis, bear a meaning - and require a construction - that is more consistent with those underpinning principles than that yielded by the constraints of narrow literalism. And no provision can be understood without some reference to its background and purpose: it is often said that the Constitution should not be construed as if it were a Finance Act, but as the law has evolved not even the most technical provisions in a taxing statute are interpreted today without regard to their overall context and object (see Dunnes Stores v. The Revenue Commissioners [2019] IESC 50, [2020] 3 IR 480 and Bookfinders Ltd. v. The Revenue Commissioners [2020] IESC 60).
62. It is not hard to find fragments in the case law favouring, in one situation or another, a ‘literal’ or ‘purposive’, a ‘harmonious’ or indeed ‘historical’ interpretation of particular constitutional Articles. In some cases, one of these is more appropriate than the other: as Murray C.J. said in Curtin v. Dáil Éireann [2006] IESC 14, [2006] 2 IR 556 at para. 73 (‘Curtin’) ‘different interpretative elements are emphasised in individual judgments according to the particular context in which questions arise and the particular types of interpretative problem’. It was for much this reason that in his separate judgment in Sinnott v. Minister for Education [2001] IESC 63, [2001] 2 IR 545 at p. 688, Hardiman J. described as ‘otiose’ much of the debate around suggested tensions between methods of constitutional construction described as ‘historical’, ‘harmonious’ and ‘purposive’: each of these words, he said, ‘connotes an aspect of interpretation which legitimately forms part, but only part, of every exercise in constitutional construction’. Noting consistent urgings in the case law for a method of interpretation that placed words in their ‘constitutional context’ (at p. 688), he approved (at p. 695) the following well-known statement of Henchy J. in The People v. O’Shea [1982] IR 384 at p. 426 (‘O’Shea’):
‘Any single constitutional right or power is but a component in an ensemble of interconnected and interacting provisions which must be brought into play as part of a larger composition, and which must be given such an integrated interpretation as will fit it harmoniously into the general constitutional order and modulation. It may be said of a constitution, more than of any other legal instrument, that ‘the letter killeth, but the spirit giveth life’. No single constitutional provision … may be isolated and construed with undeviating literalness’.
63. In the course of this judgment Henchy J. further observed (at p. 430) that a provision of the Constitution could not permit something ‘if the allowance of it would infringe any specific or more fundamental guarantee, mandate or prohibition expressed in, or postulated by, the Constitution.’ He preceded this referring to (at p. 429):
‘the principle that the Constitution must be treated as a logical whole, each provision of which being an integral component, so that it is both proper and necessary to construe any single provision in the light of the other provisions. Such an application of the doctrine of harmonious interpretation requires that the words “all decisions” be given a restricted connotation’
64. In argument in this case, as in others, there has been a tendency to stress that Henchy J. was in the minority in O’Shea and to contrast the terms in which he expressed his conclusion with those in the majority judgments. The debate, in my view, proceeds on a false premise. While the majority and minority undoubtedly diverged in the emphasis they placed on language, purpose, history and background, the actual decision in O’Shea is properly viewed as turning on a difference not of interpretative principle, but of application. The issue was whether Article 34.4.3 (‘[t]he Supreme Court shall … have appellate jurisdiction from all decisions of the High Court’) meant what it appeared to suggest - that every decision of the High Court could be appealed to this Court and, therefore, that an appeal to it lay against a directed verdict of not guilty in the Central Criminal Court. The majority (O’Higgins C.J., and Walsh and Hederman JJ.) held the Article meant what it said, while the minority (Finlay P. and Henchy J.) adopted the view that it fell to be construed in the light of the established principle at common law that no appeal lay against an acquittal following a trial before a jury. It was their view that that principle, by being absorbed within the guarantees provided for in Article 38 of the Constitution, qualified the seemingly unambiguous text of Article 34.
65. The judgments show the force of the arguments on either side of that debate, but ultimately the majority decided the case as it did, not because the text of other provisions of the Constitution was irrelevant to the interpretation of the clear language in the Article, but because those members of the Court did not believe that Article 38 of the Constitution incorporated a right not to have an acquittal appealed against (see O’Higgins C.J. at p. 403, Walsh J. at p. 420). Having regard to that conclusion, the majority was not prepared to accept what it viewed as the remaining centrepiece of the defendant’s argument: ‘existing laws, or formerly accepted legal principles or practices, cannot be invoked to alter, restrict or qualify the plain words used in the Constitution unless the authority for so doing derives from the Constitution itself’ (O’Higgins C.J. at p. 398): ‘[a]s this argument is based wholly upon practice and tradition, it cannot prevail against the express words of Article 34’ (Walsh J. at p. 420). Of course, the minority decision as to what Article 38 required was informed by the pre-existing law, but it was the content of the guarantee in Article 38 - rather than the true relationship of individual Articles of the Constitution to each other - that defined the difference between the majority and minority, Finlay P. and Henchy J. being of the view that one of the essential features of the right to trial by jury as envisaged by Article 38 was the immunity of a verdict of not guilty (see Finlay P. at p. 411 and Henchy J. at p. 431-432).
66. The majority judgments reflected (and were based upon) earlier decisions to similar effect: The State (Browne) v. Feran [1967] IR 147 (appeal lay to this Court against an absolute order of habeas corpus although the common law had laid down that it did not) and The People (Attorney General) v. Conmey [1975] IR 341 (appeal also lay against a conviction following a trial before the Central Criminal Court notwithstanding the establishment of the Court of Criminal Appeal). As it happens, the danger of over analysing judicial comment in this territory is well demonstrated by the fact that Walsh J. in the first of these cases posited a principle that is not inconsistent with the ‘harmonious interpretation’ urged by the minority in O’Shea. Walsh J. said (at p. 159):
‘In the construction of a Constitution words, which in their ordinary meaning import inclusion or exclusion, cannot be given a meaning other than their ordinary literal meaning save where the authority for so doing can be found within the Constitution itself.’
67. A test of reading up or down the product of a literal construction by reference to what is ‘within the Constitution itself’ stretches a broad canvas, one also envisaged by O’Higgins C.J. in the passage I have cited earlier. O’Higgins C.J. made the same point elsewhere in his judgment in O’Shea (at p. 397):
‘Of course, the Constitution must be looked at as a whole and not merely in parts and, where doubt or ambiguity exists, regard may be had to other provisions of the Constitution …. Plain words must, however, be given their plain meaning unless qualified or restricted by the Constitution itself.’
68. The qualifications to all of these statements are important. An individual provision of the Constitution is neither plain nor unambiguous if it conflicts with another Article, and the question of whether there is ‘doubt or ambiguity’ can sometimes be answered only with an understanding of context and object. The point is well made by Professor Casey in relation to this very passage: sometimes a provision will seem perfectly plain when viewed in isolation and that it is only when read ‘in the wider context may an ambiguity emerge’ (Casey Constitutional Law in Ireland 3rd. Ed. 2000 at p. 377). It will become clear from what I say later that a careful analysis of Article 18.4.2 itself shows the wisdom and force of this insight.
69. The judgment of Henchy J. in O’Shea has often been cited with approval or apparent approval (see The Attorney General v. X [1992] 1 IR 1 at p. 87 (per O’Flaherty J.), TD v. Minister for Education [2001] IESC 101, [2001] 4 IR 259 at p. 307 (per Denham J.) and p. 367 (per Hardiman J.), AO and DL v. Minister for Justice [2003] IESC 3, [2003] 1 IR 1 at p. 143 (per Hardiman J.), The People (Director of Public Prosecutions) v. MS [2003] IESC 24, [2003] 1 IR 606 at p. 619 (per Keane C.J.) Curtin v. Dáil Éireann [2006] IESC 14, [2006] 2 IR 556 (‘Curtin’) at p. 610 (per Murray C.J.)). Similar statements focussing on the importance of considerations of overall purpose and of reading the Constitution harmoniously and as a whole have appeared in the case law for decades (see O’Byrne v. Minister for Finance [1959] IR 1 at p. 21 (per Dixon J.) (‘it is not to be parsed with the particularity appropriate to ordinary legislation and that the intention, if it can reasonably be gathered, should prevail’), O’Donovan v. Attorney General [1961] IR 114 at p. 129 (per Budd J.) (‘the words used … must, to be properly understood, be read in the light of the Constitution as a whole’), Dillane v. Attorney General [1980] ILRM 167 at p. 170 (per Henchy J.) (‘the doctrine of harmonious interpretation … requires, where possible, the relevant constitutional provisions to be construed and applied so that each will be given due weight’), Attorney General v. Paperlink [1984] ILRM 373 at p. 385 (per Costello J.) (‘a purposive, rather than a strictly literal, approach to the interpretation of the sub-paragraphs is appropriate’), Gilchrist v. Sunday Newspapers [2017] IESC 18, [2017] 2 IR 284 at p.310 (‘[t]he Constitution was intended to function harmoniously, and where there were points of potential conflict between rights and obligations provided for, that should be sought to be resolved without the subordination or nullification of one provision’ (per O’Donnell J.).
70. Parts of that judgment have attracted some criticism [10] and there can be no doubt but that unless carefully constrained the use of concepts such as ‘constitutional order and modulation’ or the ‘fundamental purposes’ of the document as mechanisms of interpretation raises valid concerns. For my part I do not read the judgment in O’Shea – or any other decision of this Court - as necessarily requiring that anything other than the text of the instrument, rights that have been derived from, and general principles and purposes that can be related to, the text of the Constitution (together with clear evidence of context and purpose of the kind to which I refer in the next part of this judgment) can be resorted to in the process of interpretation. That reflects the limitations suggested by Clarke C.J. on the deduction of what used to be called ‘unenumerated rights’ in Friends of the Irish Environment v. Government of Ireland [2020] IESC 49, [2021] 3 IR 1 at para. 8.6 et seq. Those limitations are sharper again when it comes to the exercise of constitutional interpretation: the authorities make it clear that the text of any given Article both marks the proper starting point for, and delineates the finishing line of, the process of construction. Subject to that proviso, I do not believe that Henchy J.’s analysis can or should now be seriously questioned as the authoritative explanation of how a court should view any issue of interpretation presented by any Article of the Constitution.
71. It must, however, be stressed that this does not simply mean that it is only where there is some tension between the text of specific provisions and an ambiguity in one, that this interpretative approach can be taken. This is clear from Tormey v. Ireland [1985] IR 289. There, the issue was whether Article 34.3.1 (which described the High Court as being ‘invested with full original jurisdiction in and power to determine all matters and questions whether of law or fact’) meant that the plaintiff was entitled to a right to trial in the Central Criminal Court on a charge under the Larceny Acts. The Court found that the provision did not have the effect its undiluted language would suggest, and that the consequence of Article 34.3.1 and Article 34.3.4 was that jurisdiction could be devolved on an exclusive basis to inferior courts by legislation, provided the High Court retained a supervisory jurisdiction.
72. Noting that O’Shea was opened to the Court, the relevant interpretative principle was explained by Henchy J., giving the judgment of the Court (the other members of which were Griffin, McCarthy, Barrington and Carroll JJ.) (at p. 295-296). The paragraph merits quotation in full:
‘The rule of literal interpretation, which is generally applied in the absence of ambiguity or absurdity in the text, must here give way to the more fundamental rule of constitutional interpretation that the Constitution must be read as a whole and that its several provisions must not be looked at in isolation, but be treated as interlocking parts of the general constitutional scheme. This means that where two constructions of a provision are open in the light of the Constitution as a whole, despite the apparent unambiguity of the provision itself, the court should adopt the construction which will achieve the smooth and harmonious operation of the Constitution. A judicial attitude of strict construction should be avoided when it would allow the imperfection or inadequacy of the words used to defeat or pervert any of the fundamental purposes of the Constitution. It follows from such global approach that, save where the Constitution itself otherwise provides, all its provisions should be given due weight and effect and not be subordinated one to the other. Thus, where there are two provisions in apparent conflict with one another, there should be adopted, if possible, an interpretation which will give due and harmonious effect to both provisions. The true purpose and range of a Constitution would not be achieved if it were treated as no more than the sum of its parts.’ [11]
73. In the course of their submissions, the respondents sought to attach some significance to the fact that in Curtin, the Court did not refer to Tormey and that Murray C.J. did not expressly approve the passages from the judgment of Henchy J. in O’Shea to which I have earlier referred, while describing a passage from the judgment of O’Higgins C.J. as ‘particularly authoritative’. This is, I think, to - again - over analyse the judgment.
