It was announced last week that amidst concerns surrounding the low participation of women in Irish politics, the Government intends to introduce legislation requiring political parties to field a minimum quota of 30% of female candidates for elections. This requirement will be enforced through an unusual administrative sanction. Parties failing to meet this target would, it is proposed, lose a portion of the public funding available under the Electoral Acts. The “quota” will rise to 40% after the next general election.

It is not my intention, in this post, to consider the various policy arguments surrounding the necessity or efficacy of the proposed measure. Instead, I simply intend to consider whether, if implemented, it would likely encounter any constitutional difficulty, and the chances of success of any political party or prospective individual candidates who might challenge the constitutional validity of the relevant legislation. As will be outlined, the constitutionality of the proposed measure depends on how the purpose and political-moral quality of the freedom of association guaranteed by the Constitution are interpreted.

Given the notoriously weak constraints imposed on the legislature by the Article 40.1 equality guarantee and the deferential judicial attitude to “rational” gender distinctions contained in legislation, I believe that freedom of association, rather than the equality guarantee, represents the most significant constitutional hurdle for the proposed reform. This requires consideration of whether, and to what degree, freedom of association protects the internal organisational autonomy of political parties. This is more or less virgin territory for the somewhat meagre existing jurisprudence on the constitutional freedom of association. It also requires discussion of whether, on the one hand, political parties enjoy a heightened degree of associational freedom in relation to other categories of private association – given the constitutional sanctity of the electoral process – or, on the other, whether the overwhelming “public” interest in the candidates fielded by formally private parties in fact warrants greater regulation of and intervention in associational freedom in comparison to other types of association.

Article 40.6.1 provides: “The State guarantees liberty for the exercise of the following rights, subject to public order and morality:- …

iii. The right of the citizens to form associations and unions. Laws, however, may be enacted for the regulation and control in the public interest of the exercise of the foregoing right.”

What little we know of the requirements of constitutional freedom of association under existing case law may be summarised as follows.

Freedom of association protects the right to establish associations, subject to public order and morality, and the security of the State. No system of prior authorisation exists for the establishment of associations, which are presumptively free. Only associations threatening the security of the State, such as paramilitary associations, may be prohibited, under the terms of the Offences against the State Act 1939, ss. 16, 18.

Freedom of association also protects the right to join an association of one’s choice. This was emphatically established in the early case NUR v Sullivan [1947] IR 77, where it was held that workers in a particular industry could not be forced to join only the trade union determined by a statutory tribunal. This right to join associations is limited, however, for members of the Defence Forces, the Police, and the Civil Service.

Freedom of association also protects the right to dissociate; to leave or not join an association such as a trade union. It has been determined that industrial pickets directed at forcing employees to join a particular union are unlawful given Article 40.6.1 (Educational Co of Ireland Ltd v Fitzpatrick (No 2) [1961] IR 345).

Finally, and most salient, freedom of association guarantees associations’ autonomy with regard to the determination of membership, including, it seems, the placing of exclusionary terms on membership. Thus, it has been held that a trade union’s refusal to accept a particular member is protected by freedom of association – Tierney v Amalgamated Society of Woodworkers 1959 IR 254. The right to associate with whom one will embraces a corollary right, it appears, to refrain from associating with others.

Thus, for obvious reasons, freedom of association may conflict with laws imposing prohibiting discrimination by private associations, as they potentially curtail the aforementioned right of exclusionary autonomy with regard to membership. A straightforward prohibition on single-gender or single-religion clubs would therefore clearly be unconstitutional. However, this cannot mean that all private clubs, whatever their purpose or nature, are constitutionally free to discriminate, in admitting members, on whatever terms they please – as outlined below, they probably are constitutionally free to discriminate against applicant members only to the extent necessary to protect the aims and ends for which they were established.

Thus, in Equality Authority v Portmarnock Golf Club [2009] IESC 73, in which the respondents were held by a majority of the Supreme Court to not constitute a “discriminating club” within the terms of the Equal Status Act, Hardiman J expounded the constitutional conflict and reconciliation of equality law and freedom of association:

“The  defendant …  is a gentlemens’ golf club. All parties agree that this is perfectly legal …. There is a constitutionally recognised right to freedom of association. At least in the private sphere one may associate with whom one likes and need not associate with those with whom one does not wish to associate. Thus, one may have a ladies’ club, a gentlemens’ club, a gay club, a Catholic club, an Africans’ club and so on. And if one forms such a body, one may exclude from it those who do not meet the basic criterion for membership. Thus, a man may be excluded from a ladies’ club and so on. This is the immemorial position at law, and is also agreed by all parties.

The Equal Status Act 2000 does not alter this position … But even if a club is “a discriminating club”, that fact in no way renders the existence of the club, with its discriminating rules, unlawful …. No attempt is made in the Equal Status Act or anywhere else to do this, perhaps because it is feared that to do so would be unconstitutional”

What goes for sports clubs certainly goes for political parties (and certainly, withdrawal of funding is not the same as straightforward prohibition).  However, what goes for exclusionary prerogatives over membership need not go for the distribution of offices and statuses amongst existing members. The question unresolved in the current jurisprudence, and salient in the case of gender quotas, is whether freedom of association, in addition to protecting the right to form, join and leave associations, and autonomously determining the terms of membership, it also embraces the right to associate freely on persons’ not only through a guarantee of autonomous authority over membership rights, but also, by enjoying autonomy in allocating offices and benefits – such as election nominations in political parties. Does it go beyond the right to associate with whom one will, for objects and ends one promotes, and further protect, ancillary to this, the right to organise internally as one pleases? In other terms, does it protect associational organisational autonomy in addition to the free choice of members and objects? This turns on whether associations’ rights extend beyond the freedom to choose and exclude members in order to pursue certain values, and also embraces the freedom to rank, promote and organise members to the same end. If so, the right might embrace political parties’ autonomy with regard to the nomination of electoral candidates, as well as to their terms of membership per se.