74. Curtin is certainly an important case, but the interpretative principle for which it is actual authority is that a provision of the Constitution which is, on its face, pellucid may not merely fall to be qualified by reference to the express text of another Article, but also by reference to a well-established constitutional principle of general application viewed in the light of relevant aspects of constitutional history. Article 35.4.1 of the Constitution is, at first glance, as clear as can be: a judge is removed from office ‘upon resolutions passed by Dáil Éireann and by Seanad Éireann calling for his removal’. In striking contrast to the procedure laid down for the impeachment of the President, there is no pre-ordained process. The Article, literally construed, envisages a short, sharp outcome the only precondition to which was a duly passed parliamentary resolution.
75. None of the parties in Curtin thought that this was how the Article should be interpreted. The applicant sought to challenge the legality of a process (and the constitutionality of legislation pursuant to which it was purportedly embarked upon) commenced in the Houses of the Oireachtas and ultimately directed to the consideration of resolutions for his removal as a judge of the Circuit Court pursuant to Article 35.4.1. As explained in the judgment of Murray C.J. (at para. 72), one of the issues depended on the extent to which, by reference to history, to other provisions of the Constitution, to the independence of the judiciary, to the principles of the separation of powers, to the need to respect fair procedures or otherwise, the Court should interpret the provision as requiring the observance of particular procedures contended for by the applicant. Murray C.J., noting the need to strike the correct balance ‘between the effect to be given to the literal meaning of particular words and the need to have regard to the terms of the Constitution as a whole’ (para. 73) referred to both principal judgments in O’Shea. Specifically noting that Keane C.J. in The People (Director of Public Prosecutions) v. MS had approved the analysis conducted by Henchy J. in that case, he continued as follows:
‘Where words are found to be plain and unambiguous, the courts must apply them in their literal sense. Where the text is silent or the meaning of words is not totally plain, resort may be had to principles, such as the obligation to respect personal rights, derived from other parts of the Constitution. The historical context of particular language may, in certain cases, be helpful …’
76. It will be noted that here Murray C.J. was addressing the relationship between language that was ‘not totally plain’ and ‘principles’ reflected in the Constitution as a whole, and not the question that arises if two provisions of the Constitution are in apparent conflict. As I have already said, where two provisions of the Constitution sit uneasily together, neither can be said to be ‘totally plain’ and that tension in itself may generate an ambiguity such as to justify moving beyond the literal language of either. As Murray J. (as he then was) himself quipped in the course of his judgment in Crilly v. Farrington [2001] IESC 60, [2001] 3 IR 251 at p. 299 ‘“[a]mbiguous” is an ambiguous term’.
77. While aspects of Article 35.4.1 were found in Curtin to be clear and unambiguous, on some critical issues the provision was silent - as to the meaning of ‘misbehaviour’, as to the procedures to be followed in removing a judge, and as to whether the Houses could appoint investigating committees or (if so), the powers that could be delegated to them. Those questions, it was found, fell to be resolved by reference to the combined effect of the history of the Article, the importance of the value of judicial independence, the separation of powers and the over-riding requirements of fair procedures and constitutional justice (the need to interpolate the latter of which into the provision, was accepted by the respondents).
78. In conducting that exercise, the Court undertook a complex and nuanced review of the relationship between these general provisions, Article 35.4.1 and a number of other Articles the Constitution - notably the power of each House of the Oireachtas to make its own rules and standing orders (Article 15.10) and the mandate of judicial independence in Article 35.2. These led to a range of conclusions all of which were weighted on or around the forty-three words in Article 35.4.1 - including that the Houses of the Oireachtas were entitled to establish a committee to investigate the allegations against the judge giving rise to the resolution, that this could never be done in advance of and merely in contemplation of the possible proposal of such a resolution, that it was open to the Houses to provide for an investigating committee that would make no findings of fact or recommendations but that they could have been invested with such powers, that the Houses themselves could hear evidence and must be open to do so where fair procedures so required, and that they must accord to the applicant full rights of constitutional justice and fair procedures. The end point of that exercise was a process which bore closer resemblance to the specific procedures identified for the impeachment of the President than to the bare text of Article 35.4.1. Curtin thus, if anything, shows precisely how - and why - there will be circumstances in which a seemingly clear constitutional provision must be supplemented by reference to broader but clearly ascertainable constitutional principles, and how these may combine to produce an interpretation of a provision that is radically different from that suggested by a clinical analysis of only the bare text.
The ‘Bacik’ principles
79. I have digressed a little on these issues having regard to the extensive debate conducted in this case around the general principles governing the interpretation of the Constitution and in the light of the position adopted by the respondents as to the proper approach to the interpretation of what are said to be ‘clear’ constitutional Articles. Stating (not incorrectly) that the literal approach is the ‘default’ method of constitutional interpretation, noting that Murray C.J. had said at one point in Curtin that words denoting numbers, places or identified persons admit of no debate, they say that Article 18.4.1 and 18.4.2 are clear and unambiguous and, in particular, refer to numbers of identified persons in just the manner contemplated by that statement. They underline that in Howlin v. Morris [2005] IESC 85, [2006] 2 IR 321 at p. 364 Hardiman J. said that in relation to ‘technical’ provisions of the Constitution (in that case Article 15.10) it was appropriate to adopt ‘a “narrower and less liberal”’ approach. Article 18.4. - they emphasise - establishes the composition of a legislative chamber and is concerned with the ‘technical architecture of the State’. They place, in this regard, heavy reliance on the decision of the Divisional Court in Bacik v. An Taoiseach [2020] IEHC 313, [2021] 3 IR 283 and, in particular, on the summary of interpretative principles appearing at para. 80 of the judgment in that case.
80. In Bacik, the plaintiffs had each been elected to Seanad Éireann in the course of the general election for the Seanad but, at the time of that election, the Taoiseach had not been appointed by the newly elected Dáil Éireann. Thus, the eleven members to be appointed by the newly elected Taoiseach had not been nominated and in those circumstances, the Taoiseach from the previous Government had refused to advise the President to fix a date for the first sitting of the new Seanad. The plaintiffs contended that the eleven members to be nominated by the Taoiseach were not required to be so nominated before that first sitting, and sought orders requiring the Taoiseach to convene that sitting. While the position of the respondents altered during the currency of the proceedings, the essential issue as addressed by the Divisional Court was whether the first meeting of Seanad Éireann could lawfully take place before all sixty members identified in Article 18.1, elected and nominated, were in place.
81. That fell to be resolved having regard to the fact that Article 18.1 states that the Seanad ‘shall be composed of sixty members ...’ as a result of which, the Court found, reference in Article 18.8 to the first meeting of the Seanad was a reference to the Seanad as so composed. That conclusion was reached in a context where the applicants had presented a range of interpretative arguments based upon the Irish text, the text of various other provisions (including Article 16.2.1 and 16.2.2), the history of Article 18.3 (dealing with the appointment of the Taoiseach’s nominees) and claims that the interpretation urged by the respondents would undermine the political system and render the principal organ of representative government inoperative.
82. In that regard, the Court considered the judgments in O’Shea, Tormey, and Curtin, from which the following principles (upon which the respondents heavily relied) were deduced (and to one of which - para. (e) - I have earlier referred):
‘The following principles flow from the approach taken in [Curtin v. Dáil Éireann [2006] IESC 14, [2006] 2 IR 556] and from the case-law cited by Murray C.J. in that case:
‘(a) The starting point is to carefully consider the words used in a constitutional provision with a view to identifying their meaning.
(b) While not specifically addressed in Curtin, it may also be necessary to consider the meaning of the words in the Irish language version of the text which, in accordance with Article 25.4 of the Constitution, takes priority in the event of any conflict with the English language version.
(c) Where the words used are clear and unambiguous, they are to be construed in their literal sense. Thus, for example, words denoting numbers, places or identified persons admit of no debate.
(d) The words used in the provision and issue cannot be construed in isolation. They must be construed in the context of the Constitution as a whole.
(e) If a literal interpretation of one provision might bring it into conflict with the literal meaning of another provision, then it is legitimate to resort to the harmonious approach with a view to interpreting both provisions in a way which avoids inconsistency. In this context, while Murray C.J. in Curtin did not expressly say that the harmonious interpretation favoured by Henchy J. in O’Shea should be applied, he did not dissent from the observations of Henchy J. to that effect in the passage quoted by him. It is interesting to observe that, although [Tormey v. Ireland [1985] IR 289] was cited by counsel in Curtin, the judgment of the court does not refer to it. Having regard to the emphasis placed by the Supreme Court in Curtin on the principle of the words of the constitutional provision in issue should be the first port of call, it seems to us that the harmonious approach will only be taken in cases of apparent inconsistency. It will not be necessary to go beyond a literal interpretation of a constitutional provision unless such an interpretation gives rise to an apparent conflict with some other provision of the Constitution.
(f) In case of doubt, ambiguity, inconsistency or silence, it is legitimate to have regard to factors such as the historical context. Although the issue did not arise in Curtin, this would appear to include, for example, a relevant amendment to the Constitution, something which was considered for example in [M v. Minister for Justice [2018] 1 IR 417] [...]’
83. Paragraphs (a), (b), (d) and (f) of this most useful distillation are clear and undoubtedly correct. The remaining paragraphs - if read together - are also not an inaccurate encapsulation of the cases. However, if each one of them is construed in isolation from the other paragraphs in the summary (as the respondents sought to do here) they are liable to mislead. So, the statement in para. (c) that where words are clear and unambiguous they are to be construed in their literal sense is true only once it is appreciated that seemingly clear and unambiguous language may not actually be clear and unambiguous at all once understood in the light of other provisions of the Constitution, the general principles that underlie the document as I have explained them earlier and the overall context and purpose of the provision (as indeed I think is acknowledged in (d)). The statement in (e) that resort may be had to the ‘harmonious’ approach where there is a conflict between the literal interpretation of two provisions is right, but this is not the only circumstance in which this can happen: to rephrase what Henchy J. said in O’Shea in accordance with my earlier observations, a judicial attitude of strict construction should be avoided when it would allow the imperfection or inadequacy of the words used to defeat or pervert any of the fundamental principles of the Constitution as deduced in the manner I have earlier described.
84. The historical context may of course be relevant in any case of doubt, ambiguity, inconsistency or silence. Indeed some of the judgments in Maguire v. Ardagh [2002] IESC 21, [2002] 1 IR 385 show that in many such cases it may be quite important. But there is no case, and no provision - no matter how clear it may be - in which it is other than appropriate to understand and have regard to the overall context in which that provision was adopted. That context will often be indivisible from the purpose of the provision.
85. These various strands come together in this way. The Constitution, as the basic law of the State, is animated by a number of fundamental principles and is intended to achieve its various purposes through a sequence of provisions intended to be read as one, and designed to function collectively. That understanding dictates the proper approach to the interpretation of any individual part of the instrument. As with any legal document, the intent of those adopting it is informed in the first instance by the language they have used (and having regard to Article 25.4, that appearing in the Irish text) but the exercise of interpreting that language involves not merely deducing the meaning of the words appearing in the provision in question, but also reconciling the text of the Article with the document in its entirety while at the same time ensuring that both are analysed in the light of their underlying purpose. The predominant importance of that purpose, and the fundamental principles that inform it, means that in undertaking the exercise of constitutional interpretation the courts must incline more to flexibility in subordinating the strictly literal interpretation of the text in order to attain a construction that is both internally harmonious, and that achieves a clearly ascertainable purpose, than might be the case when conducting the exercise of interpreting ordinary legislation.
Extrinsic material and the interpretation of a constitutional Amendment
86. Much of what I have said so far is concerned with reading the text of a provision of the Constitution in the light of other provisions of the document, or matters of basic constitutional principle. But what are variously described as the ‘context’, the ‘purpose’ or the ‘historical context’ may also be important. Sometimes, these mean the same thing and/or are sought to be ascertained by reference to the same material. Often, the text will completely explain the purpose. Sometimes, depending on the precise issue in play, these matters may be established by taking account of the pre-existing common law or established pre-independence constitutional practice (as indeed happened in Curtin), or indeed by reliable accounts in authoritative legal or historical texts. Less commonly and in most cases less usefully, regard may be had to material preparatory to the drafting of the Constitution. Most strikingly, in Maguire v. Ardagh reference was made to early drafts of the 1922 Constitution in considering whether it was intended that the Houses of the Oireachtas would enjoy powers of investigation (see Denham J. at p. 562 and Geoghegan J. at p. 711), and a similar approach was suggested by Geoghegan J. in Howlin v. Morris as regards Article 15.
87. However, all of this comes with a warning. Many of the difficulties in attributing an intention to Parliament by reference to debates in the Houses of the Oireachtas or other extraneous evidence discussed in the context of the interpretation of primary legislation in Crilly v. Farrington transfer across to the process of constitutional interpretation. In particular, a court can rarely be confident that pre-enactment drafts, parliamentary debates around the Constitution or indeed contemporaneous public discussion illuminate in any way the intent of those ratifying the Amendment: ‘to rely on individual documents produced by select number of individuals as part of an ongoing and evolving process of drafting as evidence of collective intention would risk engaging in the sort of errors identified by McGuinness J in Crilly v Farrington … To these concerns might be added, in the constitutional context, the democratic dubiousness of construing the popular will as expressed in documents of which the public were generally unaware.’ (Carolan ‘Originalism Enabled? The Role of Historical Records in Constitutional Adjudication’ (2013) 36 DULJ 311 at pp. 320-321). The reference is to this statement of McGuinness J. [2001] IESC 60, [2001] 3 IR 251 at p.302:
‘The process of legislation by the Oireachtas is essentially collective. It is the Oireachtas as a whole which legislates. It would in my view be a misleading oversimplification of this process to rely, in interpreting a statute, on ministerial statements alone … For the courts to rely on ministerial statements in interpreting statutes would not, therefore, reflect the will of the Oireachtas as a whole.’
88. That this is an objection that is in truth based upon reliability and utility rather than admissibility in the strict sense is shown by the very particular position adopted in the cases to the interpretation of (especially, more recent) constitutional Amendments. There are important, if self-evident, practical differences between the process of interpreting an Amendment to the Constitution and the construction of the original text as adopted by the People. In 1937, the People adopted a single and composite text. It will usually be less than useful to seek to construe that text by reference to earlier drafts or communications between those involved in the drafting process or to contemporary public discussion of the meaning of particular Articles. As a matter of historical record the People did not have any insight into the drafting process when voting on the draft Constitution in July 1937 and, indeed, it is only in the late 1980s that access to this archival material first started to become generally available. It cannot be assumed that the electorate as a whole was concerned with the nuances of any specific Article and thus that it would have had any necessary awareness of mooted interpretations or effects attributed to it. For this reason, it is more common to see the analysis of the intent of the People conducted by reference to what ‘the framers’ (a term, perhaps curiously, imported into discussion in this jurisdiction) intended by a particular provision. For the reasons I have just identified, appeals to the ‘framers’ intent’ does not often significantly advance the interpretative process.
89. When it comes to amendments to the original text, however, the People are considering one, or a limited number, of proposals. Contemporary discussion may provide a useful guide to what they understood they were adopting. Of course, as with any extrinsic source used for the construction of a document, that discussion must be clear and reliable. Usually, this will happen when there is unequivocal and indisputable evidence of a general consensus as to what a referendum was ‘about’. But in principle, where the text inserted by an Amendment is unclear or gives rise to a doubt in the sense, or for any of the reasons, to which I have referred earlier, I can see no reason why reliable evidence as to what the Houses of the Oireachtas as institutions (as opposed to the stated positions of individual members) believed they were proposing, or what the People might reasonably have understood they were adopting, cannot be taken into account in the construction of the provision.
90. Without a doubt, when such material is tendered it is of less impact than other interpretative sources: whatever about reading down provisions of the Constitution by reference to other Articles of the instrument, or to established principles such as those of fair procedures or the requirements of other fundamental rights, and whatever about the utility of such sources in presenting a general theory of the ‘purpose’ of a provision, general evidence of such purpose must be treated cautiously where it is deployed to expand or reduce the literal meaning of an Article where the text is otherwise clear and capable of functioning without evident absurdity.
91. That being so, and given that it is both obvious and well established that in interpreting any provision of the Constitution it is important to consider what it was intended to mean as of the date that the People approved it (Sinnott v. Minister for Education per Geoghegan J. at p. 718), it is unsurprising that there have been cases in which account was taken in the way I have just described of contemporaneous understandings of the purpose of proposed constitutional Amendments. In Roche v. Roche [2009] IESC 82, [2010] 2 IR 321 one of the issues was whether the Eighth Amendment to the Constitution Act 1983 afforded a basis on which the plaintiff could require the return to her of frozen embryos created in the course of in vitro fertilisation by the parties. The plaintiff contended that the embryos were ‘unborn’ for the purposes of Article 40.3.3, and sought to rely upon that contention in advancing her claim. Denham J. (as she then was) (at para. 125) stressed the importance of the statutory and constitutional context in which the Amendment was introduced, and the mischief to which it was addressed, that context leading her to conclude that the provision was concerned with the termination of pregnancy by abortion (para. 134). Geoghegan J. reaching the same conclusion (and contrasting the process of constitutional Amendment with that of construction of a statute) observed (at para. 210):
‘Judges play no part in the drafting of a statute, still less in the voting of it into law. Judges, however, are ordinary citizens and do participate in referenda. It would seem to me to be highly artificial if a judge could not also take judicial notice of and, to some extent at least, use as an aid to interpretation, the ordinary common understanding of what in context was involved in the referendum.’
92. This passage was cited with approval in IRM v. Minister for Justice [2018] IESC 14, [2018] 1 IR 417, where it was described as being of relevance in the approach to be taken by a court in placing a constitutional amendment in its context as of the time when enacted (at para. 215). There, the Court in determining the extent of the rights invested in the unborn, took into account the discerned object of the Thirteenth and Fourteenth Amendments (at para. 194) (‘to prevent restrictions on travel or the provision of information or [sic] travel, and in particular to preclude any interpretation of the Constitution which could lead to the grant of any order restraining the provision of such information or undertaking of such travel.’). Similarly, in O’Doherty v. Attorney General [2009] IEHC 516, [2010] 3 IR 482, at paras. 24-25) Birmingham J. (as he then was) looked to reports of relevant Oireachtas Committees when considering the overall purpose of Article 28A of the Constitution.
The evidence in this case
93. At the trial, conflicting evidence was given as to the purpose of the Seventh Amendment. Dr. Cahillane, a lecturer in constitutional law at UL, gave evidence on behalf of the applicant. Her evidence is addressed at length in the High Court judgment. She treated in some detail the history of the Seanad and the debate around its composition at the time of the framing of the Constitution, noting in particular the resistance of Mr. de Valera to a directly elected Senate. The Divisional Court recorded part of that evidence as follows (at para. 60):
‘Dr. Cahillane offers the opinion that there was no general consensus in terms of the appropriate composition for the electorate of the Seanad. This opinion was elaborated upon in her oral evidence: Dr. Cahillane agreed that there had been a significant retooling of the Upper House for the purpose of the 1937 Constitution. The witness explained that the direct elections were considered too complicated because of the massive electorate involved. The drafters of the 1937 Constitution were in agreement with the idea of functional or vocational representation but simply could not decide how they would to that. It was further stated it presented too many problems to put a very clear method of Seanad elections into the 1937 Constitution and they decided to do so by legislation.’
94. Her written testimony regarding the 1979 Amendment to the Constitution included the following:
‘It is clear the 7th amendment had a dual purpose. While the necessity for the amendment arose in the first place because of the planned dissolution of the NUI, it is evident from the debate in the Oireachtas at the time, as well as the newspaper coverage that the necessity to provide equality of representation for all third level graduates was an equal purpose of the Amendment. The people clearly expected legislation to follow through with this after the referendum.’
95. Dr. Eoin O’Malley, an associate professor of politics in the School of Law and Government at Dublin City University, and who gave evidence on behalf of the respondents, disagreed. Dr. O’Malley’s evidence was that the 1979 referendum did not constitute a mandate by the People to extend the franchise to all graduates. He said that the initial motivation for the Seventh Amendment to the Constitution was the need to accommodate the potential dissolution of NUI and the establishment of independent colleges, although he accepted that there was also discussion at the time of the possibility that subsequent legislation would lead to an extension of the franchise in respect of university seats. He said that the Government’s advertising set out as the primary purpose of the proposed Amendment the need to accommodate the potential dissolution of the NUI. He noted, in particular, that the sponsoring Minister was not the Minister for the Environment but the Minister for Education. The judgment records part of his evidence, as follows (at paras. 72 and 73):
‘The third issue on which Dr. O’Malley provided a witness statement was whether there was consensus “as to what reforms there should be to the university seats?” He felt there was no such consensus, pointing out that Sinn Féin “would remove them altogether”. The Manning Report “called for their effective abolition”, TCD had suggested the retention of the six seats “to be supplemented by another four graduate seats”, three Fianna Fáil parliamentarians had sponsored a recent bill seeking “the extension of the Seanad seats to all graduates”, and that the interests campaigning in 2013 to retain the Seanad did not have any “agreement as to how the university seats would be addressed”.
Finally, in the event that there was a unitary Seanad university constituency, Dr. O’Malley was asked how this would be organised. He felt this would create difficulties, and would involve significant resources, a legislative basis for the collection and storage of data, and significant cooperation among the higher education institutions.’
96. Some of this evidence was supported by reference to newspaper advertisements at the time of the referendum and to contemporary commentary. For example, Dr. Cahillane gave evidence that it was reported in numerous newspapers (she named ten specific articles) that on foot of the Amendment, if passed, the Government ‘intended to redistribute the seats of the existing university representation in order to allow graduates of all third level colleges a vote in the elections to those seats’. She quoted the then leader of the Seanad as stating that the ‘likely’ representation would be two seats each for TCD and University College Dublin, a seat between University College Cork and University College Galway, and one seat for ‘other institutes of higher education’. She also referred to the Minster for Education explaining that it was intended that other institutions of higher education in the State would enjoy votes, and to his intention ‘to go ahead with the legislation as quickly as possible’. A government advertisement stated that the Minister for Education would introduce Bills to dissolve the NUI and establish new independent universities and that ‘[t]he Minister … will introduce a Bill to provide by law for the election of six members to the Seanad by the institutions of higher education specified by law’. This material, Dr. Cahillane said under cross-examination, showed that the extension of the franchise ‘potentially became the more dominant purpose of the amendment’ as the coverage focussed ‘almost exclusively on the extension of the franchise rather than on the issue of the dissolution’.
97. In response, Dr. O’Malley quoted the Minister for Education as describing the purpose of the referendum as being to remove an obstacle standing in the way of the introduction of Bills enabling the reorganisation of university education (the point being that if NUI were dissolved the arrangements for the election of senators by that university would automatically lapse). He noted that Professor Kelly TD had also spoken in the Dáil in favour of the extension of the franchise to NIHE Limerick [12] or the Regional Technical Colleges, observing that he could see ‘there is not any such intention here’. His evidence was that the advertising spoke only in terms of the reorganisations of the university sector (‘the main purpose of this Referendum is to allow for the reorganisation of university education’), and he noted the then President of the Students’ Union in UCD as observing that the real issue at stake was not the minor one of the reallocation of Senate seats, but the dissolution of the NUI. His conclusion based on this information was clear: ‘[t]he Bill had one main purpose which was, as the Minister had stated and as the public debate went on, the Government put to the people, and that was the dissolution of the NUI and the subsequent reorganisation of the Seanad seats as a result of the dissolution of the NUI.’
98. It is obvious from this summary that much of the contextual information put before the Court was presented by reference to (conflicting) testimonial evidence subject to inevitable disputation. The newspaper archive provided to the High Court comprised some 110 pages of articles, letters published in newspapers and advertisements which supported one, other (or in a few cases, both) versions of what the public were being told the objective of the referendum was, and what was going to occur were it passed. I mean no disrespect whatsoever to the highly distinguished witnesses who gave their time and specialist knowledge to tender evidence in this case, when I say that it is hard to my mind to conceive of circumstances in which the Court could comfortably embark upon the exercise of interpreting the Constitution based upon evidence of this kind. It is harder still to dispute the suggestion that to conclude otherwise would involve enveloping the meaning of the State’s fundamental law in a fog of confusion. Nor, for that matter, is it desirable that parties should feel the need to instruct, or that the process of constitutional litigation should be either clouded or prolonged by the evidence of, expert witnesses purporting to testify as to the purpose or meaning of a constitutional provision. If material otherwise relevant to the construction of the Constitution is not sufficiently clear to be understood in its own terms and without the benefit of expert explanation, it is difficult to my mind to see how it can ever be of assistance in the process of interpretation. If information tendered in support of a suggested object of a proposal to amend the Constitution is not close to unanimous in its distillation of the purpose of the referendum, it cannot be of any real utility in clarifying any ambiguity or doubt that may attend the resulting provision.