In order to consider whether freedom of association guarantees this level of internal autonomy to political parties, it is necessary to explore different interpretations of the purpose of the norm, and its broader location within the overall scheme of constitutional values, with each of these political-moral interpretations liable to yielding different answers to the problem at hand.

One possible interpretation of freedom of association holds that its purpose is to promote the internal vitality of associations in order to ensure a vibrant “public sphere” as a counterweight to the power of the state (the “associational vitality” interpretation). A second conceives of political freedom of association in a more instrumental sense, as securing the integrity of the political process against partisan manipulation in the conduct of competition for public power (the “political fairness” interpretation). A third, more tightly circumscribed interpretation of this constitutional value suggests that its purpose is simply to enable citizens to act collectively and associate so as to promote, pursue and exercise public and private objects, their ends, and conceptions of the good (the “moral personality” interpretation). On this view, freedom of association merely protects the collective dimensions, and is a corollary of, the personal intellectual freedoms of opinion, expression, conscience and religion, protecting the associative structures giving life and reality to these in social form.

While these values and interpretations underpinning the constitutional norm of freedom of association are by no means to be regarded as mutually exclusive, they may yield different requirements as to the degree of internal organisational autonomy guaranteed to political parties.

If the primary purpose of constitutional freedom of association is to enable citizens to associate for the purpose of collectively pursuing and promoting certain private and public ends, whether in politics or otherwise, then it makes little sense to argue that attaching gender quotas to party candidacy nominations, as the price for public funding, violates freedom of association – because such regulatory measures have not bearing on the aim, objects and ends for which political parties are established to pursue, and do not penalise or burden the pursuit of any such aims. Therefore, if we privilege the “moral personality” interpretation of freedom of association, gender quotas for political parties are unlikely to run afoul of its constitutional guarantee. While the measure may impose a certain regulatory burden on parties, this bears no relation to the aims, ends values which they are established to pursue and promote. Put otherwise, the imposition of a minimum quota of candidates for either gender interferes in no way in the values, ends and conceptions of the good for which citizens may associate with each other in order to promote; whatever these values, they are not especially constrained by the required of including a particular gender in the machinery and processes through which they are pursued. The ends and objectives for which any given political party might be established, and which freedom of association is designed to protect, may presumably be as easily pursued while ensuring greater representation and esteem for a particular gender within the associational framework established for the pursuit of these goals. A possible, rather hypothetical counter-example to this point is the possibility of parties established in order to promote the interests of a particular gender, and for whom the imposition of gender quotas might thus represent an interference in the objects and ends for which they were established, and place them at a discriminatory disadvantage in relation to groups established for the pursuit of gender-neutral ends. An analogy might lie in the imposition of religious-equality norms on religious groups; being forced to accept or promote members who do not share the identity or ethos to whose pursuit and promotion the very existence, the raison d’être of the group, is directed. No registered ‘Men’s party’ exists, but the denial of equal funding to such a group on the basis of the very values or ends it would promote admittedly points to a fascinating constitutional hypothesis which strains the conceptual distinction sketched here (whether the measure would jeopardise the associational freedom of a Southern equivalent of the “Women’s Alliance” – evidently less hypothetical – would depend on whether the proposed legislation will require candidates quotas prescribing minimum candidates of each gender, or of women specifically.)

In any case, this interpretation of freedom of association as protecting associative autonomy with regard to the choice of ends and members, but not internal organisational autonomy for the pursuit of these ends, is probably supported by and echoed in the rather muddy compromise between equality and associative freedom struck in the Equal Status Act 2000, upon which the Portmarnock ruling hinged. Private clubs, as associations, were exempt from the requirement of non-discrimination in the admission of members only to the extent that their “primary purpose” was to cater for the members of particular group, and therefore, they were entitled to confine membership exclusively to that group. The Portmarnock majority infamously defined the “primary purpose” of the respondent golf club, through somewhat torturous reasoning, to be not the playing of golf, but the playing of golf in a gender-exclusive environment – this animated Hardiman J’s perhaps somewhat misplaced analogy with women’s book clubs. In either case, the point is that the legislative draftspersons seemed to have assumed that the conflict between equality legislation and freedom of association is lesser than what might be assumed in the sense that freedom of association does not encompass the right to complete exclusionary freedom with regard to the choice of members,  but simply a limited exclusionary associative freedom with regard to membership where such exclusion is necessary to preserve the ends, purposes and objects for which the association is established. This view was tacitly accepted by the Supreme Court, as the constitutionality of the relevant sections of the Equal Status Act were not challenged. While this analytical framework imposed by the reconciliation of equality and associative freedom was admittedly strained in a deluge of absurd hair-splitting literal distinctions in Portmarnock, it supports the argument advanced here surrounding gender quotas for political parties. Just as associations’ exclusionary prerogatives with regard to membership are constitutionally protected only insofar as the exclusionary choice of members is directly linked the objects and aims which the association exists to pursue, their exclusionary prerogatives with regard to the distribution of power and offices within the association would similarly be protected only insofar as this level of associational autonomy could be considered necessary to preserve the associations’ aims, ends and conceptions of the good. While this autonomy-for-exclusion might be necessary in the case of a religion, for example – few would propose forcing churches to ordain women – it is hardly likely to be necessary in the case of political parties.