The statement provided for in the Referendum (Amendment) Act, 1979
99. However, in this case there was information relevant to the meaning and intendment of the Seventh Amendment of a very particular kind. To explain: the procedure governing the Amendment of the Constitution is sketched in Article 46. Every proposal for such an Amendment must be initiated in Dáil Éireann as a Bill. Upon being passed by both Houses of the Oireachtas, it is submitted by Referendum to the decision of the People. If the President is satisfied that the provisions of the Article have been complied with and that such a proposal has been duly approved by the People in accordance with Article 47.1 of the Constitution, it shall be signed by him or her and then duly promulgated as a law.
100. At the time of the Seventh Amendment in 1979, the conduct of referendums was regulated by the provisions of the Referendum Act, 1942 and the Electoral Act, 1963, as amended. Section 64 of the latter provided that at a referendum the local returning officer for a constituency shall send to every elector a polling card, while Rule 18 of the First Schedule to the Referendum Act, 1942 made provision for enabling incapacitated voters to be apprised of the proposal.
101. The Referendum (Amendment) Act, 1979 made specific provision for the constitutional referendums in relation to the Sixth Amendment of the Constitution (Adoption) Bill, 1978 and the Seventh Amendment of the Constitution (Election of Members of Seanad Éireann by Institutions of Higher Education) Bill, 1979. It provided that the polling cards for each of these referendums must contain explanatory statements, the form of which was set forth in an Appendix to s. 1 of that Act. It required that these statements be displayed by the presiding officer in the precincts of his or her polling station. It further stated (s. 1(f)) that in applying Rule 18 the presiding officer could assist an incapacitated voter by reading the proposal to him or her, but that if the voter failed to understand this he or she was required to:
‘read out to the voter such statement of the proposal which is the subject of the referendum as is set out in paragraph 1 of the Appendix to section 1 of the Referendum (Amendment) Act, 1979’
102. The ‘statement of the proposal’ applicable to the Seventh Amendment and contained in the Appendix to s. 1, was as follows:
‘The Seventh Amendment of the Constitution (Election of Members of Seanad Éireann by Institutions of Higher Education) Bill, 1979, proposes the election by universities and other institutions of higher education specified by law of such number of members of Seanad Éireann, not exceeding 6, as may be specified by law. Those so elected would [13] be in substitution for an equal number of the members elected at present (3 each) by the National University of Ireland and the University of Dublin. The Bill also proposes that nothing in Article 18 of the Constitution shall prohibit the dissolution by law of those Universities.’
(Emphasis added).
103. For reasons I have outlined earlier I agree with the conclusion reached by the Divisional Court that most of the evidence adduced by the applicant to support the construction of Article 18.4.1 for which he contended was too equivocal to be of any real use in understanding the provision. However, the statement attached to s. 1 of the Referendum (Amendment) Act, 1979 is in a different category. While it is certainly the case that the provision refers to ‘such number of members of Seanad Éireann, not exceeding 6, as may be specified by law’ (my emphasis), in no sense could it be said that this ‘statement of the proposal’ communicates that the effect of the Amendment, if adopted, will be to vest in the Oireachtas an unbridled discretion to add, or not add, additional institutions to the Seanad franchise. Nor does it suggest that in fact the effect of the Amendment if adopted would be to enable the further concentration of the franchise for all the university members in a single institution: on the contrary, all references to universities and institutions of higher education in the statement are framed in the plural.
104. The title of the 1979 Act - ‘Election of Members of Seanad Éireann by Institutions of Higher Education’ – suggests without qualification an intention to broaden the franchise beyond the universities (of which, in 1979, there were still only two and to which the legislature was exclusively referring when it spoke of ‘universities’). In that regard - and critically (if unsurprisingly) - it mirrored the title of the relevant Referendum Bill itself (‘Seventh Amendment of the Constitution (Election of Members of Seanad Éireann by Institutions of Higher Education) Bill, 1979): that title appeared on the actual Ballot Paper, together with the words ‘SEANAD REPRESENTATION’.
105. The Referendum (Amendment) Act, 1979 does not relate the proposal in any way to the dissolution of NUI. What is proposed is not conditional, but seemingly definite: ‘the election by universities and other institutions of higher education specified by law of such number of members of Seanad Éireann, not exceeding 6’ (my emphasis). Certainly, while a trained lawyer might understand that this could mean that the Oireachtas was being given a widely drawn power as to whether it would, in fact, enact any such laws so that the phrase ‘specified by law’ had wound into it a necessary discretion, this is far from obvious. It reads more naturally as stating that the ‘proposal’ envisages the extension of the franchise to ‘universities and other institutions’ with the Oireachtas enjoying the power to determine by law which institutions would so benefit. And this part of the statement suggests a similar representation: ‘[t]hose so elected would be in substitution for ...’ (again, my emphasis). On any reasonable reading, it promises a change as evident, if nothing else, from the stated contrast between the position at the time of the referendum (‘at present’) and the future (‘those so elected’).
106. It is not simply the case that the statement in the Act and the actual proposal now reflected in Article 18.4.2 as contended for by the respondents were (as counsel for the respondents accepted in oral argument) ‘different’. The difference alleged is between, on the one hand, a constitutional Article which (the respondents say) confers an unfettered discretion on the Oireachtas to legislate so as to expand the franchise for the university seats and which allows the Oireachtas to continue the status quo or indeed to further concentrate the franchise in a single institution, and, on the other, a statement to be delivered to the People which nowhere says any of this, and which is instead wholly consistent with the effect of the Amendment being to produce a change through implementing legislation and a consequent extension of the university panels to include institutions of higher education other than the two then existing universities.
107. This is potentially relevant to the disposition of this case at two levels. First, this was the statement which, as prescribed by law, was sent to every elector at their home. It was publicly displayed at every polling station. And it was read by the presiding officer to any incapacitated elector who was unable to understand the proposal when read to him or her. It is as reliable an account of what the People might reasonably have believed they were voting for, as one could conceive.
108. But second, and at least as importantly, this statement reflected the understanding of the Houses of the Oireachtas themselves as to what was being proposed. The Houses, it must be recalled, initiated and passed the wording of the Seventh Amendment in the form of a Bill as envisaged by Article 46 of the Constitution. Their perception of what was intended was formulated in order to be, and was, enshrined in law by statute. That solemn understanding as communicated by the text of the provision is one of proposed extension of the franchise, not of a legislative carte blanche to maintain, or not to maintain, the then status quo. It affords a strong indication that what was intended when the Seventh Amendment was adopted, was that the Oireachtas would proceed to legislate so as to enable graduates of institutions of higher education other than TCD and NUI to participate in the Seanad university panel. It is only a strained interpretation of the statement appended to s. 1 of this Act that would point to a belief that Article 18.4.2 was purely permissive. Nor can it be easily squared with an interpretation by which the Oireachtas could legislate so as to preserve the then status quo. [14]
109. Obviously, the Oireachtas cannot, whether through the device of a summary of a proposed constitutional amendment or otherwise, change the meaning of a provision of the instrument, just as newspaper articles, advertisements or other commentary on a proposed change to the Constitution can determine what that change, when implemented, actually means. But when there are two possible constructions of an Article as adopted following an Amendment, the Court is quite entitled to have regard to evidence of this kind in order to determine the context in which the Amendment was proposed and adopted, and the likely purpose and intent. In this case, I think this evidence - because of its source and purpose– is unusually and particularly important to that end.
III THE EQUALITY ISSUE
‘Equality in the electoral process’
110. In the course of his submissions, the applicant postulates what he describes as ‘the principle that the Constitution protects equality in the electoral process’. He deduces this from Article 40.1 and Article 5. He notes observations of Budd J. in O’Donovan v. Attorney General [1961] IR 114 at p. 137 (‘a “democratic state” denotes one in which all citizens have equal political rights’), of Denham J. in McKenna v. An Taoiseach (No.2) [1995] 2 IR 10 at p. 53 (‘[t]here is a right to equal treatment in the political process’), of McKechnie J. in Kelly v. Minister for the Environment [2002] IESC 73, [2002] 4 IR 191 at p. 218 (‘the State must in its electoral laws have regard to the concept of equality’) and of Murray C.J. in King v. Minister for the Environment (No.2) [2006] IESC 61, [2007] 1 IR 296 at p. 316 (‘any intervention by the State by way of legislation in the electoral process must serve a legitimate purpose, be proportionate to that purpose and avoid invidious discrimination’). The principle he deduces from these statements, it is said, dictates that the legislature ‘must treat graduates of UL, who presumptively come within the ambit of Article 18.4, equally, save where there is a justification for different treatment’. And, he says, the State has never stood over the merits of any distinction between NUI and TCD graduates to the exclusion of UL graduates.
111. The applicant relied in a similar way on Article 6 of the Constitution. That provision, he says, shows that because the People are, under the Constitution, sovereign, it must follow that the intention of the People at referendum is always relevant to constitutional interpretation involving or concerning an Amendment. Here, the applicant seeks to draw a distinction between what the People endorsed at referendum, and what the government and legislature chose to put before them by way of referendum proposal.
112. In 1979, it is said, the People were given a choice between retaining what are described as the ‘discriminatory’ provisions of Article 18.4 or approving an Amendment to that provision aimed at remedying that discrimination. Stressing that the People were not given the option of removing Article 18.4.1 altogether, they chose, the argument is, the latter. They did not, it is said, ‘vote for inequality … [t]hey voted to remedy it’. Thus, it is said, the High Court erred when it said (as it did) that the People decided that it was constitutionally permissible for the six seats to remain the shared preserve of NUI and TCD.
113. I have noted earlier that the High Court rejected the proposition that principles of equality had a role in the resolution of this dispute because to apply those principles in the manner contended for by the applicant would involve taking the general principles of equality and using them to override a specific provision which permits graduates of NUI and TCD to be preferred over graduates of other higher education institutions. As a result, the Court held that it could not under any circumstances draw a ‘blue pencil through a provision of the Constitution’, especially in circumstances such as these where the People have ratified the continued operation of the provision by enacting the Seventh Amendment.
Does a ‘principle of equality’ require that Article 18.4 be interpreted so as to impose on the Oireachtas an obligation to legislate so as to include UL in the university panel?
114. While I have explained earlier why the application of fundamental principles that can be located in or clearly related to the text of the Constitution might properly be taken into account in interpreting provisions of the instrument, and while there is undoubtedly authority both in the case law, and potentially in the form of the equality guarantee in Article 40.1, for a concept of ‘equality in the elecotral process’ it is axiomatic that principles of this kind will only be of assistance in elucidating the meaning of a constitutional Article where their application can be accommodated within the text and intendment of the provision in question. In my view, the principle relied upon by the applicant cannot be deployed in the manner suggested by his principal argument so as to require that Article 18.4.2 be construed to impose an affirmative obligation to legislate in the manner contended for by him.
115. Much has been written about, and indeed a great deal of helpful evidence tendered in the course of this case explaining, the history of the Seanad from the equivalent institution as provided for in the 1922 Constitution through the dissolution of that body in May 1936, the reluctance on the part of the then Government to reinstate an upper house, to the ambition at the time of the drafting of the Constitution to establish a body constituted along ‘vocational’ lines. As is well known, there have been criticisms of the manner in which the institution has been composed, proposals for (and a failed further referendum intended to bring about) its abolition, and much talk of its reform.
116. Whatever about the merits of the various positions adopted in those debates, what is relevant here - and indisputable - is that the Constitution itself provides for an institution which is not in the most usual sense of the term ‘democratic’ or representative of the electorate as a whole. In this respect it is similar to the second house in many bicameral legislatures. Speaking of that ‘representative’ function, Garvin in his perceptive analysis (The Irish Senate Institute of Public Administration, 1969) puts the matter particularly well (at p. 88-89):
‘the class or stratum which it was intended to represent is undefinable: it exists, but it is not a coherent group; rather it is a medley of groups culled from different sections of Irish society, with few interests in common and liable to take particularist lines on national issues: in this sense vocationalism is the direct and irreconcilable opponent of political activity: in fact, politics in our sort of society exists to soften the clash of vocational interests’.
117. It is one thing to say that provisions of the Constitution should, generally, be interpreted with a view to implementing general constitutional principles of fairness, or of respect for fundamental rights, or indeed of equality, but quite another to articulate how these principles might be applied to the provisions of Article 18. The decisions in McKenna v. An Taoiseach (No.2), and indeed O’Donovan v. Attorney General, show how these principles may have a role within the democratic process, and indeed the fact of those principles is not contested by the respondent. This is because those aspects of that process have at their root a concept of equality: every citizen should have the right to participate in elections on terms that are similar to those who are like positioned and every citizen should have the right to participate in referenda on the same basis.
118. But the Seanad as conceived in the Constitution is different. It does not mandate a system of universal suffrage, and it does not envisage an electoral process in which all citizens are entitled to participate equally or otherwise. Citizens may differ as to whether this is desirable or not. But it is undeniably the structure that was put in place in 1937. This ‘vocational’ feature of the Seanad’s composition, inherent in the architecture put in place by the Constitution itself, is not a structure onto which a constitutional mandate of ‘equality in the political process’ or for that matter the unvarnished terms of Article 40.1 can be easily mapped. Indeed, were it otherwise, those citizens who have obtained the benefit of higher education (and needless to say that in 1937 they were a very small and privileged minority) might not have been singled out for special treatment in the design of the electoral process. [15] There is a difference in treatment of citizen (as the respondents put it in their submissions) ‘baked into the legislative and constitutional architecture of the Upper House’. Differences of franchise and representation, as the respondents also rightly put it, ‘are woven into the fibre of the Seanad itself.’ The application of principles of equality to mandate a particular version of that constitutional structure raises impossible conundrums.
119. So, if this version of the applicant’s argument were well placed and the principle of ‘equality in the democratic process’ were to mean that all legislative discretions conferred by the Constitution in respect of the composition of the Seanad (a) had to be exercised and (b) had to be exercised so as to bring about some version of equal treatment in that process, the courts would be drawn into a radical redesign of the Seanad. This would not merely result - unavoidably - in the imposition on to the Constitution of a version of the Seanad which deviates fundamentally from that actually provided for, but would involve the courts in negotiating and trading what are essentially political judgements without the benefit of any clear justiciable standards.
120. Two specific examples - I think - show that, and why, this is the case. The provision for the election of candidates to the vocational panels envisaged by Article 18.7 could not be described in a sentence that used in any affirmative sense the words ‘equal’, ‘equality’ or, for that matter, ‘democratic’. There is no doubt but that the process might be made ‘more equal’ ‘more democratic’ or for that matter ‘more representative’ by the use of the process envisaged and provided for by Article 19 (the English language version of which is, as I have noted, similar to that appearing in Article 18.4.2). But there is no version of the judicial function that could be reconciled with the design of such a process and no method of constitutional interpretation that could be said to justify superimposing it on Article 19.
121. A similar issue might have presented itself if Article 18.4 had never been amended. Since 1979, new universities have of course been established in the State. It might have been said that once these institutions were brought into being they were so similar to TCD and NUI in status, legal form and function, that the principle of equality in the democratic process demanded that graduates of those institutions also be allowed to vote in elections for the university panel. This is, of course, the argument advanced here as based upon the amended provision, but the facility, in point of fact, could have been squeezed into the text of Article 18.6 as it originally stood: as I have noted while the Constitution identifies NUI and TCD as the universities for the purposes of Article 18.4, it does not say who can or cannot vote in the elections for those panels. There is no reason in theory why graduates of other universities could not have been given a vote on one or other of them. But this would have involved imposing on Article 18 a meaning which it was never intended to bear and which it could not be rationally construed so as to bear. As with the vocational panels, the university panels were intended to represent stated universities and it would turn the constitutional design of the Seanad on its head to conclude that Article 40.1 or a broader concept of political equality, could invalidate legislation because it limited the franchise to persons having some association with the institutions iterated in Article 18.4 as originally adopted by the People.
122. Neither Article 40.1 of the Constitution nor principles of political equality can be credibly overlaid on Article 18 or, for that matter, Article 19 of the Constitution so as to mandate the exercise of legislative powers governing the composition of the Seanad in such a way as to render the electorate for that institution ‘equal’ or, for that matter, no more unequal than they had to be. The text of the provisions would not bear this, nor would the structure of the Seanad as envisaged by the Constitution. Moreover, and as I explain now, the imposition of a constitutional principle of equality of the kind urged would inevitably create impossible uncertainty around any version of the composition of the second House of the Oireachtas.
Article 40.1 and legislation enacted under Article 18.4.2
123. While it would not therefore not be appropriate to overlay Article 18.4 with a duty to legislate by reason of a principle of equality of the kind contended for by the applicant, any legislation that is passed pursuant to any provision of Article 18 is itself in theory amenable to challenge under Article 40.1. So, to take that example again, while it could not be said that there was any duty on the Oireachtas to exercise the undoubted power conferred by Article 19, if it did exercise that power the resulting legislation would in theory be susceptible to a challenge under any other provision of the Constitution. In this case the respondents did not challenge that proposition, or at least did not do so insofar as Article 18.4.2 was concerned. And, the applicant’s case suggests, if legislation introduced pursuant to Article 18.4.2 can be challenged pursuant to Article 40.1, this necessarily means that Article 18.4.2 itself is subject to a limitation corresponding to what would, and what would not, comprise constitutionally permissible distinctions between different institutions. But, the interests and considerations to which I have just referred would suggest that Article 40.1 must have but a limited application to such legislation. It is important to explain why this is so.
124. The general principles governing the application of the guarantee in Article 40.1 that all citizens shall as human persons be held equal before the law as read in the light of the proviso that the State shall have due regard in its enactments to differences of capacity, physical and moral, and of social function, have been recently summarised by O’Malley J. in the course of her judgment in Donnelly v. Minister for Social Protection [2022] IESC 31, [2022] 2 ILRM 185 as follows (at para. 188):
‘(i) Article 40.1º provides protection against discrimination that is based on arbitrary, capricious or irrational considerations.
(ii) The burden of proof rests upon the party challenging the constitutionality of a law by reference to Article 40.1º.
(iii) In assessing whether or not a plaintiff has discharged that burden, the court will have regard to the presumption of constitutionality.
(iv) The court will also have regard to the constitutional separation of powers, and will in particular accord deference to the Oireachtas in relation to legislation dealing with matters of social, fiscal and moral policy.
(v) Where the discrimination is based upon matters that can be said to be intrinsic to the human sense of self, or where it particularly affects members of a group that is vulnerable to prejudice and stereotyping, the court will assess the legislation with particularly close scrutiny. Conversely, where there is no such impact, a lesser level of examination is required.
(vi) The objectives of a legislative measure, and its rationality (or irrationality) and justification (or lack of justification) may in some cases be apparent on its face. Conversely, in other cases it may be necessary to adduce evidence in support of a party’s case.’
125. Noting this helpful reduction of the applicable principles, a graduate of an institution excluded from the Seanad franchise by legislation enacted under Article 18.4.2 would, save in the most striking of circumstances face insurmountable hurdles in seeking to agitate a challenge under Article 40.1. Such a graduate could not assert a constitutional right to be a member of the electorate for Seanad Éireann: there is no such right, and their claim would accordingly fall to be determined as a ‘pure’ discrimination claim (as indeed was the case in Donnelly). Such a claim would not benefit from the close scrutiny referenced at para. (v) of this summary, as none of the attributes referred to there are engaged. A court in appraising such a challenge would necessarily afford substantial deference to a legislative choice made within an area of significant constitutional sensitivity. This is not only because the Constitution has clearly and expressly vested the function of crafting the Seanad electorate in the Oireachtas in terms that are broadly drawn (‘one or more of the following’), but because a failure to extend that deference would result in the very uncertainty in the composition of a House of the Oireachtas to which I have earlier referred.
126. To enjoy any prospect of success, it would be necessary for a claimant advancing such a case to establish that the exclusion of their institution in the light of the inclusion of others was based on arbitrary, capricious or irrational considerations. This would be, for the reasons to which I have already referred, an extraordinarily a heavy burden. In deciding whether that burden had been discharged, a court would be forced to conclude that differences in the legal status of, or range of academic study within, an institution whether alone or combined with the fact that the larger and broader the franchise the greater the logistical challenges in administering a Seanad election, would in and of themselves present strong and clear justifications for that differential treatment. When Article 18.4 is placed in the context of the form and structure of the Seanad as envisaged by the Constitution (which lacks any internal equality standard), the Oireachtas would have extremely broad latitude in having regard to factors such as geographical spread or the age profile of electors in exercising the choices available to it. It might decide to prefer larger institutions over smaller ones, or vice versa, or it might decide to lean in favour of institutions that are weighted to science over humanities, or vice versa.
127. All of this follows from the practical reality that, at least unless carefully circumscribed, the application to legal provisions governing the composition of the university constituency for election to Seanad Éireann of a very broadly drawn concept of equality could cast a long shadow. Unless legislation enacted to give effect to Article 18.4.2 were to extend to all graduates of all institutions of higher education in the State (and, even if that constituency could be formulated with any sense of reliability and certainty, it is very important to again stress that there is no version of the constitutional text that could mandate that approach), the identification of any one institution of higher education as part of the franchise inevitably invites graduates of another to claim that there is no real distinction between the institutions so that the designation of one, but not the other, is in breach of this provision.
128. Thus, if the power were exercised so as to extend the franchise to all universities (as that term was traditionally understood) in the State, it might well be said that it would be discriminatory not to extend it also to the Technological Universities. Whether not this was done, other institutions might stake a claim of discrimination also. Graduates of degree awarding institutions such as the Royal College of Surgeons of Ireland, or the Honourable Society of the Kings Inns (both of which had, as of 1937, been conferring degrees for a very long time) might stake a plausible claim of unjustifiable discrimination. Other bodies which might view themselves as institutions of higher education comparable to one or other of these bodies say that their graduates should also be included. Private universities might legitimately complain if they were left outside the net. The list is, almost, endless. Indeed, even within those institutions to which the franchise has been extended, the application of the provisions of Article 40.1 in the manner contended for by the applicant would invite constitutional chaos: as matters stand a graduate of NUI could very plausibly complain that the large divergence in the number of graduates of NUI and TCD (112,000 NUI and 65,000 TCD) demanded a reconfiguration of the franchise to render it ‘equal’, ‘representative’ and ‘democratic’. The problem is thus not with the theory of applying a principle of equality so as to nudge the interpretation of Article 18.4.2, it is that (short of a universal suffrage for all graduates of all institutions of higher education - which if practically possible could never be imposed having regard to the text and intent) there is no direction in which the interpretation could be nudged that does not itself, arguably, involve an unjustifiable discrimination.
129. These practical considerations are, I think, readily accommodated within the text and spirit of Article 18.4.2. While I deal in some detail in the next section of this judgment with the meaning of this difficult provision, the one feature of it that is absolutely clear is that the legislative power is remarkably widely drawn - ‘one or more of the following’. Whatever ‘one’ and ‘the following’ mean, the Oireachtas has been given a wide berth in identifying the institutions in which the franchise can be vested, and clearly it is expressly envisaged that there may, quite permissibly, be a very small number of institutions chosen. The doctrine of harmonious interpretation as I have earlier described it would suggest that this constitutional consignment and description of function would itself out-rule - at least in all but the most egregious of cases - any viable claim under Article 40.1. It has been said that that which is categorically permitted by one provision of the Constitution cannot be in breach of another (and see in that specific regard Dillane v. Ireland [1980] ILRM 167 at p. 170 per Henchy J.). Of course, it is the case that equal treatment is a fundamental principle postulated by the Constitution. But the exercise of harmonious interpretation inevitably requires that the application of two provisions that appear to pull in different directions be resolved in the light of the purpose of each. And the constitution of the Seanad in general, and the provision made for the university seats in particular, simply do not sit with the application of that principle in anything other than circumstances disclosing blatant, arbitrary, unjustified and wholly irrational differentiation between institutions that are different but identically situated in terms of their precise legal status and academic range. Even then, the reality that the Constitution itself envisages the franchise being confined to a small number of institutions, and the practical consideration that the more institutions that are added to the franchise, the more difficult the organisation of any election and the more likely other institutions would claim that they too have a right to be included in the electorate would all afford weighty grounds on which differential treatment might be objectively justified. It follows that neither that provision, nor a general principle of equality in the political process, can affect in any meaningful way the interpretation of Article 18.4.2.
IV WHAT DOES ARTICLE 18.4.2 MEAN?
Is Article 18.4.2 mandatory or permissive?
130. There is no sentence using the word ‘may’ that is - without further explanation or elaboration - unambiguously permissive. Undoubtedly, the law proceeds on the basis that normally the term has the effect of conferring a power, not of imposing a duty (see State (Sheehan) v. The Government of Ireland [1987] IR 550 and Kenny v. Dental Council [2004] IEHC 105, [2009] 4 IR 321). But sometimes, ‘may’ when placed in context is in fact clearly intended to describe a mandatory obligation, and sometimes even when it is permissive, circumstances can arise in which in a particular situation a power becomes a duty.
131. It is not appropriate that the minutiae of the rules governing the interpretation of primary legislation be uncritically transported into the process of construing the State’s basic law. But decisions in the construction of statutes illustrate the acknowledgement by the law of the fact that sometimes when the word ‘may’ is viewed in the light of the circumstances in which it is uttered, it is understood and intended to be understood not as the extension of an opportunity, but as a command. Similarly, in the construction of a legal instrument permissive language will in certain circumstances be construed so as to impose a duty (and indeed the converse is also the case). Doyle v. Hearne [1987] IR 601 is one example. There, the Court held that the use of the word ‘may’ in a provision relating to the adjournment of proceedings in the Circuit Court while a question of law was referred to this Court, was mandatory. Finlay C.J. said (at p. 607):
‘I have come to the conclusion that the terms of s. 16 of the [Courts of Justice] Act of 1947 are not so unambiguous as to prohibit an interpretation of them aided by a consideration of the apparent intention of the legislature in enacting these provisions. I accept that the provision for the adjournment of the pronouncement of the judgment or order must be construed as mandatory. Any other construction would create a total absurdity for it would be giving to a Circuit Court judge a power to consult the Supreme Court as to the determination of a question of law, but leaving him free to decide the case in which it arose and thus, presumably, the question of law as well, prior to that determination.’
132. That case affords an example of a seemingly permissive provision which was construed as mandatory because to conclude otherwise would have been to enable an absurdity. There are other situations in which the same conclusion has followed for different reasons. Provisions intended to protect the rights of the public are said to be in a category in which it is more likely the court will find the provision to be mandatory (Dodd Statutory Interpretation in Ireland (2008) at para. 12.11), but this same conclusion can also arise from the overall context of the provision (O’Donnell v. South Dublin Co. Co. [2015] IESC 28 at para. 47). Indeed the law has developed some special rules, so that a power which exists for the benefit of a class of persons subject to exhaustive conditions will be treated as imposing a duty where those conditions are shown to have been met (see Application of Dunne [1968] IR 105).
133. These cases are all concerned with statutes, and with widely differing circumstances. For the reason I have already observed, the rules that have developed around the categorisation of those circumstances are not material here. But the reality they reflect must be taken into account in the interpretation of a constitutional provision. And in the case of the provision under consideration here, I have concluded that Article 18.4.2 presents one of those provisions in which the word ‘may’ should be construed as imposing a duty rather than as providing for a discretionary power. That provision imposed an obligation on the Oireachtas to legislate as to the composition of the Seanad university franchise after the adoption of the Seventh Amendment (I will deal shortly with how). In this regard, the following features of the text, purpose and context of the provision are particularly relevant.
134. First, it follows from everything I have said earlier in this judgment that the consequence of the Seventh Amendment is to render Article 18.4 as a whole a uniquely difficult and opaque provision of the Constitution. Far from being clear, simple and capable of literal interpretation, Article 18.4 requires implication, interpolation and reconstruction to make it work. Article 18.4.1 (i) and (ii) (but not (iii)) must be read as if they included a proviso along the lines ‘unless and until the power vested in the Oireachtas by Article 18.4.2 is exercised ...’. Article 18.6 may have to be rewritten so that the term ‘Universities’ includes new universities other than TCD and NUI and, more challengingly, ‘institutions of higher education’. Even in doing this, questions arise as to why Article 18.4.2 referred to the franchise being governed ‘in the manner to be provided by law’ when Article 18.6 made similar provision, and whether Article 18.6 was intended to fall by the wayside upon the exercise by the Oireachtas of its function to enact legislation under the former provision. As I have earlier noted and explain further shortly, the manner in which seats were to be distributed between institutions lacks clarity, and it is not obvious why the institutions referred to in Article 18.4.2 were identified separately.
135. It would, of course, be crude and illogical to conclude that simply because other parts of Article 18.4 require re-writing or implication, that therefore Article 18.4.2 confers an obligation rather than a power. But it would be equally wrong not to acknowledge that it requires for this reason, very close attention and that the conclusion that words carry their more usual meaning, might follow less easily in this provision, than in others.
136. Second, the more general difficulties with the text of the Article to which I have referred take the matter further: as I have explained earlier, on any version, the word ‘shall’ in Article 18.4.1 actually means ‘may’, while the manner in which the same provision has been retained following the Seventh Amendment has the consequence that in one situation - the mooted dissolution of NUI - may in Article 18.4.2 actually would have meant ‘must’. The conclusion that it did so for all purposes, might not follow far behind. This is particularly so given that, at the same time, the word ‘may’ in the penultimate clause of Article 18.4.2, in specifying the number of replacement seats following legislation, actually meant ‘shall’ or, as the Irish text records ‘will’. Article 18.4 as a whole does not consistently use the term ‘shall’ to connote an obligation, and it does not consistently use the term ‘may’ to describe a power. The context, accordingly, does not lend itself to the assumption that would normally follow from the fact that the provision uses language to define the function of the Oireachtas that is generally construed as being permissive.
137. Third, it follows from everything I have said earlier about the rules governing the interpretation of the Constitution, that the primary obligation of the Court in seeking to construe any such provision is to afford it a meaning that reflects the intention of the People as evident by the language they have adopted having regard to the reason they adopted it. In that regard the context is key. The People must be taken to have amended Article 18.4 for a purpose. They did not change the Constitution in vain. The purpose urged by the respondents - the enabling of structural changes to NUI - while certainly facilitated by Article 18.4.3, is nowhere identified as either the precondition to, or sole object of, a change to the Seanad franchise. Indeed, as I have explained earlier, it is not entirely clear how the provision would actually function in that eventuality.
138. Where, as in the case of this provision, an amendment to the Constitution is in none of its parts self-executing being instead entirely dependent on legislative intervention, the Court should incline to resolve any ambiguity in that provision in favour of an obligation to bring about that change, not to allow it to be postponed indefinitely. And in this case, for the reasons I have outlined, the provision is ambiguous.
139. Fourth, all of this is reinforced by the ‘statement of the proposal’ contained in the Referendum (Amendment) Act, 1979. For the reasons I have considered earlier, it is clearly admissible to an understanding of the background to and reason for the provision, its purpose as generally understood and the objective of the Houses of the Oireachtas in proposing the Amendment to the Constitution in the first place.
140. This statutory statement provides in the very particular context that arises here, strong support for the proposition that Article 18.4.2 was intended to do more than to vest in the Oireachtas a temporally open ended and substantively unbounded discretion whether to extend the university constituencies. I do not believe that any elector reading the proposal could have believed that this was what the Amendment was intended to achieve, and it can only be concluded that the Houses of the Oireachtas were operating on the same assumption. This Amendment was presented to the People by the institutions of State responsible for its formulation and proposal as being one that would be followed by legislation extending the Seanad university franchise from TCD and NUI to other institutions of higher education. It was billed as a proposal for ‘the election by universities and other institutions of higher education specified by law of such number of members of Seanad Éireann, not exceeding 6’ (emphasis added). It communicated an intention to change, not to amend the Constitution so that everything could stay as it was. The provisions of the Referendum (Amendment) Act, 1979 and the statement appended to s. 1 thereof affords a uniquely powerful pointer that if Article 18.4.2 said that the Oireachtas ‘may’ legislate following the adoption of the Amendment, it actually meant that it would, and therefore, it must.
141. It should be again said that the fact that Article 18.4.2 was introduced by way of Amendment to the Constitution is central to this conclusion. Had this text been in the document as originally enacted, the case for analogising the Article to Article 19 (which, as I have observed, was clearly intended to be discretionary) would have been strong. But the provision was not included in the original text, and must be judged as an amendment introduced to enable a particular purpose. Even though the draftsman may have chosen to mirror some of the language used in Article 19, the context was - for the reasons I have outlined - entirely different.
The meaning of ‘one or more of the following institutions’
142. It follows that the statutory statement is an equally powerful pointer to the conclusion that the obligation to legislate had to be exercised through an extension of the franchise to institutions other than the two universities referred to in the 1979 Act. This, however, does not at first glance easily sit with the text of the relevant part of the Article:
‘Provision may be made by law for the election, on a franchise and in the manner to be provided by law, by one or more of the following institutions, namely:
i. the universities mentioned in subsection 1° of this section,
ii. any other institutions of higher education in the State …
of so many members of Seanad Éireann as may be fixed by law in substitution for an equal number of the members to be elected pursuant to paragraphs i and ii of the said subsection 1’
143. I have earlier noted the oddity attending the structure of this provision - there is no obvious reason why the universities and the other institutions have been separately categorised, and it is unclear whether - in so doing - it was intended that NUI and TCD would be characterised as a single institution, or separate institutions for the purposes of the provision.
144. Looking at the words alone, it seems to me that the latter must be the proper construction. Each university is a separate institution and it would be highly artificial to designate them in any other way. The ‘other institutions of higher education’ are also categorised together, but the text makes it clear that they are treated as separate institutions (‘any other’). It is impossible to see why the institutions identified in the first paragraph would be presented as a single institution, while those in the second clearly were not. Indeed, had it been intended that (i) and (ii) were, for some reason, defined as distinct - if fictional - ‘institutions’ for the purposes of Article 18.4.2 the term ‘or more’ would have had to have read ‘one or both’. There could never have been a ‘more’. Support for that conclusion, I believe, can also be found in some features of Article 18.4.2 noted by Hogan J. in his judgment. The fact that the dissolution of NUI was contemplated at the time of the adoption of the Seventh Amendment suggests to me that NUI and TCD were viewed as separate institutions rather than being treated jointly (otherwise the dissolution of one would at least raise the prospect that the single institution had necessarily also been dissolved) and I think that this is also supported by the final clause of Article 18.4.2 (‘grouped together or by a single institution’).
145. Reading the words alone and in this way, the franchise could be vested in TCD alone, in NUI alone, in TCD and NUI jointly and unequally, in TCD and NUI jointly and equally, or in any other institutions of higher education alone, or in combination with each other or in combination with TCD and/or NUI.
146. If that is what the provision means, it would facilitate the most remarkable of outcomes. A proposed Amendment to the Constitution for the purposes of changing the composition of the university seats would, in fact, allow the then current position to be continued exactly as it was. An Amendment that was proposed with a view to extending that franchise could be used so as to reduce the number of institutions that could elect members to the Seanad, and indeed to concentrate it in graduates of a single institution. It would have allowed the vote for the university seats to be vested in a number of institutions, none of which were necessarily universities at all. And it would have done this while retaining Article 18.6 intact, with its seemingly clear assumption that members elected by ‘the Universities shall be elected’ on a franchise and in a manner to be provided for by law (emphasis added).
147. An interpretation that allows all of the foregoing is not merely difficult to square with the purpose of the Amendment as understood, but also leaves unexplained many of the apparent anomalies in the provision. In particular, it leaves unexplained why it is that the two sets of institutions (the universities and the other institutions of higher education) have been separated and enumerated in the text as they have. If the provision means what I have just suggested, the only explanation for splitting the institutions in this way is that TCD and NUI are separated from the other institutions simply because they were, by reason of the pre-existing franchise, in a different position. Framing the provision in this manner was one way of making clear that the Oireachtas could continue to include them as part of the franchise. On that basis, Article 18.4.2 has resorted to a very cumbersome means of expressing the otiose.
148. I think that the use of the word ‘substitution’ immediately after (ii) provides a good starting point in resolving some of these issues. This can only mean that the distribution of the franchise required by the legislation to be enacted under Article 18.4.2 was to be different from the allocation provided for in Article 18.4.1. The making of provision for three members to be elected by NUI and three by TCD would not involve any ‘substitution’ because it would replicate what was already there. That would amount to an absurdity.
149. Literally construed, the provision might allow that legislation of this kind could state that one of the universities mentioned in Article 18.4.1 (say NUI) would obtain the entitlement to elect one or more additional members provided for in that law, these being ‘in substitution’ for an equal number of those previously elected by TCD. But I do not believe it can credibly be contended that the Seventh Amendment was adopted by the People so as to mandate the reallocation of the Seanad franchise between TCD and NUI. No-one has ever suggested that this was the sole purpose of the exercise. Something quite different was envisaged. And that additional feature of the anticipated new regime was the extension of the franchise to other institutions.
150. This means that there is a collision between one view of the language of Article 18.4.2 and the purpose of the Referendum by which it was introduced. To that extent, some of the interpretative principles I have identified earlier in this judgment come into play. Literally construed, while the law I have found to be mandated by this provision could have simply moved one or more of the seats referred to in Article 18.4.1 between NUI and TCD thereby excluding the other institutions entirely, this was not the object. That object could be attained by reading the provision as mandating the vesting of one or more of those seats in one or more institutions other than NUI and TCD, but this would involve treating one part of Article 18.4.2 - that in (ii) - as mandatory, and the other ((i)) as optional. It would also leave unresolved the question of why these provisions have been separately iterated in the first place.
151. The text and object can be more easily reconciled if the provisions of Article 18.4.2 (i) and (ii) are read as travelling together. It would make some sense that the institutions are split in this way if the intention was that the Oireachtas had to ensure the inclusion of institutions from both (i) and (ii) in any final legislative prescription of the Seanad university constituencies. It is, in fact, hard to conceive any other substantive (as opposed to presentational) reason the provision would have been so framed. Thus construed, the words ‘by one or more of the following institutions’ assumes that what follows are read as conjoined so that the reference is to one or more institutions from (i) and from (ii). So, on this basis (and subject of course to the proviso that, having regard to Article 18.4.3, both TCD and NUI had not been dissolved) the end result would be that the franchise could be vested in NUI and TCD and one or more other institutions, in NUI and one or more other institutions, or in TCD and one or more other institutions. The third of these reflected what would have occurred had NUI been dissolved, as was proposed at the time of the Amendment, while the second arises because that suggested dissolution was addressed not in express terms but in neutral language. This would have been the clear effect had the Article expressly joined paragraphs (i) and (ii), and it would have been out-ruled had the institutions been designated without separate enumeration or, for that matter, with separate enumeration but as alternatives. The text, obviously, did neither.
152. This is not the literal or for that matter, obvious, interpretation of this provision. But the literal interpretation would allow the Oireachtas to defeat the object of the Amendment by legislating to reallocate, but not to extend, the Seanad franchise. It is an interpretation that assumes the word ‘and’ joins (i) and (ii) and it sits uneasily with the manner in which the noun ‘one’ appears in Article 18.4.2. But the stated object of the Amendment did not contemplate the vesting of the franchise in a single institution - on the contrary everything in the statement of the proposal it is framed in the plural. [16] And this is the only interpretation that explains other difficult aspects of the provision. That being so, I have for the following reasons concluded that this is how the Court must construe the provision.
153. First, as I have already said, it reflects the only plausible explanation of why (i) and (ii) appear in Article 18.4.2 in that form. The respondents were unable to identify any reason why this had been done. Second, it is an interpretation that would have precluded the Oireachtas from passing legislation (as I have found it was required to do) that simply repeated the status quo. Third, it would preclude the Oireachtas from further concentrating the franchise in a single institution (which, clearly, was never the intent). Fourth, it is an interpretation which might explain why Article 18.6 was never rephrased: on this construction (again, subject to NUI and TCD both not having been dissolved) there was always going to be at least one of these universities forming part of the constituency. Fifth, this interpretation provides some explanation for the fact that the Article did not simply say that the Oireachtas could legislate so as to provide for six members of the Seanad by way of replacement of the seats referred to in Article 18.4.1: the intent was that at least to some extent, the franchise was always going to remain vested in either TCD or NUI.
154. Sixth, this is the interpretation that implements the proposal as it was explained in the Referendum (Amendment) Act, 1979. The process whereby the Constitution is amended is underpinned by a pact of sorts between the Houses of the Oireachtas and the People. The People cannot themselves propose an Amendment to the Constitution and, where the Houses of the Oireachtas do so, the People cannot change the wording of the provision so proposed. They can only vote in favour of it or vote to reject it. Where the Houses of the Oireachtas, having proposed an Amendment, tell the People in solemn form and in plain language what the proposal those Houses have themselves devised means, it seems to me to follow from the exclusive role of the Houses in framing and proposing an amendment that the Court - at least in interpreting the resulting Amendment - should hold the Houses to that construction to as great an extent as the language of the provision will allow. And the interpretation that follows from the proposal contained in the Referendum (Amendment) Act, 1979 was that the Seanad university panel would be extended to other institutions of higher education. The interpretation I have suggested is the only one that implements that promise within the language of the provision. It is an interpretation which ensures change rather than stasis, which implements the purpose of the provision and which most sensibly gives effect to what must be presumed to have been the deliberate choice of the People to frame Article 18.4.2 in the manner they did.
155. Having regard to the comments I made earlier regarding the roundabout manner in which the substitution provision in Article 18.4.2 is expressed, the mechanics of this might be noted. It follows from that provision that the constitutional obligation provided for in Article 18.4.2 could be discharged in one of two ways. First, a law could be passed dividing the franchise to elect the six members between ‘one or more’ of TCD and NUI, and ‘one or more’ of the ‘other institutions of higher education’, with the result that those six members of Seanad Éireann would be elected in substitution for the six members referred to in Article 18.4.1. This is, on any version, the neater version. It could also be achieved by a law which simply vested the franchise for one, two, three, four or five of the seats referred to in Article 18.4.1 (i) and (ii) in one or more of those ‘other institutions of higher education’ in substitution for all but one of the seats vested in NUI and TCD under Article 18.4.1. If either of these approaches is adopted, ‘provision’ will have been made ‘by law’ in accordance with the mandate in Article 18.4.2. Each ensures that Article 18.6 continues to have some meaning.
Conclusion
156. Having regard to the foregoing, I conclude that (i) Article 18.4.2 properly construed required the Oireachtas to legislate so as to extend the franchise for the Seanad university panel and (ii) the options available to the Oireachtas in so legislating are to ensure that the franchise is vested (a) in NUI and TCD and one or more other institutions of higher education, (b) in NUI and one or more other such institutions, or (c) in TCD and one or more other such institutions. This is the only interpretation of the provision which implements the purpose of the Amendment as represented to the People while at the same time accommodating the constitutional text. Having regard to the final clause of Article 18.4.2, one or more of these members can be elected by institutions collected together. It is a matter for the Oireachtas to determine in which proportion the relevant seats should be allocated.
V REMEDIES AND ORDERS
157. Because the effect of ss. 6 and 7 of 1937 Act is to limit the Seanad university franchise to graduates of NUI and of TCD, these provisions are not consistent with the provisions of Article 18.4.2, which mandates that the franchise include at least one other institution of higher education. On any reasonable construction of the provision, Article 18.4.2 did not immediately have this effect: the very retention of Article 18.4.1 (i) and (ii) acknowledged that some time would be required to effect the changes envisaged by the new constitutional provision. What, exactly, that time was does not matter now: on any version it has long since expired. And since that point, these provisions have been contrary to the constitutional Article now governing the composition of the university seats.
158. This, in itself, gives rise to a significant issue. Until recently, the conventional understanding was that the issue of a declaration that a provision or provisions of an Act of the Oireachtas is or are unconstitutional operated from the point at which the legislation was enacted. It is well established that this does not mean that all actions undertaken on foot of that legislation are null and void, but the working assumption, traditionally, was that the declaration reached back. The critical difficulty in applying that assumption here is obvious: every Oireachtas composed since 1979 has operated on the basis of the Seanad university franchise as defined by the 1937 Act, and at any point from now on there might be an election before the Oireachtas has the opportunity to remedy the invalidity identified in this judgment.
159. Looking backwards first, I do not think that there can be any serious doubt but that the effect of any declaration of invalidity in this case must be wholly prospective: indeed counsel for the applicant quite properly accepted as such in the course of her oral submissions. No-one has sought to challenge the composition of Seanad Éireann, the legislature has placed clear and necessary deadlines on the bringing of proceedings intended to impugn the validity of an election, and the proposition that laws otherwise duly and properly enacted would be retrospectively invalidated based upon an issue with the composition of one or other of the Houses of the Oireachtas long after the fact enjoys no basis in law, theory or reality. As Hogan J. has put it in his judgment, the validity of the results of all past elections since the enactment of the Seventh Amendment must be regarded as standing beyond legal challenge.
160. The question of what happens from the point at which this judgment is delivered in circumstances such as these was prefigured in the judgment of O’Donnell J. in PC v. Minister for Social Welfare [2018] IESC 57. This followed from the decision in PC v. Minister for Social Protection [2017] IESC 63, [2017] 2 ILRM 369 in which it was found that s. 249(1) of the Social Welfare (Consolidation) Act, 2005 (which operated so as to preclude persons serving a term of imprisonment from receiving certain social welfare benefits) contravened principles of the separation of powers and the administration of justice and where the Court, rather than immediately issue a declaration of invalidity, proceeded to adjourn the matter in order to enable the making of submissions as to the remedy that should issue. In so deciding, it followed a course set in NHV v. Minister for Justice [2017] IESC 35, [2018] 1 IR 246.
161. When the matter came back before the Court, the question presented itself as to the form any declaration of invalidity should assume, and as to whether the applicant was entitled to damages to reflect the loss of benefits to him caused by the application of the provision during his period of imprisonment. In the course of considering those questions, O’Donnell J. (with whose judgment Clarke C.J., McKechnie J. and O’Malley J. agreed) noted that ‘pressing situations’ would inevitably arise in which the grant of a declaration of invalidity with immediate effect would ‘cause a very serious problem, which might be considered more damaging, at least in the short term, than the unconstitutionality identified’ (at para. 14). He continued (at para. 15):
‘Even more extreme circumstances could be envisaged and have arisen in other jurisdictions. One example is where the flaw relates to the election, or legal constitution, of a legislature and where immediate invalidity might be simply incapable of remedy, or worse, might remove the only mechanism for remedying the flaw.’
162. Ultimately, it was not necessary in that case to decide whether to suspend any such declaration for a period of time, but the judgment was clear in its conclusion that such a declaration could, in an appropriate case, issue. He said (at para. 21):
‘The objection that flexibility of a remedy in constitutional challenges offends against principle therefore, is not in my view persuasive. Experience of litigation and disputes more generally, suggests that some cases may not provide the clear cut innocent victim/malicious wrongdoer narrative, more regularly encountered in popular entertainment. More realistically, the system of administration of justice goes to some length to seek to remove unnecessary issues and isolate clear cut issues to which definitive answers can be given. Justice permits, and indeed may require, that a very clear cut decision be rendered in such cases. But it would be foolish not to recognise that there are many situations where the issues require complex and measured solutions … The system established by the Constitution, as judicially interpreted, is a balanced one, which recognises other values as well as the identification of legislation in some respect repugnant to, or inconsistent with, the Constitution … The obligation to render invalid any offending provision of legislation, which is determined to be repugnant to or inconsistent with the Constitution is, a function of the highest importance. As emphasised by Clarke C.J. in the ruling in N.H.V., the normal remedy when unconstitutionality is identified would be the consequential declaration of invalidity of the provision with immediate effect, and that is the position from which the court should be slow to depart, and against which any other remedy should be measured and justified. But I see no justification for an a priori rule that this is the only remedy available … The precise circumstances in which it is appropriate to make any other order, and in particular to suspend a declaration of invalidity, is however, a matter to be considered carefully, cautiously, and on a case by case basis, and will be exceptional. I would, however, reject the argument that it is in principle impermissible for a court to make any other order other than one of an immediate declaration of invalidity.’
163. The practice of granting declarations of this kind is generally related back to a series of decisions of the Supreme Court of Canada in which s. 52 of the Constitution Act, 1982 was construed as enabling the suspension of orders declaring legislation to be unconstitutional. While s. 52 states, baldly, that any law that is inconsistent with the provisions of the Constitution is, to the extent of the inconsistency, ‘of no force and effect’, the Supreme Court of Canada has, since the decision in the Manitoba Language Reference [1985] 1 SCR 721, consistently enabled the grant of such orders where a grave and significant injury to a clear and defined public interest would follow from the immediate issuing of a declaration of invalidity. In Manitoba Language Reference the Court found that the Constitution Act, 1867 and the Manitoba Act, 1870 had the effect that laws that were not in both French and English languages were of no force and effect. However, it decided that then current laws were deemed temporarily valid until such a time where translations could be re-enacted in order to avoid a legal vacuum in Manitoba and to ensure the continuity of the rule of law. The effect was, as Professor Hogg has described it ‘a radical exercise of judicial power, because a body of unconstitutional law was maintained in force solely by virtue of the Court’s order’ (Constitutional Law of Canada 5th Ed. Vol. 2 at p. 179).
164. Nonetheless, a series of subsequent decisions (R. v. Swain [1991] 1 SCR 933; Schachter v. Canada [1992] 2 SCR 679), have shown how the application of a doctrinaire theory of constitutional invalidity could undermine key features of the Constitution itself. These have identified as amongst the circumstances in which suspensory orders may be warranted cases in which striking down the legislation without enacting something in its place would pose a danger to the public, in which striking down the legislation without enacting something in its place would threaten the rule of law, or where legislation was deemed unconstitutional because of under-inclusiveness rather than overbreadth, and therefore striking down the legislation would result in the deprivation of benefits from deserving persons without thereby benefitting the individual whose rights have been violated.
165. Dixon v British Columbia (AG) [1989], 59 DLR (4th) 247, shows how these principles fall to be applied in the context of electoral laws. There, the British Columbia Supreme Court invalidated a system of provincial electoral boundaries found to violate the Charter right to vote but suspended its declaration so that a functional electoral system would remain in place in the event of an election. The possibility that, in a system of parliamentary democracy, an election could be called at any time was found to constitute an ‘emergency’ justifying a suspended declaration in line with the Manitoba Language Reference decision. McLachlin C.J. (as she then was) emphasised the propriety of allowing the legislature to determine the precise features of a new system of electoral boundaries. This consideration addressed the concern that an immediate declaration of invalidity could precipitate an electoral crisis, justifying the issuance of a suspended declaration. She explained (at p. 56):
‘The effect of a declaration that a law is inconsistent with the Charter is to render it of no force and effect under s. 52 of the Charter. In most cases where a particular provision falls under s. 52 the result is to restore the law in question to the status of conformity with the Charter. The effect in this case is arguably the reverse. If the provisions prescribing electoral districts in British Columbia are set aside, the electoral districts vanish. Should an election be required before they are restored, it would be impossible to conduct it. The result would be the disenfranchisement of the citizens of the Province.’
166. Referring to the decision in the Manitoba Language Rights case, she continued (at p. 60):
‘The absence of the machinery necessary to conduct an election in a system where in theory an election can be required at any time, qualifies as an emergency of the magnitude of suspension of all provincial legislation. In my view, it is open to this Court to specify a temporary period during which the existing legislation remains valid and during which the Legislature enacts and brings into force an apportionment scheme which complies with the Charter’.
167. As Hogan J. observes in his concurring judgment, similar conclusions have been reached in Australia (Cormack v. Cope (1974) 131 CLR 432; Victoria v. Commonwealth (1975) 134 CLR 81) and are reflected in the jurisprudence of the German Constitutional Court. The theoretical foundation for this approach is well laid in this jurisdiction by those authorities acknowledging the limitations that must necessarily attach to a finding of constitutional invalidity (The State (Byrne) v. Frawley [1978] IR 326, Murphy v. Attorney General [1982] IR 241 and A. v. Governor of Mountjoy Prison [2006] IESC 45, [2006] 4 IR 88). The Canadian decisions, and the comments of O’Donnell J. in PC v. Minister for Social Protection to which I have earlier referred show that that theory can be projected forwards as well as backwards. On the facts, the case for making similar orders here is unanswerable. A failure to do so would threaten the stability of the democratic system. That this occurs in a context in which the unconstitutionality identified in this judgment is wholly structural and involves no impairment of the constitutional rights of the applicant, or any other citizen, speaks to the overwhelming justification for adopting this unusual course of action, in this singular case.
168. The exercise in reconstitution of the Seanad electorate necessitated by this judgment will, obviously, require time. The Court will receive submissions from the parties as to the length of time required to allow the issues identified here to be so addressed. In the meantime, the suspension should operate in the first instance until 31 July 2023.
169. Having regard to the remedy that will be granted pursuant to the Constitution, it is neither necessary nor appropriate to address the applicant’s claims under ECHR. In these circumstances, this appeal should be allowed, the final orders to abide the further hearing to which I have referred.
[1] The other substantive claim was, in essence, that the exclusion of the plaintiff (but inclusion of members of the Oireachtas or local government) from voting in Seanad general elections disproportionately interfered with his rights under the Constitution.
[2] ‘Féadfar socrú a dhéanamh le dlí ionas go bhféadfadh aon dream feidhme nó gairme beatha, nó aon chomhlacht nó comhairle feidhme nó gairme beatha, an oiread comhaltaí do Sheanad Éireann a thoghadh go lomdíreach agus a chinnfear leis an dlí sin, in ionad an oiread chéanna de na comhaltaí a thoghfar as na comhrollaí d'iarrthóirí a chóireofar faoi Airteagal 18 den Bhunreacht seo.’
[3] It appears that at that point the suggestion was made in the context of a mooted merger of NUI and TCD, see para. 82 (‘Recent developments relating to the Universities may require some consultation with interested bodies as to the allocation of the six University seats between the different colleges’).
[4] Curiously, while ‘universities’ thus appears in the English text, in the Irish text it is capitalised (‘na hOllscoileanna’). In Article 18.6 it is capitalised in both language versions.
[5] It might be said that this, more properly, ought to have referred to ‘institution’ in the singular.
[6] Although the proceedings were commenced pursuant to Order 84 of the Rules of the Superior Court, by Order dated 6 July 2020 it was directed that the action be treated as though it had been commenced by plenary summons. Affidavits and witness statements were delivered by Mr. Barry Ryan, a principal officer in the Department of Housing, Planning and Local Government, Ms. Sheila de Burca, an assistant principal officer in the Department of An Taoiseach, the applicant’s solicitor, Ms. Sinead Lucey, Mr. Aengus Ó Corráin, Professor Gary Murphy (of the School of Law and Government in Dublin City University), Dr. Attracta Halpin, the Registrar of NUI, and Senator Gerard Craughwell. Expert evidence was given on behalf of the applicant by Dr. Laura Cahillane of the Law School at the University of Limerick, and on behalf of the respondents by Dr. Eoin O’Malley, of the School of Law and Government at Dublin City University. These affidavits and witness statements (save for those of Dr. Cahillane and Dr. Eoin O’Malley) were admitted without cross-examination. Those two witnesses were cross examined on their written evidence.
[7] 4 1° Na comhaltaí a thoghfar do Sheanad Éireann, is ar an gcuma seo a leanas a thoghfar iad:—
i Toghfaidh Ollscoil na hÉireann triúr.
ii Toghfaidh Ollscoil Bhaile Átha Cliath triúr.
iii Toghfar triúr is daichead as rollaí d'iarrthóirí a chóireofar ar an gcuma a shocraítear anseo inár ndiaidh.
2° Féadfar foráil a dhéanamh le dlí chun go dtoghfar de réir toghchórais, agus ar an modh, a shocrófar le dlí, ag ceann amháin nó níos mó de na forais seo a leanas, eadhon:
i na hOllscoileanna a luaitear i bhfo-alt 1° den alt seo,
ii aon fhorais eile ardoideachais sa Stát,
an líon sin comhaltaí de Sheanad Éireann a shocrófar le dlí in ionad líon comhionann de na comhaltaí a bheas le toghadh de bhun míreanna i agus ii den fho-alt sin 1°.
Féadfar comhalta nó comhaltaí de Sheanad Éireann a thoghadh faoin bhfo-alt seo ag forais a bheas tiomsaithe le chéile nó ag foras aonair.
3° Ní cead aon ní dá bhfuil san Airteagal seo a agairt chun toirmeasc a chur le hOllscoil a luaitear i bhfo-alt 1° den alt seo a lánscor de réir dlí.
[8] The Divisional Court was clearly of the view that the provision treated each as a separate institution, and indeed this is the most natural interpretation of the Article: ‘Article 18.4.2° … allowed six senators to be elected by the NUI, by TCD, by any other institution of higher education in the State, or any combination of these …’ (at para. 138).
[9] The provision neither defines nor provides any guidance as to the meaning of ‘institutions of higher education’, a further and equally surprising omission. At the time of the adoption of the Seventh Amendment these were defined in statute law as meaning a university, a college of a university or an institution which the Minister designates by regulation as such (Higher Education Authority Act, 1971, s. 1(1)). As of 1979, various orders made under the Higher Education Authority Act, 1971 had the effect that the Royal College of Surgeons in Ireland, the College of the Pharmaceutical Society of Ireland, the National Council for Educational Awards, the National Institute of Higher Education (‘NIHE’) Dublin and NIHE Limerick, the National College of Art and Design and the Royal Irish Academy were designated as Institutions of Higher Education for the purposes of that Act (some of these were and are, in fact, nominating bodies for the purposes of the Cultural and Educational panel for Seanad elections). Since 1979 this list has been added to so as to include Dún Laoghaire Institute of Art, Design and Technology, and Dundalk Institute of Technology. Section 1 of the Higher Education Authority Act, 1971 was subsequently amended in 2006 and 2018 so as to include within the definition of ‘institution of higher education’ the Technological Universities (‘TU’) established under the Technological Universities Act 2018 (TU Dublin, TU Shannon: Midlands Midwest, Atlantic TU, Munster Technological University, and South East TU), a college to which the Institutes of Technology Acts 1992 to 2006 apply and the Dublin Institute of Technology. The Royal College of Surgeons in Ireland (established in 1784) is not a university, but is the holder of a ‘university authorisation order’ which, pursuant to s. 53 of the Universities Act 1997, allows it to describe itself as a university. The applicant contended that only institutions coming within the definition of ‘institutions of higher education’ in s. 1 of the 1971 Act could have been in contemplation in 1979.
[10] See in particular Kelly ‘The Irish Constitution’ (5th Ed. 2018) at para. 1.1.36; Doyle ‘Interpretation: the unrealisable ideal of judicial constraint’ in Carolan ‘Judicial Power in Ireland’ (IPA 2018) 110 at pp. 119-120; Doyle and Hickey ‘Constitutional Law: Text, Cases and Materials’ (2nd Ed. 2019) at para. 4-11.
[11] The analysis here has also been cited with approval in subsequent decisions see Roche v. Roche [2009] IESC 82, [2010] 2 IR 321 at para. 147; Zalewski v. Workplace Relations Commission [2021] IESC 24 at para. 106 of the judgment of O’Donnell J. and para. 129 of the judgment of MacMenamin J.
[12] The evidence before the High Court was that while by 1979 NIHE Limerick was operating, NIHE Dublin did not open until 1980. NIHE Limerick became the University of Limerick under s. 2(1) of the University of Limerick Act, 1989, and NIHE Dublin became Dublin City University under s. 2(1) of the Dublin City University Act, 1989.
[13] It should, if only for the sake of complete accuracy, be noted that this is misquoted in the Divisional Court judgment, which records (at para. 32) the polling card as stating ‘[t]hose so selected will be in substitution for ...’. The statutory text is ‘would’.
[14] It is unsurprising in these circumstances that even the most informed of commentary has (if casually) described the effect of the Seventh Amendment in similar terms: see The Report of the Implementation Group on Seanad Reform 2018 at para. 1.10 (‘[i]n July 1979 the electorate voted to extend the franchise in Seanad elections to graduates of other universities’). The 2015 Manning Report on Seanad Reform similarly referred consistently to the ‘implementation’ of the 1979 Referendum by extending the franchise.
[15] The six university seats are a product of the particular history of the institution and indeed of university representation in Parliament generally: originally it was intended that the Senate of the Irish Free State would contain four university seats (indeed the original Heads of Agreement for the Constitution of the Irish Free State envisaged an additional two seats being afforded to Queens University Belfast in the event that the six counties of Northern Ireland were to remain in the Free State). In the course of the Committee Stage of the Bill to enact the Constitution of the Irish Free State these were transferred to the Dáil, in which six seats were allocated to the two universities (see O’Sullivan The Irish Free State and its Senate (1940) p. 84-85). Article 27 of the 1922 Constitution thus provided that each university in the Irish Free State which was in existence at the date of the coming into operation of the Constitution would be entitled to three representatives to Dáil Éireann. It appears to be generally accepted that the inclusion of six university senators in Article 18.4 was to compensate for the abolition of that representation in the Dáil in 1937 and ‘initially at any rate to ensure a voice for the ex-Unionist minority through the three Trinity seats’ (Manning ‘The Senate’ in ‘The Houses of the Oireachtas: Parliament in Ireland’ eds. MacCartaigh and Manning (IPA 2010)).
[16] The final clause of Article 18.4.2, it should be said, does not change this. This is concerned with ensuring that individual members or groups of members could be elected either by single institutions or institutions as grouped. It does not mean that all members can be elected by a single institution.